This is interesting... I had never thought about that.

http://futurity.org/science-technology/human-speech-is-music-to-our-ears/

-- 
Saludos,
Felipe Sateler

> http://futurity.org/science-technology/human-speech-is-music-to-our-ears/

"Purves says the evidence suggest the main biological reason we 
appreciate music is because it mimics speech, which has been critical to 
our evolutionary success."

That's the kind of pseudo-scientific sentence I just despise (being 
educated as a scientist myself).

Evolutionary success; whatever this means, considering that evolutionary 
failure is nowhere to be seen, and thus every single living being on 
Earth is an evolutionary success (a worm, for example); what does 
evolutionary success have to do with Erik Satie's Gymnopedies ?

Or: do we appreciate bird songs because they mimics speech ? which 
speech ? do birds appreciate our speech ? what about whales ?

Pathetic.


"Though they only worked with western music and spoken English"

... which kills the whole thing IMO.


Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On Thu, 2009-12-03 at 18:49 +0100, Stéphane Rollandin wrote:
> > http://futurity.org/science-technology/human-speech-is-music-to-our-ears/
> 
> "Purves says the evidence suggest the main biological reason we 
> appreciate music is because it mimics speech, which has been critical to 
> our evolutionary success."
> 
> That's the kind of pseudo-scientific sentence I just despise (being 
> educated as a scientist myself).

Well, this is a press release, so it is bound to have this sort of
commentary.

> 
> Evolutionary success; whatever this means, considering that evolutionary 
> failure is nowhere to be seen, and thus every single living being on 
> Earth is an evolutionary success (a worm, for example); what does 
> evolutionary success have to do with Erik Satie's Gymnopedies ?

Evolutionary failure is plenty seen in fossils. 

> 
> Or: do we appreciate bird songs because they mimics speech ? which 
> speech ? do birds appreciate our speech ? what about whales ?

I guess the main point is that music sensing is related to speech
recognition/sensing. Given that speech recognition is crucial to human
survival, it would explain why such a large portion of the population
enjoys music. Western music is then very likely to develop patterns
similar to western speech.

> 
> Pathetic.
> 
> 
> "Though they only worked with western music and spoken English"
> 
> ... which kills the whole thing IMO.

Not really. I would expect a similar relationship with spoken chinese
and chinese music.


-- 
Saludos,
Felipe Sateler

I think this statement is too general. Pitch recognition is a  
necessary aspect of speech recognition and that is indeed why pitch is  
so important in music of almost all nations.

But I am not sure about mimicking speech.

Best

Peiman

On 3 Dec 2009, at 18:03, Felipe Sateler wrote:

> On Thu, 2009-12-03 at 18:49 +0100, Stéphane Rollandin wrote:
>>> http://futurity.org/science-technology/human-speech-is-music-to-our-ears/
>>
>> "Purves says the evidence suggest the main biological reason we
>> appreciate music is because it mimics speech, which has been  
>> critical to
>> our evolutionary success."
>>
>> That's the kind of pseudo-scientific sentence I just despise (being
>> educated as a scientist myself).
>
> Well, this is a press release, so it is bound to have this sort of
> commentary.
>
>>
>> Evolutionary success; whatever this means, considering that  
>> evolutionary
>> failure is nowhere to be seen, and thus every single living being on
>> Earth is an evolutionary success (a worm, for example); what does
>> evolutionary success have to do with Erik Satie's Gymnopedies ?
>
> Evolutionary failure is plenty seen in fossils.
>
>>
>> Or: do we appreciate bird songs because they mimics speech ? which
>> speech ? do birds appreciate our speech ? what about whales ?
>
> I guess the main point is that music sensing is related to speech
> recognition/sensing. Given that speech recognition is crucial to human
> survival, it would explain why such a large portion of the population
> enjoys music. Western music is then very likely to develop patterns
> similar to western speech.
>
>>
>> Pathetic.
>>
>>
>> "Though they only worked with western music and spoken English"
>>
>> ... which kills the whole thing IMO.
>
> Not really. I would expect a similar relationship with spoken chinese
> and chinese music.
>
>
> -- 
> Saludos,
> Felipe Sateler



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

> Evolutionary failure is plenty seen in fossils. 

It was called "success" at the time of their living ;)

> Not really. I would expect a similar relationship with spoken chinese
> and chinese music.

Possibly, but this is not in the study, thus its conclusion is bogus.

By the way, I am not denying by principle a likely relationship between 
speech and music. I would actually be surprised if such relationship was 
proven not to exists at all. It's the evolution idiocy that I despise: 
we have studies coming out very regularly and showing that everything is 
the way it is (or, more frequently, the way it seems to be) just because 
of some evolutionary advantage. In most cases, the argument is just 
laughable, as the one we have here. More unpleasant, this line of 
reasoning denies any intrinsic validity to the field it applies to, 
making everything a simple artifact of the evolutive process.

Again, in the view of evolution itself, just *everything* comes from 
evolution, and *everything* can be labelled a success. So what ? That's 
just tautological.

best,

Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On Thu, 2009-12-03 at 19:46 +0100, Stéphane Rollandin wrote:
> > Evolutionary failure is plenty seen in fossils. 
> 
> It was called "success" at the time of their living ;)

Touche :p

> 
> > Not really. I would expect a similar relationship with spoken chinese
> > and chinese music.
> 
> Possibly, but this is not in the study, thus its conclusion is bogus.

Yes, I agree that the extrapolation is a bit of an overstretch. But, if
more studies like this one appear for other cultures it would provide
(or take away, depending on the results) support for the theory.

> 
> By the way, I am not denying by principle a likely relationship between 
> speech and music. I would actually be surprised if such relationship was 
> proven not to exists at all. It's the evolution idiocy that I despise: 
> we have studies coming out very regularly and showing that everything is 
> the way it is (or, more frequently, the way it seems to be) just because 
> of some evolutionary advantage.

You yourself explained why is this (everything comes from, or at least
is a side-effect of [as in this case], evolution). The point is to
discover what reason(s) made it a success (or lack thereof).

>  In most cases, the argument is just 
> laughable, as the one we have here. More unpleasant, this line of 
> reasoning denies any intrinsic validity to the field it applies to, 
> making everything a simple artifact of the evolutive process.

I could not disagree more. Explaining why stuff is the way it is does
not deprive them of any value. Does the explanation of why a peacock's
tail is so colorful deprive it of its beauty? Does explaining why we
like sugar make it any less tasty? I think it actually makes it more
interesting.

> Again, in the view of evolution itself, just *everything* comes from 
> evolution, and *everything* can be labelled a success. So what ? That's 
> just tautological.

Again, the point is to discover what reason(s) made it a success (or
lack thereof).

-- 
Saludos,
Felipe Sateler

Couldn't agree with you more.

I want to like music because I like it, or even more, because it is
good, not because it's somehow going to help me have children (too
late for that anyway)....

Regards,
Mike

On 12/3/09, Stéphane Rollandin  wrote:
>> http://futurity.org/science-technology/human-speech-is-music-to-our-ears/
>
> "Purves says the evidence suggest the main biological reason we
> appreciate music is because it mimics speech, which has been critical to
> our evolutionary success."
>
> That's the kind of pseudo-scientific sentence I just despise (being
> educated as a scientist myself).
>
> Evolutionary success; whatever this means, considering that evolutionary
> failure is nowhere to be seen, and thus every single living being on
> Earth is an evolutionary success (a worm, for example); what does
> evolutionary success have to do with Erik Satie's Gymnopedies ?
>
> Or: do we appreciate bird songs because they mimics speech ? which
> speech ? do birds appreciate our speech ? what about whales ?
>
> Pathetic.
>
>
> "Though they only worked with western music and spoken English"
>
> ... which kills the whole thing IMO.
>
>
> Stef
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>


-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

 > The point is to
> discover what reason(s) made it a success (or lack thereof).

That's the tricky part, and clearly it's where we disagree. Since this 
is wildly OT, I invite you to email me privately if you wish to continue 
this discussion.

regards,

Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Stéphane Rollandin wrote:
>  what does
> evolutionary success have to do with Erik Satie's Gymnopedies ?
> 

Maybe the Gymnopedies got him laid?


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

> Number 7 has gotten me laid before.

there are only three Gymnopedies, you must be confusing with another 
aphrodisiac


Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I meant #3. Apologies for any confusion/blue what-have-you as a  
result  of prescriptive use of my research.
Occasionally I get my low prime numbers confused.

Greg

On Dec 7, 2009, at 9:10 PM, Stéphane Rollandin wrote:

>> Number 7 has gotten me laid before.
>
> there are only three Gymnopedies, you must be confusing with  
> another aphrodisiac
>
>
> Stef
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Victor Lazzarini wrote:
> Not even the 4ths or 5ths or 9ths. Only the octave escapes.
> 
> 

And not even that on a piano with stretched octaves!

Richard Dobson




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

   2 cents corresponds to a ratio of rouchly 1:1.00115.  Try rendering the following CSD, and see if you can hear 2 cents.  Most people should be able to hear that.  Even though these are sines, and the effect is pretty drastic, even with a piano's tone, 2 cents can usually be heard when the two notes are played at the same time. 





-odac




sr = 44100
kr = 4410
ksmps = 10
nchnls = 1
0dbfs = 1

instr 1 ; 440 Hz

iamp  	=	p4
aosc	oscil	1, 440, 1
kenv	linen	iamp, 0.1, p3, 0.1

	out	aosc * kenv

endin 

instr 2 ;440.50864

iamp	=	p4
aosc	oscil	1, 440.50864, 1
kenv	linen	iamp, 0.1, p3, 0.1

	out	aosc * kenv

endin 



f1 0 4096 10 1 

; 440 Hz 
i 1 1 4 0.75

; 440.50864 Hz
i 2 6 4 0.75

; Both 
i 1 11 4 0.375
i 2 11 4 0.375

e 


 

--- On Mon, 12/7/09, gmschroeder  wrote:

> From: gmschroeder 
> Subject: [Csnd] Re: Re: Re: Re: Re: Re: [OT] Human speech is music to out ears
> To: csound@lists.bath.ac.uk
> Date: Monday, December 7, 2009, 8:10 AM
> 
> Can you people hear 2 cents?
> On Dec 7, 2009, at 11:08 PM, Victor Lazzarini
> wrote:
> There is nothing 'natural' about
> 12TET, it's all construction. Although I like it from
> the purely theoretical point of view.2 cents off
> is still 2 cents off.
> Victor
> On 7 Dec 2009, at 13:14, cameron bobro
> wrote:
> 2 cents off of 2:3 and 3:4, and 4
> cents from 8:9 is certainly close enough to establish that
> these intervals are indeed derived from the harmonic series.
> (Though of course the ninth would generally be a fifth of a
> fifth historically, but whatever)
> 
> At any rate, a great deal of Western "classical"
> music comes from the meantone era in which the fifths were
> more heavily tempered in favor of purer thirds,
> bass-ackwards from 12-tET. And the cultural imperialists who
> try to claim 12-tET as "natural" always drag in
> Mozart- who, of course, used and taught meantone. But
> don't get me started on this...:-)
> 
> 
> 
> --- On Mon, 12/7/09, Victor Lazzarini 
> wrote:
> 
> From: Victor Lazzarini 
> Subject: [Csnd] Re: Re: Re: [OT] Human speech is music to
> out ears
> To: csound@lists.bath.ac.uk
> Date: Monday, December 7, 2009, 4:44 AM
> 
> Not even the 4ths or 5ths or 9ths.
> Only the octave escapes.
> 
> On 7 Dec 2009, at 03:33, cameron bobro
> wrote:
> But in 12-tET it is only the octave and
> fifth/fourth, and arguably the ninth, that match the
> harmonic series. The thirds, which are the actual focus of
> the article, are dissonances with the harmonic series. 
> Attempts to sell 12-tET as "natural" have
> increased in number and stupidity (dishonesty?) in recent
> years- I won't give my full opinion on this, cf.
> Godwin's Law.
> 
> --- On Thu, 12/3/09, Stéphane Rollandin 
> wrote:
> 
> From: Stéphane Rollandin 
> Subject: [Csnd] Re: [OT] Human speech is music to out ears
> To: csound@lists.bath.ac.uk
> Date: Thursday, December 3, 2009, 9:49 AM
> 
> > http://futurity.org/science-technology/human-speech-is-music-to-our-ears/
> 
> "Purves says the evidence suggest the main biological
> reason we appreciate music is because it mimics speech,
> which has been critical to our evolutionary success."
> 
> That's the kind of pseudo-scientific sentence I just
> despise (being educated as a scientist myself).
> 
> Evolutionary success; whatever this means, considering that
> evolutionary failure is nowhere to be seen, and thus every
> single living being on Earth is an evolutionary success (a
> worm, for example); what does evolutionary success have to
> do with Erik Satie's Gymnopedies ?
> 
> Or: do we appreciate bird songs because they mimics speech
> ? which speech ? do birds appreciate our speech ? what about
> whales ?
> 
> Pathetic.
> 
> 
> "Though they only worked with western music and spoken
> English"
> 
> ... which kills the whole thing IMO.
> 
> 
> Stef
> 
> 
> 
> 
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body
> "unsubscribe csound"
> 
>        
> 
>        
> 
> 





Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

mark jamerson wrote:
> 2 cents corresponds to a ratio of rouchly 1:1.00115.  Try rendering
> the following CSD, and see if you can hear 2 cents.  Most people
> should be able to hear that.  Even though these are sines, and the
> effect is pretty drastic, even with a piano's tone, 2 cents can
> usually be heard when the two notes are played at the same time.
> 
> 

That's demonstrating beats of two ~almost~ identical frequencies. The 2 
cents in the thread refers to the difference between a pure fifth (ratio 
3/2, ~702 Cents) and a tempered fifth (700 Cents). It is even more 
subtle than that on the piano because of the stretched octave situation. 
Octaves are stretched to reduce the beating between harmonics - the 
strings are so stiff that higher harmonics are sharp relative to the 
fundamental. So ironically, on a piano the fifths may be closer to 
perfect than ET theory suggests. Much of the art of tuning, 
historically, is based not so much on comparing fundamental pitches per 
se, but on getting the "right" amount of beating between notes. An 
essential tool for the tuner was a pendulum, with which to measure 
beats. On a harpsichord, one can imagine, that mattered a lot.

I attended a demonstration a long time ago at the Menuhin School (UK), 
by a French piano tuner called Serge Cordier. He was proposing (and 
demoinstrating) a new method of tempered tunign based on perfect fifths; 
arguments included better sound with string players (string being tuned 
also to perfect fifths), faster tuning, and "a systematic stretching of 
all octaves". It certainly sounded no worse than 12T ET, and the 
demonstration performance of a piano with string quartet did, as I 
recall, sound very 'harmonious". I see this system is definitely "out 
there", but mostly documented in French:

http://fr.wikipedia.org/wiki/Temp%C3%A9rament_%C3%A9gal_%C3%A0_quintes_justes


The worst interval on the piano (as in 12T ET in general) is the 
2oct+maj 3rd; where the upper note beats with the 5th harmonic of the 
lower. The difference is considerable - more like 14 Cents (the 
pythagorean third was classed as a dissonance for this reason - built 
from 5 perfect fifths it was an even wider interval than we get in 12T ET).

I have a suspicion, untested and totally unproven, that on some pianos 
the hammers are positioned so as to somewhat suppress this tiresome 5th 
harmonic.


Richard Dobson








Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Are there opcodes to detune by a particular number of cents? From, say
pch or cps?

On 12/7/09, Richard Dobson  wrote:
> mark jamerson wrote:
>> 2 cents corresponds to a ratio of rouchly 1:1.00115.  Try rendering
>> the following CSD, and see if you can hear 2 cents.  Most people
>> should be able to hear that.  Even though these are sines, and the
>> effect is pretty drastic, even with a piano's tone, 2 cents can
>> usually be heard when the two notes are played at the same time.
>>
>>
>
> That's demonstrating beats of two ~almost~ identical frequencies. The 2
> cents in the thread refers to the difference between a pure fifth (ratio
> 3/2, ~702 Cents) and a tempered fifth (700 Cents). It is even more
> subtle than that on the piano because of the stretched octave situation.
> Octaves are stretched to reduce the beating between harmonics - the
> strings are so stiff that higher harmonics are sharp relative to the
> fundamental. So ironically, on a piano the fifths may be closer to
> perfect than ET theory suggests. Much of the art of tuning,
> historically, is based not so much on comparing fundamental pitches per
> se, but on getting the "right" amount of beating between notes. An
> essential tool for the tuner was a pendulum, with which to measure
> beats. On a harpsichord, one can imagine, that mattered a lot.
>
> I attended a demonstration a long time ago at the Menuhin School (UK),
> by a French piano tuner called Serge Cordier. He was proposing (and
> demoinstrating) a new method of tempered tunign based on perfect fifths;
> arguments included better sound with string players (string being tuned
> also to perfect fifths), faster tuning, and "a systematic stretching of
> all octaves". It certainly sounded no worse than 12T ET, and the
> demonstration performance of a piano with string quartet did, as I
> recall, sound very 'harmonious". I see this system is definitely "out
> there", but mostly documented in French:
>
> http://fr.wikipedia.org/wiki/Temp%C3%A9rament_%C3%A9gal_%C3%A0_quintes_justes
>
>
> The worst interval on the piano (as in 12T ET in general) is the
> 2oct+maj 3rd; where the upper note beats with the 5th harmonic of the
> lower. The difference is considerable - more like 14 Cents (the
> pythagorean third was classed as a dissonance for this reason - built
> from 5 perfect fifths it was an even wider interval than we get in 12T ET).
>
> I have a suspicion, untested and totally unproven, that on some pianos
> the hammers are positioned so as to somewhat suppress this tiresome 5th
> harmonic.
>
>
> Richard Dobson
>
>
>
>
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>

Aidan Collins wrote:
> Are there opcodes to detune by a particular number of cents? From, say
> pch or cps?
> 

If there aren't, you can compute it:  there are 1200 Cents per octave 
(100 per ET semitone). So something along the lines of pow(2.0,1/1200) 
should give you 1 Cent. Somewhere around 1.00057779 according to my 
calculator.

Richard Dobson





Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

> Aidan Collins wrote:
>> Are there opcodes to detune by a particular number of cents? From, say
>> pch or cps?
>>


cent

cent — Calculates a factor to raise/lower a frequency by a given amount of
cents.

Description

Calculates a factor to raise/lower a frequency by a given amount of cents.

Syntax

cent(x)

This function works at a-rate, i-rate, and k-rate.

Initialization

x -- a value expressed in cents.

Performance

The value returned by the cent function is a factor. You can multiply a
frequency by this factor to raise/lower it by the given amount of cents.




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

 Richard, I see the difference you have pointed out.  The difference is much more subtle in this case, especially with pure sine tones.  But in the broader spectrum tones of the piano, the upper harmonics make the difference much more apparent.  For instance, when tuning the 3/2 fifth A4-E5, the tuner is not using the fundamentals, but rather comparing the coincident partials; the 3rd partial of A4 and the 2nd partial of E5, or in this case, the pitch E6.  If A4=440Hz and E5=660Hz, the coincident partials would line up at 1320Hz, and thus there would be no beating in the tuned fifth.  
 However, if the tuner wants to tune a tempered fifth(700 cents) the fundamentals at A4 and E5 would be 440Hz and ~659.255Hz, respectively.  In this case the third partial of A4 would be 1320Hz, but the second partial of E5 would be 1318.51.  This results in 1.5 beats per second, which is easily heard by the tuner. 

  The previous was the theoretical approach to the problem, and did not account for the real world effects of inharmonicity, and common tuning practice.  The following, however, does account for these: 

  The temperament octave most commonly used by tuners, is the F3-F4 octave.  In this range, the 702 cents tempered fifth of F3-C4 theoretically beats at .59Hz.  In order for the tuner to achieve this, he compares the coincident partials and gets them beating to the desired speed.  Given theoretical values for F3 and C4, the coincident partials would fall at 523.842 and 523.252 Hz, yielding the .59Hz beat.  Working backwards from these values, the frequencies of the fundamentals can be found.  
  Inharmonicity stretches the partials upwards, but if the tuner uses to partials to find the beat speeds, the inharmonicity would result in the fundamentals being stretched downward.  Using arbitrary values, let's say the inharmonicity stretches the 2nd partial up 1 cent, and the 3rd partial up 3 cents(the stretching gets worse as the partials increase). Using these values, we can calculate the fundamentals as follows:

   F3:
  523.842 Hz / 3 = 174.614 Hz 
  174.614 Hz - 3 cents = 174.312 Hz 

   C4
  523.252 Hz / 2 = 261.626 Hz
  261.614 Hz - 1 cent = 261.475 Hz

  The difference between these adjusted values is 702.00134 cents, still a tempered fifth. Using inharmonicity values of 0.5 cents and 1 cent for the 2nd and 3rd partials respectively will yield a difference of 700.992 cents.  2 cents and 5 cents, will yield 702.992 cents.  
  Therefore, when the tuner sets an interval to a specified beat speed, the inharmonicity of the strings is a big factor in how pure or tempered the final tuning becomes.  
  As you said: "the art of tuning, historically, is based not so much on comparing fundamental pitches per se, but on getting the "right" amount of beating between notes."  The modern tuner has many tricks up his sleeve to account for the imperfections in pianos in order to create a balanced or "tempered" tuning.  A battery of beat comparison tests are used to adjust for the imharmonicity of a given piano, one of which is the stretching of octaves.  Many times, the tuner must go back and adjust notes already tuned to yield a better final tuning, and the F3-C4 fifth, preciously tuned to 0.59 beats becomes 0.5 Hz, or 0.63 Hz.  The overall goal being a musically pleasing piano.   



--- On Mon, 12/7/09, Richard Dobson  wrote:

> From: Richard Dobson 
> Subject: [Csnd] Re: Re: Re: Re: Re: Re: Re: Re: [OT] Human speech is music to out ears
> To: csound@lists.bath.ac.uk
> Date: Monday, December 7, 2009, 4:50 PM
> mark jamerson wrote:
> > 2 cents corresponds to a ratio of rouchly
> 1:1.00115.  Try rendering
> > the following CSD, and see if you can hear 2
> cents.  Most people
> > should be able to hear that.  Even though these
> are sines, and the
> > effect is pretty drastic, even with a piano's tone, 2
> cents can
> > usually be heard when the two notes are played at the
> same time.
> > 
> > 
> 
> That's demonstrating beats of two ~almost~ identical
> frequencies. The 2 cents in the thread refers to the
> difference between a pure fifth (ratio 3/2, ~702 Cents) and
> a tempered fifth (700 Cents). It is even more subtle than
> that on the piano because of the stretched octave situation.
> Octaves are stretched to reduce the beating between
> harmonics - the strings are so stiff that higher harmonics
> are sharp relative to the fundamental. So ironically, on a
> piano the fifths may be closer to perfect than ET theory
> suggests. Much of the art of tuning, historically, is based
> not so much on comparing fundamental pitches per se, but on
> getting the "right" amount of beating between notes. An
> essential tool for the tuner was a pendulum, with which to
> measure beats. On a harpsichord, one can imagine, that
> mattered a lot.
> 
> I attended a demonstration a long time ago at the Menuhin
> School (UK), by a French piano tuner called Serge Cordier.
> He was proposing (and demoinstrating) a new method of
> tempered tunign based on perfect fifths; arguments included
> better sound with string players (string being tuned also to
> perfect fifths), faster tuning, and "a systematic stretching
> of all octaves". It certainly sounded no worse than 12T ET,
> and the demonstration performance of a piano with string
> quartet did, as I recall, sound very 'harmonious". I see
> this system is definitely "out there", but mostly documented
> in French:
> 
> http://fr.wikipedia.org/wiki/Temp%C3%A9rament_%C3%A9gal_%C3%A0_quintes_justes
> 
> 
> The worst interval on the piano (as in 12T ET in general)
> is the 2oct+maj 3rd; where the upper note beats with the 5th
> harmonic of the lower. The difference is considerable - more
> like 14 Cents (the pythagorean third was classed as a
> dissonance for this reason - built from 5 perfect fifths it
> was an even wider interval than we get in 12T ET).
> 
> I have a suspicion, untested and totally unproven, that on
> some pianos the hammers are positioned so as to somewhat
> suppress this tiresome 5th harmonic.
> 
> 
> Richard Dobson
> 
> 
> 
> 
> 
> 
> 
> 
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk
> with body "unsubscribe csound"
> 





Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Michael Gogins wrote:
> Thanks for your comments.
>
> About the first point, I did know about IRCAM's mixed bag model of
> software support. My point is merely that it shouldn't be that way. If
> IRCAM is funded by the public, the public should get what IRCAM makes.
> Otherwise, in my view, IRCAM is ripping off the public.
>   

That why here in our "composition departement" of
CNSMD-Lyon (Conservatoire National Supérieur de Musique de Lyon, France).
we use PWGL !!
We have some OM software that our public french state governement paid 
for us,
but which interest for students when they go away from here.

A couple of hours with no goal-directedness, gives me the opportunity  
to chime in on this thread, which much more interesting than the  
original article - in the same way as the music made in 12-tet is  
often more interesting than the theory of 12-tet.  A pattern, perhaps?

If the only aural affects one is capable of describing are "happy" and  
"sad", it seems understandable that one might so classify sounds one  
hears into these two tone-buckets. The practice of theoretical  
classification is rarely an end in itself. Calling it what is, dogma,  
is it used to "beat other classifications about the head" and in this  
way resembles that other great scourge of humanity: religion, which  
is, as Percy Granger once observed is "the ruination of life. All our  
dark & needless belief in sin stems from religion, while our  
disastrous ambitions stem from the fighting spirit of the Old  
Testament."*

Somehow the jump from dissonance _with_ the harmonic series that  
Cameron Bobro wrote about seems to get translated in (harmonic)  
dissonance _per se_. That is a conceptual elision IMO which comes  
about from too-heavy an emphasis on an Pythagorean/Helmholtz ideology/ 
theory as distinct from any of a number of other theories: phenomenal,  
gestalt or even psychophysical etc. My point is not to deny that these  
others do not have an ideological basis, simply to point out that more  
that one (ideological basis), even a mixture of them, is possible. The  
pracitice of theorising only really being intellectully useful when  
undertaken comparatively (as opposed to the mistaken belief that one  
is describing a Reidian "world as it really is"). One of the most  
unpleasant, and I would also say divisive, precessions of the non-tet  
"movements" is the way they try to defend a sort of moral high ground  
over some ideology of tuning "naturalness". If a phenomena can be  
demonstrated in a world, it is natural in that world, including 17.707- 
tet, stretched octaves or whatever. All arguments of general  
consonance/dissonance/naturalness/ etc etc are just dogmatic and have  
nothing to do with an empirical science of music. Music, not sound,  
music.

So when I read in the original article
> Although there are literally millions of scales that could be used  
> to divide the octave, most human music is based on scales  
> compromised of only five to seven tones. The researchers argue the  
> preference for these particular tone collections is based on how  
> closely they approximate the harmonic series of tones produced by  
> humans.
I don't think - gee isn't that amazing that thing that Scientists have  
discovered. Instead I think:
1. Cart before horse. "Argue"? Where is the argument. As Adian Collins  
said: "the classic pitfall of forgetting that correlation does not  
equal causation".

2. A scale is an abstraction - what Xenakis calls "an outside time  
structure" and most human music is not based on abstractions but on  
embodied practice. Abstractions like scales are things which are  
derived from such practices. Where have these scientists been when  
such representationalism  has been so systematically examined as  
dismissed in the philosophy of perception? Worshipping at the Church  
of Locke, no doubt.

3. Gee isn't it amazing that some people, and their editors, who  
"being educated as a scientist [themselves]" ( to appropriate  
Cameron's phrase) are so ignorant of the history of the science that  
they think that they can get away with such drivel as being anything  
other that mental onanism.

4. How sad that that music-ology is so debased a practice in today's  
intellectual mileaux that it is not even referenced by these fakir- 
scientists.

5. Will empirical musicology eventually lead to a more intellectually  
rigourous "musiconomy" (al la G. Loy) that can used to argue with  
these fakirs in a way that is taken seriously, or are we doomed to an  
academic world in which intellectual rigour is ignored in the pursuit  
of careers and other "disastrous ambitions"?

And then I sink into a deep depression when I realise that even most  
computer music composers seem to be still more enchanted by timbre- 
fetishisms than music per se.
Sigh,

David

* From "Things I Dislike", written on a train from Chicago to New York  
August 1, 1954.
On 08/12/2009, at 3:35 AM, Aidan Collins wrote:

> I think the press release does seem to fall for the classic pitfall  
> of forgetting that correlation does not equal causation.
> There probably is a pretty good correlation between perception of of  
> 'sad' or 'happy' speech with intervals approximating major and minor  
> thirds, but that's all there is to say about it. It seems very  
> likely to me that this relationship could arise from being exposed  
> to those intervals throughout life and having a consistent  
> association between those sounds with certain emotions. Parents  
> clearly expect and evoke certain responses to music in their children.
> I think they would need to design some experiment that removes  
> culture and life experience (or if they found similar results  
> testing with animals) before I'd give these results much weight.
>
> Aidan
>
>
>
> On Mon, Dec 7, 2009 at 11:04 AM, Peiman Khosravi  > wrote:
> Yes but music did not grow out of mathematical principles. We may  
> wish to approach it that way as composer's/researchers but the  
> existence of music in the first place is based on what we can hear,  
> not abstract ratios.
>
> In addition, we shouldn't mistake western music notation with the  
> actual music. In most cases performers (e.g. strings or vocals) do  
> play these intervals closer to the natural harmonics. Also we cannot  
> deny that a large part of cultural construct grows out of mental  
> misrepresentations (refer to David Huron's Sweet Anticipation). For  
> instance the Tonic itself does not lead to a sense of resolution (it  
> is actually the relationship between tonic and dominant in the  
> larger context of an implied harmonic sequence that counts: the  
> dominant leading to the tonic is the resolution), nevertheless we  
> attribute the tonic itself with the sense of resolution. Whether our  
> musical perception is based on mistaken attributions is not the  
> question here.
>
> With regard to Michael Gogins' point:
>
>> I want to like music because I like it, or even more, because it is
>> good, not because it's somehow going to help me have children (too
>> late for that anyway)....
>
>
> Evolution is not a question of conscious desire. One's intentions  
> are not necessarily relevant here. For instance do we go to the  
> restaurant because we want to survive, thus continuing the human  
> race?!
>
> Also the article (whether I agree with it or not) does not suggest  
> that music is ruled by evolutionary success, it merely points out  
> that our musical preferences are developed culturally as a byproduct  
> of our evolutionary success. Well I would be surprised if that was  
> not the case.
>
> Best,
>
> Peiman
>
>
>
>
> On 7 Dec 2009, at 15:03, Victor Lazzarini wrote:
>
>> Never mind what we can hear, it's a question of principle. Some  
>> people can't hear 100 cents or more (I can think of some singers).
>> 1.49... is not 1.5 in my book.
>>
>> Victor
>> On 7 Dec 2009, at 14:10, gmschroeder wrote:
>>
>>> Can you people hear 2 cents?
>>>
>>> On Dec 7, 2009, at 11:08 PM, Victor Lazzarini wrote:
>>>
>>>> There is nothing 'natural' about 12TET, it's all construction.  
>>>> Although I like it from the purely theoretical point of view.
>>>> 2 cents off is still 2 cents off.
>>>>
>>>> Victor
>>>>
>>>> On 7 Dec 2009, at 13:14, cameron bobro wrote:
>>>>
>>>>> 2 cents off of 2:3 and 3:4, and 4 cents from 8:9 is certainly  
>>>>> close enough to establish that these intervals are indeed  
>>>>> derived from the harmonic series. (Though of course the ninth  
>>>>> would generally be a fifth of a fifth historically, but whatever)
>>>>>
>>>>> At any rate, a great deal of Western "classical" music comes  
>>>>> from the meantone era in which the fifths were more heavily  
>>>>> tempered in favor of purer thirds, bass-ackwards from 12-tET.  
>>>>> And the cultural imperialists who try to claim 12-tET as  
>>>>> "natural" always drag in Mozart- who, of course, used and taught  
>>>>> meantone. But don't get me started on this...:-)
>>>>>
>>>>>
>>>>>
>>>>> --- On Mon, 12/7/09, Victor Lazzarini   
>>>>> wrote:
>>>>>
>>>>> From: Victor Lazzarini 
>>>>> Subject: [Csnd] Re: Re: Re: [OT] Human speech is music to out ears
>>>>> To: csound@lists.bath.ac.uk
>>>>> Date: Monday, December 7, 2009, 4:44 AM
>>>>>
>>>>> Not even the 4ths or 5ths or 9ths. Only the octave escapes.
>>>>>
>>>>>
>>>>> On 7 Dec 2009, at 03:33, cameron bobro wrote:
>>>>>
>>>>>> But in 12-tET it is only the octave and fifth/fourth, and  
>>>>>> arguably the ninth, that match the harmonic series. The thirds,  
>>>>>> which are the actual focus of the article, are dissonances with  
>>>>>> the harmonic series.  Attempts to sell 12-tET as "natural" have  
>>>>>> increased in number and stupidity (dishonesty?) in recent  
>>>>>> years- I won't give my full opinion on this, cf. Godwin's Law.
>>>>>>
>>>>>> --- On Thu, 12/3/09, Stéphane Rollandin  
>>>>>>  wrote:
>>>>>>
>>>>>> From: Stéphane Rollandin 
>>>>>> Subject: [Csnd] Re: [OT] Human speech is music to out ears
>>>>>> To: csound@lists.bath.ac.uk
>>>>>> Date: Thursday, December 3, 2009, 9:49 AM
>>>>>>
>>>>>> > http://futurity.org/science-technology/human-speech-is-music-to-our-ears/
>>>>>>
>>>>>> "Purves says the evidence suggest the main biological reason we  
>>>>>> appreciate music is because it mimics speech, which has been  
>>>>>> critical to our evolutionary success."
>>>>>>
>>>>>> That's the kind of pseudo-scientific sentence I just despise  
>>>>>> (being educated as a scientist myself).
>>>>>>
>>>>>> Evolutionary success; whatever this means, considering that  
>>>>>> evolutionary failure is nowhere to be seen, and thus every  
>>>>>> single living being on Earth is an evolutionary success (a  
>>>>>> worm, for example); what does evolutionary success have to do  
>>>>>> with Erik Satie's Gymnopedies ?
>>>>>>
>>>>>> Or: do we appreciate bird songs because they mimics speech ?  
>>>>>> which speech ? do birds appreciate our speech ? what about  
>>>>>> whales ?
>>>>>>
>>>>>> Pathetic.
>>>>>>
>>>>>>
>>>>>> "Though they only worked with western music and spoken English"
>>>>>>
>>>>>> ... which kills the whole thing IMO.
>>>>>>
>>>>>>
>>>>>> Stef
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> Send bugs reports to this list.
>>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>>>> "unsubscribe csound"
>>>>>>

________________________________________________
Dr David Worrall.
- Experimental Polymedia:	  worrall.avatar.com.au
- Sonification: www.sonifiction.com.au








Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

> And then I sink into a deep depression when I realise that even most 
> computer music composers seem to be still more enchanted by 
> timbre-fetishisms than music per se.

no need to be depressed, this is just a moment in music history. the 
software tools with have now allow for a deep manipulation of timbre, 
but we sorely lack tools for musical composition proper. I would go as 
far as saying that while we have plenty of tools for making sound, we 
have close to no tool for making music. this is why I ended up 
developing my own environment for composition, and in the process of 
doing so I came to realize how poor are the abstract musical concepts 
currently implemented in music software. there is a lot of work to be 
done in that area.

a good step for a computer music composer IMO is to write a piece for 
piano in the form of a plain MIDI file associated with a piano 
soundfont: low music spec, low sound quality. only good music could pass 
that test.

do your music software let you compose such pieces easily ?

...my own contribution would be there:
http://www.zogotounga.net/TGG/zik/var/06-Derive.ogg


Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On 10/12/2009, at 7:46 PM, Stéphane Rollandin wrote:

>> And then I sink into a deep depression when I realise that even  
>> most computer music composers seem to be still more enchanted by  
>> timbre-fetishisms than music per se.
>
> no need to be depressed, this is just a moment in music history.

No it isn't- it's _our_ moment  :-)
Else you end up with a reductionist argument of the same form as  
whether or not you eat today, you'll be hungry tomorrow so why bother  
eating today?After all, it would be cheaper that way. Music history  
can bother with itself after I'm no longer around, if _anyone_ is  
around after we melt the earth :-(

> the software tools with have now allow for a deep manipulation of  
> timbre, but we sorely lack tools for musical composition proper. I  
> would go as far as saying that while we have plenty of tools for  
> making sound, we have close to no tool for making music. this is why  
> I ended up developing my own environment for composition, and in the  
> process of doing so I came to realize how poor are the abstract  
> musical concepts currently implemented in music software. there is a  
> lot of work to be done in that area.
>

I agree completely. It's not the fault of the current software,  
however. I've always thought of csound (as opposed to cmusic) as a  
sound synthesis not a music generation system. Not trying to be  
everything, but trying to do well what it purports to do, is one of  
its strengths, IMO.
It's our collective responsibility to develop tools for how we want to  
make music and that in turn is related to what we perceive music to  
be. And, for me at least, that comes from the doing and reflecting,  
not just 'theoretical' speculation.
The timbral synthesis period was exciting in as much as we learned a  
lot about the dynamic nature of musical sound (cf what the acoustics  
books told us, eg Helmholtz, Wood etc) but it is time to move on, IMO.
One thing that happened is we (I'm speaking generally, of course,  
there are always exceptions) got caught up in cognitivistic  
computational modelling - which had the effect of de-emphasising the  
proprioceptive and vestibular aspects of music perception. Bonjour  
René Descartes. This objectification of a parametric conceptualisation  
of sound allows us to say, without it being questioned, things like  
"the sound moved around the room" when in fact nothing of the sort  
happened. And if one reads the current philosophy of sound, much of it  
is a woeful attempt to translate the tradition of projecting theories  
of visual perception into the sonic domain, when the senses are quite  
different and function quite differently - else why would we have them  
both (all)?

> a good step for a computer music composer IMO is to write a piece  
> for piano in the form of a plain MIDI file associated with a piano  
> soundfont: low music spec, low sound quality. only good music could  
> pass that test.

  Somehow we missed the simple fact that there has always been always  
been plenty of music made with crude "synthesis engines"  that was  
very musical. Today, whist I have timbre fetish to match anyone, I've  
abandoned the idea that music is primarily about timbre, even just  
about aural an experience.

> do your music software let you compose such pieces easily ?
I'm continue to develop my software toolkit try to assist me to  
compose what interests me. Or to reveal things in which I take an  
interest. "Ease" has never been an important criteria for me.  
Truthfully, I'm probably more wary of it than I should be.
>
> ...my own contribution would be there:
> http://www.zogotounga.net/TGG/zik/var/06-Derive.ogg
>
>
> Stef
>
Thanks for posting this Stef. I'll have look. For me two of the issues  
I've had to deal with wrt software on a very conscious level are   
those of extensibililty and obsolescence. I've spent so much time  
rewriting (and wishing I had the time to rewrite) some software in all  
sorts of languages simply because the underlying tools/OS are no  
longer supported, that, tempting as is often is from a curiosity POV,  
I'm loathe to strike out towards the latest interesting software  
development until there is a considerable momentum behind it. Witness  
the interest in Haskell, for example. It might turn out to be  
wonderful, but I leave it to the young-guns amongst us to explore and  
report back while I get on with trying to improve the tools and  
thoughts I've got. There's plenty of variety out there, but integrity  
is not so easily acquired.

I can't quiet believe, for example, that there isn't a public domain  
API solution to the task of FFT to score notation with the filter  
parameters under algorithmic control. Or there is and I've missed it  
perhaps? Even a half-decent FFT to MIDI would fill a gap.

David

________________________________________________
Dr David Worrall.
- Experimental Polymedia:	  worrall.avatar.com.au
- Sonification: www.sonifiction.com.au
- Education for Financial Independence: www.mindthemarkets.com.au








Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Oops, that is if you think that music can be decomposed of all of its  
expression to fall into a discrete lattice of pitch-duration- 
instrument. I happen to think that you can't do that. If on the other  
hand you have said just 'a piano piece to be played by a pianist',  
then prospects start to improve.

On 10 Dec 2009, at 08:46, Stéphane Rollandin wrote:

>  good step for a computer music composer IMO is to write a piece for  
> piano in the form of a plain MIDI file associated with a piano  
> soundfont: low music spec, low sound quality. only good music could  
> pass that test.



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

> Oops, that is if you think that music can be decomposed of all of its 
> expression to fall into a discrete lattice of pitch-duration-instrument. 
> I happen to think that you can't do that. If on the other hand you have 
> said just 'a piano piece to be played by a pianist', then prospects 
> start to improve.

no, what I meant is much simpler: by cornering the composition activity 
into a very limited form, MIDI notes (so, a crude representation) played 
by a piano soundfont (not a pianist, precisely), then we can better 
appreciate the sheer musicality of the compositions.

an example in my mind is the interpretation of Robert Wyatt melodies by 
Pascal Comelade, using toy instruments with really cheesy sounds. worth 
listening (I think it's called "September Song").

I certainly did not intent to define what music is nor limit its 
expressions.

Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Because, I guess this is quite a hard thing to do. One thing is the  
FFT, the other is how we perceive sound and a third one is how things  
are notated. Not a straight one-to-one mapping, I'm afraid. Like  
polyphonic pitch-tracking and instrument segregation, these are very  
hard things. Also for us: consider how many years of training it takes  
to transcribe down a 4-part Bach chorale from a recording. I also have  
the impression that each one of us hears the combination of tones/ 
instrument/timbre in a complex mix slightly differently and objects do  
not all get segregated in the same way in each of our heads. I might  
be wrong of course.

Last week, I heard of a mobile phone applet that can tell you what a  
piece of music is if it is played to it. Not having a mobile phone  
myself, I can't say it's true or otherwise. That seems to me to at  
least partially negate what I have just said above, so it just shows  
how much I know about these things... ;)

Victor

On 10 Dec 2009, at 10:39, DavidW wrote:

> I can't quiet believe, for example, that there isn't a public domain  
> API solution to the task of FFT to score notation with the filter  
> parameters under algorithmic control. Or there is and I've missed it  
> perhaps? Even a half-decent FFT to MIDI would fill a gap.



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

But my point is that as far as I am concerned, I can't see that the  
environment you suggested allows for much musicality to be expressed.
That's only my opinion.

Victor

On 10 Dec 2009, at 11:58, Stéphane Rollandin wrote:

>> Oops, that is if you think that music can be decomposed of all of  
>> its expression to fall into a discrete lattice of pitch-duration- 
>> instrument. I happen to think that you can't do that. If on the  
>> other hand you have said just 'a piano piece to be played by a  
>> pianist', then prospects start to improve.
>
> no, what I meant is much simpler: by cornering the composition  
> activity into a very limited form, MIDI notes (so, a crude  
> representation) played by a piano soundfont (not a pianist,  
> precisely), then we can better appreciate the sheer musicality of  
> the compositions.
>
> an example in my mind is the interpretation of Robert Wyatt melodies  
> by Pascal Comelade, using toy instruments with really cheesy sounds.  
> worth listening (I think it's called "September Song").
>
> I certainly did not intent to define what music is nor limit its  
> expressions.
>
> Stef
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

>>> And then I sink into a deep depression when I realise that even most 
>>> computer music composers seem to be still more enchanted by 
>>> timbre-fetishisms than music per se.
>>
>> no need to be depressed, this is just a moment in music history.
> 
> No it isn't- it's _our_ moment  :-)
> Else you end up with a reductionist argument of the same form as whether 
> or not you eat today, you'll be hungry tomorrow so why bother eating 
> today?After all, it would be cheaper that way. Music history can bother 
> with itself after I'm no longer around, if _anyone_ is around after we 
> melt the earth :-(

maybe you have to accept the idea that an infinity of very interesting 
thing will happen after your death anyway, so in some sense you're bound 
to miss about everything. strangely this idea pleases me a lot :)


>> do your music software let you compose such pieces easily ?
> I'm continue to develop my software toolkit try to assist me to compose 
> what interests me. Or to reveal things in which I take an interest. 
> "Ease" has never been an important criteria for me. 

same here. by ease, I meant practicability. some concepts can really be 
played with only when the software supports them comprehensively, else 
it gets very tedious. we (as composers and software developer) have to 
reify the lower structural aspects of our composition in order to use 
them effectively; and while these are lower in the view of the 
composition, they are very high-level in terms of software engineering.


Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Victor Lazzarini a écrit :
> But my point is that as far as I am concerned, I can't see that the 
> environment you suggested allows for much musicality to be expressed.
> That's only my opinion.

then what do you thing of the music I linked to as example ? isn't it 
musical ? (don't be afraid to say exactly what you think, I'm not 
attached to it, it's research :)

Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Where can I listen to it?
On 10 Dec 2009, at 12:11, Stéphane Rollandin wrote:

> Victor Lazzarini a écrit :
>> But my point is that as far as I am concerned, I can't see that the  
>> environment you suggested allows for much musicality to be expressed.
>> That's only my opinion.
>
> then what do you thing of the music I linked to as example ? isn't  
> it musical ? (don't be afraid to say exactly what you think, I'm not  
> attached to it, it's research :)
>
> Stef
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I certainly can't think of any languages related to csound that once  
relied almost entirely on non-realtime composition techniques with  
compile times measured in days, and . . .
wait, isn't that what csound was until some time in the early  
nineties, and even later for those of us with very tight budgets?

I hear someone limiting music by demanding it have variations in some  
fixed number of qualities more than I hear someone limiting music by  
declaring it might be valuable to try something I think sounds awful.

Still writing most of his stuff in vim without realtime control and  
trimming his k-rate to make it stop hiccuping,
Greg

On Dec 10, 2009, at 9:04 PM, Victor Lazzarini wrote:

> But my point is that as far as I am concerned, I can't see that the  
> environment you suggested allows for much musicality to be expressed.
> That's only my opinion.
>
> Victor
>
> On 10 Dec 2009, at 11:58, Stéphane Rollandin wrote:
>
>>> Oops, that is if you think that music can be decomposed of all of  
>>> its expression to fall into a discrete lattice of pitch-duration- 
>>> instrument. I happen to think that you can't do that. If on the  
>>> other hand you have said just 'a piano piece to be played by a  
>>> pianist', then prospects start to improve.
>>
>> no, what I meant is much simpler: by cornering the composition  
>> activity into a very limited form, MIDI notes (so, a crude  
>> representation) played by a piano soundfont (not a pianist,  
>> precisely), then we can better appreciate the sheer musicality of  
>> the compositions.
>>
>> an example in my mind is the interpretation of Robert Wyatt  
>> melodies by Pascal Comelade, using toy instruments with really  
>> cheesy sounds. worth listening (I think it's called "September  
>> Song").
>>
>> I certainly did not intent to define what music is nor limit its  
>> expressions.
>>
>> Stef
>>
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Victor Lazzarini a écrit :
> Where can I listen to it?

directly there (Ogg format):
http://www.zogotounga.net/TGG/zik/var/06-Derive.ogg

Stef





Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

That brings us to an interesting question: Need computer music  
instruments (software or hardware) be 'easy to play'? This is  
something I have always asked myself. Often we hear about how  
something is either hard or easy, and whether in people's opinions  
this makes it good or bad.

In the case of traditional music instruments you don't seem to see the  
same things. OK, players can complain some music is hard to play,  
students complain that their instrument is difficult to master, etc.  
But you don't see people going to redesign a violin to make it easier  
to play; or attempts to do something like this seemed to have taken  
away so much of the expressive possibilities that they are disregarded  
as serious.

This links to another question: should we not be regarding 'computer  
music' as a field of study and development that is every bit as  
complex and with long-term-development goals as standard musical  
instruction?

Victor


On 10 Dec 2009, at 12:06, Stéphane Rollandin wrote:

> same here. by ease, I meant practicability. some concepts can really  
> be played with only when the software supports them comprehensively,  
> else it gets very tedious. we (as composers and software developer)  
> have to reify the lower structural aspects of our composition in  
> order to use them effectively; and while these are lower in the view  
> of the composition, they are very high-level in terms of software  
> engineering.



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I like the rhythmic manipulations and the overall concept, but somehow  
this is despite the nature of the medium in which the piece is  
written. In other words, after a little while, I started to hear the  
piece as a demo of what could be if this was played by (definitely  
more than one) pianos with real pianists behind them (disregarding  
whether this is possible or not).
That feeling seems to be inevitable to me (but it's personal).

Victor

On 10 Dec 2009, at 12:19, Stéphane Rollandin wrote:

> Victor Lazzarini a écrit :
>> Where can I listen to it?
>
> directly there (Ogg format):
> http://www.zogotounga.net/TGG/zik/var/06-Derive.ogg
>
> Stef
>
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

So your opinion renders that not music?

On Dec 10, 2009, at 9:39 PM, Victor Lazzarini wrote:

> I like the rhythmic manipulations and the overall concept, but  
> somehow this is despite the nature of the medium in which the piece  
> is written. In other words, after a little while, I started to hear  
> the piece as a demo of what could be if this was played by  
> (definitely more than one) pianos with real pianists behind them  
> (disregarding whether this is possible or not).
> That feeling seems to be inevitable to me (but it's personal).
>
> Victor
>
> On 10 Dec 2009, at 12:19, Stéphane Rollandin wrote:
>
>> Victor Lazzarini a écrit :
>>> Where can I listen to it?
>>
>> directly there (Ogg format):
>> http://www.zogotounga.net/TGG/zik/var/06-Derive.ogg
>>
>> Stef
>>
>>
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Victor Lazzarini a écrit :
> I like the rhythmic manipulations and the overall concept, but somehow 
> this is despite the nature of the medium in which the piece is written. 
> In other words, after a little while, I started to hear the piece as a 
> demo of what could be if this was played by (definitely more than one) 
> pianos with real pianists behind them (disregarding whether this is 
> possible or not).
> That feeling seems to be inevitable to me (but it's personal).

ok, thanks for listening.

I interpret what you say here as a confirmation of my view: musicality 
can be perceived via a crude medium. the fact that you can't help 
hearing the piece as a demo only demonstrates your good taste :)

in other words: there is some music here, even if it needs to be more 
fully expressed (by real intruments, by real players).

so, back to my initial idea: isn't the exercise of producing such MIDI 
pieces an interesting benchmark for a musical composition software ?

Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I agree with most of what you say. But can you clarify this point? I  
have never heard of timbre-fetishism.

P

On 10 Dec 2009, at 04:53, DavidW wrote:

> most computer music composers seem to be still more enchanted by  
> timbre-fetishisms than music per se.
> Sigh,



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On 10 Dec 2009, at 08:46, Stéphane Rollandin wrote:

>> And then I sink into a deep depression when I realise that even  
>> most computer music composers seem to be still more enchanted by  
>> timbre-fetishisms than music per se.
>
> no need to be depressed, this is just a moment in music history. the  
> software tools with have now allow for a deep manipulation of  
> timbre, but we sorely lack tools for musical composition proper.

We still have our ears.

P


> I would go as far as saying that while we have plenty of tools for  
> making sound, we have close to no tool for making music. this is why  
> I ended up developing my own environment for composition, and in the  
> process of doing so I came to realize how poor are the abstract  
> musical concepts currently implemented in music software. there is a  
> lot of work to be done in that area.
>
> a good step for a computer music composer IMO is to write a piece  
> for piano in the form of a plain MIDI file associated with a piano  
> soundfont: low music spec, low sound quality. only good music could  
> pass that test.
>
> do your music software let you compose such pieces easily ?
>
> ...my own contribution would be there:
> http://www.zogotounga.net/TGG/zik/var/06-Derive.ogg
>
>
> Stef
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On 10/12/2009, at 11:01 PM, Victor Lazzarini wrote:

> Because, I guess this is quite a hard thing to do. One thing is the  
> FFT, the other is how we perceive sound and a third one is how  
> things are notated. Not a straight one-to-one mapping, I'm afraid.  
> Like polyphonic pitch-tracking and instrument segregation, these are  
> very hard things. Also for us: consider how many years of training  
> it takes to transcribe down a 4-part Bach chorale from a recording.  
> I also have the impression that each one of us hears the combination  
> of tones/instrument/timbre in a complex mix slightly differently and  
> objects do not all get segregated in the same way in each of our  
> heads. I might be wrong of course.
>
I understand that what you describe is complex. But it wasn't what I  
was asking. I don't want the software to make its decisions for me. I  
want algorithmic access to the parameter space so I can determine, say  
how many partials to render to as chords in the m/s, how many  
different dynamic levels etc.

Also, I'm not sure the Bach chorale eg is a good one. How many years  
of training would it take to one calculate 1 million decimal places of  
pi. And how many seconds for a computer?
> Last week, I heard of a mobile phone applet that can tell you what a  
> piece of music is if it is played to it. Not having a mobile phone  
> myself, I can't say it's true or otherwise. That seems to me to at  
> least partially negate what I have just said above, so it just shows  
> how much I know about these things... ;)
>
me too.
> Victor
>
> On 10 Dec 2009, at 10:39, DavidW wrote:
>
>> I can't quiet believe, for example, that there isn't a public  
>> domain API solution to the task of FFT to score notation with the  
>> filter parameters under algorithmic control. Or there is and I've  
>> missed it perhaps? Even a half-decent FFT to MIDI would fill a gap.
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"

________________________________________________
Dr David Worrall.
- Experimental Polymedia:	  worrall.avatar.com.au
- Sonification: www.sonifiction.com.au
- Education for Financial Independence: www.mindthemarkets.com.au








Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On 10 Dec 2009, at 11:58, Stéphane Rollandin wrote:

>> Oops, that is if you think that music can be decomposed of all of  
>> its expression to fall into a discrete lattice of pitch-duration- 
>> instrument. I happen to think that you can't do that. If on the  
>> other hand you have said just 'a piano piece to be played by a  
>> pianist', then prospects start to improve.
>
> no, what I meant is much simpler: by cornering the composition  
> activity into a very limited form, MIDI notes (so, a crude  
> representation) played by a piano soundfont (not a pianist,  
> precisely), then we can better appreciate the sheer musicality of  
> the compositions.
>

But that is like reading the Bible as law. Music does not exist in a  
"very limited form" (e.g. MIDI notes). Music notation does not contain  
the Music and that is precisely why I feel the need to move away from  
it.

> an example in my mind is the interpretation of Robert Wyatt melodies  
> by Pascal Comelade, using toy instruments with really cheesy sounds.  
> worth listening (I think it's called "September Song").
>
> I certainly did not intent to define what music is nor limit its  
> expressions.
>
> Stef
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

>> but we sorely lack tools for musical composition proper.
> 
> We still have our ears.

I don't compose with my ears, just as I don't paint with my eyes :)

Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

again, it's just an exercise, not a claim that all music is reducible to 
music notation.

here is another piece of mine, guaranteed 100% notation-proof:
http://www.zogotounga.net/TGG/zik/III/01-enfin%20libres.ogg

Stef





Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

No but your ears are the tools that you rely on more than anything  
else during the process of composition. So they are the most important  
part of your compositional tools. As long as you have them you needn't  
worry about software, or pencil sharpeners. Of course we all want  
better software environments that suite our particular creative  
patterns and preferences. But my point was that you are still guided  
by your ears when actually using the software. We sure don't need  
software to compose good music, whatever that is.

I am thinking of Grisey who created the most spectrally rich works  
with no spectral analyzers or FFT to MIDI convertors (on that note you  
should take a look at this: http://recherche.ircam.fr/equipes/repmus/carpentier/orchidee.html) 
.

P

On 10 Dec 2009, at 13:13, Stéphane Rollandin wrote:

>>> but we sorely lack tools for musical composition proper.
>> We still have our ears.
>
> I don't compose with my ears, just as I don't paint with my eyes :)
>
> Stef
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On 10/12/2009, at 11:20 PM, Victor Lazzarini wrote:

> That brings us to an interesting question: Need computer music  
> instruments (software or hardware) be 'easy to play'? This is  
> something I have always asked myself. Often we hear about how  
> something is either hard or easy, and whether in people's opinions  
> this makes it good or bad.
>
> In the case of traditional music instruments you don't seem to see  
> the same things. OK, players can complain some music is hard to  
> play, students complain that their instrument is difficult to  
> master, etc. But you don't see people going to redesign a violin to  
> make it easier to play; or attempts to do something like this seemed  
> to have taken away so much of the expressive possibilities that they  
> are disregarded as serious.
>
My experience with acoustic instruments, also backed-up by far better  
players than I, is that good instruments are harder to play than el- 
cheapo ones because they are more responsive, had a greater tonal  
range etc; these characteristics make them difficult for a beginner to  
play, not less.
So the issue fro me is not ease of play, it is a sort of ratio of  
potentiality / skill. As skill develops software potentiality can be  
increased for more expressive power.
What I don't find interesting, except in a perverse party-trick kind  
of way,  is a piece of software written and locked off by an anonymous  
3rd party that generates a "composition" for you on the click of  
mouse. Or, at the other extreme, the sort of software that makes tasks  
harder than they should be, given the required level of skill;  
requiring dogged determination because of the way the SW forces you to  
deal with it.

> This links to another question: should we not be regarding 'computer  
> music' as a field of study and development that is every bit as  
> complex and with long-term-development goals as standard musical  
> instruction?
>
Do you mean the standard music performance instruction? If so, I'm  
sure there's as much work involved - probably more - but it is  
different class of work - very little concerned with learning through  
embodied memory, for example.

> Victor
>
>
> On 10 Dec 2009, at 12:06, Stéphane Rollandin wrote:
>
>> same here. by ease, I meant practicability. some concepts can  
>> really be played with only when the software supports them  
>> comprehensively, else it gets very tedious. we (as composers and  
>> software developer) have to reify the lower structural aspects of  
>> our composition in order to use them effectively; and while these  
>> are lower in the view of the composition, they are very high-level  
>> in terms of software engineering.
>
>
Yes, I think that's what I was meaning above re brute force.

David
________________________________________________
Dr David Worrall.
- Experimental Polymedia:	  worrall.avatar.com.au
- Sonification: www.sonifiction.com.au
- Education for Financial Independence: www.mindthemarkets.com.au








Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

>  We sure don't need software to compose 
> good music, whatever that is.

well, I do. I can't play any instrument, neither can I read or write 
music, or sing. I have no ear treaning and a very limited knowledge of 
music theory ...now I'm depressed :)

the software is for me the unavoidable operating step allowing the 
actual implementation of a musical idea. the software has to grok my 
idea's structures, so it must be equipped with rather high-level 
representation of the musical concepts I use, which eventually means 
that the software has to be a very personal tool (either by 
construction, or by adaptation).

> I am thinking of Grisey who created the most spectrally rich works with 
> no spectral analyzers or FFT to MIDI convertors (on that note you should 
> take a look at this: 
> http://recherche.ircam.fr/equipes/repmus/carpentier/orchidee.html).

thanks for the link,

Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

In many styles of music the ears are indeed the most important tool --
ears understood metaphorically as including some kind of musical
perception and taste that surely occurs in the brain not the ears.

However, in other styles the mind is the most important tool. I would
include process music and algorithmic composition. In these cases
without the ideas there is nothing for the ears. The ideas have a
distinctly musical component that is heard, if at all, only in the
mind. It may not even be heard at all, only "understood" in a musical
sense.

Regards,
Mike

On 12/10/09, Peiman Khosravi  wrote:
> No but your ears are the tools that you rely on more than anything
> else during the process of composition. So they are the most important
> part of your compositional tools. As long as you have them you needn't
> worry about software, or pencil sharpeners. Of course we all want
> better software environments that suite our particular creative
> patterns and preferences. But my point was that you are still guided
> by your ears when actually using the software. We sure don't need
> software to compose good music, whatever that is.
>
> I am thinking of Grisey who created the most spectrally rich works
> with no spectral analyzers or FFT to MIDI convertors (on that note you
> should take a look at this:
> http://recherche.ircam.fr/equipes/repmus/carpentier/orchidee.html)
> .
>
> P
>
> On 10 Dec 2009, at 13:13, Stéphane Rollandin wrote:
>
>>>> but we sorely lack tools for musical composition proper.
>>> We still have our ears.
>>
>> I don't compose with my ears, just as I don't paint with my eyes :)
>>
>> Stef
>>
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>> "unsubscribe csound"
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"


-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On 11/12/2009, at 12:05 AM, Peiman Khosravi wrote:

> I agree with most of what you say. But can you clarify this point? I  
> have never heard of timbre-fetishism.
>
> P
>
> On 10 Dec 2009, at 04:53, DavidW wrote:
>
>> most computer music composers seem to be still more enchanted by  
>> timbre-fetishisms than music per se.
>> Sigh,

A term I invented a few years ago to describe a kind of musical  
practice in which the effort and thus resulting interest is almost  
exclusively timbral.
For example, an enormous effort has gone into the building and  
tweaking of physical models to produce the sounds musical instruments  
make when activated by some dumb activator, but little effort in the  
building and tweaking of physical models to a gesturally rich sequence  
of activations of such models.
The underlying assumption is that the sound of the instrument is more  
important that the way it is played. Well that isn't the case in the  
acoustic instrumental music world, where an accomplished musician is  
evaluated primarily according to their "musicianship", not the  
complexity of their instrument.

It is this concentration on the synthesis of sounds synthesised from  
such models, and the almost exclusive ignoring of the temporal,  
gestural means of activating them , that I label "timbre-fetishism".

It's historical origins are in the European practices of dissolving  
melody and rhythm into "resonances" (late piano works of Liszt, 2nd  
Viennese School and the serialists etc.


David
________________________________________________
Dr David Worrall.
- Experimental Polymedia:	  worrall.avatar.com.au
- Sonification: www.sonifiction.com.au
- Education for Financial Independence: www.mindthemarkets.com.au








Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Not at all, but it makes not quite the final product.

I think I have to qualify what I said earlier in relation to pitch- 
duration-instrument lattices. This does not really stop music-making,  
perhaps I should have said that, but makes it difficult for a complete  
musical experience. And for some composers, like myself, it really  
makes it hard to write music for, because some of the elements we are  
used to manipulate are absent.

Victor


On 10 Dec 2009, at 12:41, gmschroeder wrote:

> So your opinion renders that not music?
>
> On Dec 10, 2009, at 9:39 PM, Victor Lazzarini wrote:
>
>> I like the rhythmic manipulations and the overall concept, but  
>> somehow this is despite the nature of the medium in which the piece  
>> is written. In other words, after a little while, I started to hear  
>> the piece as a demo of what could be if this was played by  
>> (definitely more than one) pianos with real pianists behind them  
>> (disregarding whether this is possible or not).
>> That feeling seems to be inevitable to me (but it's personal).
>>
>> Victor
>>
>> On 10 Dec 2009, at 12:19, Stéphane Rollandin wrote:
>>
>>> Victor Lazzarini a écrit :
>>>> Where can I listen to it?
>>>
>>> directly there (Ogg format):
>>> http://www.zogotounga.net/TGG/zik/var/06-Derive.ogg
>>>
>>> Stef
>>>
>>>
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>> "unsubscribe csound"
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

For the record I love those late Liszt works!!

I see your point. I would agree with you as far as you mention the use  
of MIDI-based instrumental timbres. I just wasn't sure if you were  
referring to more spectromorphological listening attitudes as well.

Best

Peiman

On 10 Dec 2009, at 14:03, DavidW wrote:

> On 11/12/2009, at 12:05 AM, Peiman Khosravi wrote:
>
>> I agree with most of what you say. But can you clarify this point?  
>> I have never heard of timbre-fetishism.
>>
>> P
>>
>> On 10 Dec 2009, at 04:53, DavidW wrote:
>>
>>> most computer music composers seem to be still more enchanted by  
>>> timbre-fetishisms than music per se.
>>> Sigh,
>
> A term I invented a few years ago to describe a kind of musical  
> practice in which the effort and thus resulting interest is almost  
> exclusively timbral.
> For example, an enormous effort has gone into the building and  
> tweaking of physical models to produce the sounds musical  
> instruments make when activated by some dumb activator, but little  
> effort in the building and tweaking of physical models to a  
> gesturally rich sequence of activations of such models.
> The underlying assumption is that the sound of the instrument is  
> more important that the way it is played. Well that isn't the case  
> in the acoustic instrumental music world, where an accomplished  
> musician is evaluated primarily according to their "musicianship",  
> not the complexity of their instrument.
>
> It is this concentration on the synthesis of sounds synthesised from  
> such models, and the almost exclusive ignoring of the temporal,  
> gestural means of activating them , that I label "timbre-fetishism".
>
> It's historical origins are in the European practices of dissolving  
> melody and rhythm into "resonances" (late piano works of Liszt, 2nd  
> Viennese School and the serialists etc.
>
>
> David
> ________________________________________________
> Dr David Worrall.
> - Experimental Polymedia:	  worrall.avatar.com.au
> - Sonification: www.sonifiction.com.au
> - Education for Financial Independence: www.mindthemarkets.com.au
>
>
>
>
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I thought things like OpenMusic did something of the kind. There must  
be something like this around. Otherwise it is a good tool to be  
developed.
On 10 Dec 2009, at 13:09, DavidW wrote:

>
> On 10/12/2009, at 11:01 PM, Victor Lazzarini wrote:
>
>> Because, I guess this is quite a hard thing to do. One thing is the  
>> FFT, the other is how we perceive sound and a third one is how  
>> things are notated. Not a straight one-to-one mapping, I'm afraid.  
>> Like polyphonic pitch-tracking and instrument segregation, these  
>> are very hard things. Also for us: consider how many years of  
>> training it takes to transcribe down a 4-part Bach chorale from a  
>> recording. I also have the impression that each one of us hears the  
>> combination of tones/instrument/timbre in a complex mix slightly  
>> differently and objects do not all get segregated in the same way  
>> in each of our heads. I might be wrong of course.
>>
> I understand that what you describe is complex. But it wasn't what I  
> was asking. I don't want the software to make its decisions for me.  
> I want algorithmic access to the parameter space so I can determine,  
> say how many partials to render to as chords in the m/s, how many  
> different dynamic levels etc.
>
> Also, I'm not sure the Bach chorale eg is a good one. How many years  
> of training would it take to one calculate 1 million decimal places  
> of pi. And how many seconds for a computer?
>> Last week, I heard of a mobile phone applet that can tell you what  
>> a piece of music is if it is played to it. Not having a mobile  
>> phone myself, I can't say it's true or otherwise. That seems to me  
>> to at least partially negate what I have just said above, so it  
>> just shows how much I know about these things... ;)
>>
> me too.
>> Victor
>>
>> On 10 Dec 2009, at 10:39, DavidW wrote:
>>
>>> I can't quiet believe, for example, that there isn't a public  
>>> domain API solution to the task of FFT to score notation with the  
>>> filter parameters under algorithmic control. Or there is and I've  
>>> missed it perhaps? Even a half-decent FFT to MIDI would fill a gap.
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>
> ________________________________________________
> Dr David Worrall.
> - Experimental Polymedia:	  worrall.avatar.com.au
> - Sonification: www.sonifiction.com.au
> - Education for Financial Independence: www.mindthemarkets.com.au
>
>
>
>
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I'm having a hard time following your argument somewhere that leads  
to a conclusion that allows digital manipulation and music to overlap.
After all, you *are* ultimately just shifting around ones and zeroes  
in "pitch-duration-instrument lattices" whenever you write music with  
a computer.
Or am I ultimately misunderstanding you here?
Greg

On Dec 10, 2009, at 11:08 PM, Victor Lazzarini wrote:

> Not at all, but it makes not quite the final product.
>
> I think I have to qualify what I said earlier in relation to pitch- 
> duration-instrument lattices. This does not really stop music- 
> making, perhaps I should have said that, but makes it difficult for  
> a complete musical experience. And for some composers, like myself,  
> it really makes it hard to write music for, because some of the  
> elements we are used to manipulate are absent.
>
> Victor
>
>
> On 10 Dec 2009, at 12:41, gmschroeder wrote:
>
>> So your opinion renders that not music?
>>
>> On Dec 10, 2009, at 9:39 PM, Victor Lazzarini wrote:
>>
>>> I like the rhythmic manipulations and the overall concept, but  
>>> somehow this is despite the nature of the medium in which the  
>>> piece is written. In other words, after a little while, I started  
>>> to hear the piece as a demo of what could be if this was played  
>>> by (definitely more than one) pianos with real pianists behind  
>>> them (disregarding whether this is possible or not).
>>> That feeling seems to be inevitable to me (but it's personal).
>>>
>>> Victor
>>>
>>> On 10 Dec 2009, at 12:19, Stéphane Rollandin wrote:
>>>
>>>> Victor Lazzarini a écrit :
>>>>> Where can I listen to it?
>>>>
>>>> directly there (Ogg format):
>>>> http://www.zogotounga.net/TGG/zik/var/06-Derive.ogg
>>>>
>>>> Stef
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>> "unsubscribe csound"
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>> "unsubscribe csound"
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Well, not for everyone. If your music is based on things that cannot  
be capture in that medium, then it can't be a good benchmark.


On 10 Dec 2009, at 13:00, Stéphane Rollandin wrote:

> so, back to my initial idea: isn't the exercise of producing such  
> MIDI pieces an interesting benchmark for a musical composition  
> software ?



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Yes OM does it. I've also done this:

export partial tracks from SPEAR in txt format and import them in  
PWGL. After that it's easy to convert the data to midi.

http://www2.siba.fi/pwgl/

Many of the openmusic libraries have been ported to PWGL.

Best

Peiman

On 10 Dec 2009, at 14:13, Victor Lazzarini wrote:

> I thought things like OpenMusic did something of the kind. There  
> must be something like this around. Otherwise it is a good tool to  
> be developed.
> On 10 Dec 2009, at 13:09, DavidW wrote:
>
>>
>> On 10/12/2009, at 11:01 PM, Victor Lazzarini wrote:
>>
>>> Because, I guess this is quite a hard thing to do. One thing is  
>>> the FFT, the other is how we perceive sound and a third one is how  
>>> things are notated. Not a straight one-to-one mapping, I'm afraid.  
>>> Like polyphonic pitch-tracking and instrument segregation, these  
>>> are very hard things. Also for us: consider how many years of  
>>> training it takes to transcribe down a 4-part Bach chorale from a  
>>> recording. I also have the impression that each one of us hears  
>>> the combination of tones/instrument/timbre in a complex mix  
>>> slightly differently and objects do not all get segregated in the  
>>> same way in each of our heads. I might be wrong of course.
>>>
>> I understand that what you describe is complex. But it wasn't what  
>> I was asking. I don't want the software to make its decisions for  
>> me. I want algorithmic access to the parameter space so I can  
>> determine, say how many partials to render to as chords in the m/s,  
>> how many different dynamic levels etc.
>>
>> Also, I'm not sure the Bach chorale eg is a good one. How many  
>> years of training would it take to one calculate 1 million decimal  
>> places of pi. And how many seconds for a computer?
>>> Last week, I heard of a mobile phone applet that can tell you what  
>>> a piece of music is if it is played to it. Not having a mobile  
>>> phone myself, I can't say it's true or otherwise. That seems to me  
>>> to at least partially negate what I have just said above, so it  
>>> just shows how much I know about these things... ;)
>>>
>> me too.
>>> Victor
>>>
>>> On 10 Dec 2009, at 10:39, DavidW wrote:
>>>
>>>> I can't quiet believe, for example, that there isn't a public  
>>>> domain API solution to the task of FFT to score notation with the  
>>>> filter parameters under algorithmic control. Or there is and I've  
>>>> missed it perhaps? Even a half-decent FFT to MIDI would fill a gap.
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>> "unsubscribe csound"
>>
>> ________________________________________________
>> Dr David Worrall.
>> - Experimental Polymedia:	  worrall.avatar.com.au
>> - Sonification: www.sonifiction.com.au
>> - Education for Financial Independence: www.mindthemarkets.com.au
>>
>>
>>
>>
>>
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I'm not sure, but your notion then will include some of the things I  
consider very interesting, and are purely based on timbral-tonal  
evolution ('Antony' by Wessel comes to mind, but there is plenty of  
other examples).

As far as I am concerned, I am sometimes more worried about 'note- 
fetishisms' than timbral ones. I think I see your point, though,  
however I just class these things as 'music that does not interest me'.

Victor


On 10 Dec 2009, at 14:03, DavidW wrote:

> On 11/12/2009, at 12:05 AM, Peiman Khosravi wrote:
>
>> I agree with most of what you say. But can you clarify this point?  
>> I have never heard of timbre-fetishism.
>>
>> P
>>
>> On 10 Dec 2009, at 04:53, DavidW wrote:
>>
>>> most computer music composers seem to be still more enchanted by  
>>> timbre-fetishisms than music per se.
>>> Sigh,
>
> A term I invented a few years ago to describe a kind of musical  
> practice in which the effort and thus resulting interest is almost  
> exclusively timbral.
> For example, an enormous effort has gone into the building and  
> tweaking of physical models to produce the sounds musical  
> instruments make when activated by some dumb activator, but little  
> effort in the building and tweaking of physical models to a  
> gesturally rich sequence of activations of such models.
> The underlying assumption is that the sound of the instrument is  
> more important that the way it is played. Well that isn't the case  
> in the acoustic instrumental music world, where an accomplished  
> musician is evaluated primarily according to their "musicianship",  
> not the complexity of their instrument.
>
> It is this concentration on the synthesis of sounds synthesised from  
> such models, and the almost exclusive ignoring of the temporal,  
> gestural means of activating them , that I label "timbre-fetishism".
>
> It's historical origins are in the European practices of dissolving  
> melody and rhythm into "resonances" (late piano works of Liszt, 2nd  
> Viennese School and the serialists etc.
>
>
> David
> ________________________________________________
> Dr David Worrall.
> - Experimental Polymedia:	  worrall.avatar.com.au
> - Sonification: www.sonifiction.com.au
> - Education for Financial Independence: www.mindthemarkets.com.au
>
>
>
>
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I think you are misunderstanding me. I mean pitch-duration lattice in  
terms of music that primarily concentrates on the  discrete instances  
of these two parameters, disregarding all other possible elements that  
can make up music. Think of a MIDI rendition of 'My favourite things'  
against John Coltrane's (to demonstrate an obvious idea). I am using  
MIDI as an example because it's an easy one, but the issues go beyond  
MIDI.

Victor


On 10 Dec 2009, at 14:13, gmschroeder wrote:

> I'm having a hard time following your argument somewhere that leads  
> to a conclusion that allows digital manipulation and music to overlap.
> After all, you *are* ultimately just shifting around ones and zeroes  
> in "pitch-duration-instrument lattices" whenever you write music  
> with a computer.
> Or am I ultimately misunderstanding you here?
> Greg
>
> On Dec 10, 2009, at 11:08 PM, Victor Lazzarini wrote:
>
>> Not at all, but it makes not quite the final product.
>>
>> I think I have to qualify what I said earlier in relation to pitch- 
>> duration-instrument lattices. This does not really stop music- 
>> making, perhaps I should have said that, but makes it difficult for  
>> a complete musical experience. And for some composers, like myself,  
>> it really makes it hard to write music for, because some of the  
>> elements we are used to manipulate are absent.
>>
>> Victor
>>
>>
>> On 10 Dec 2009, at 12:41, gmschroeder wrote:
>>
>>> So your opinion renders that not music?
>>>
>>> On Dec 10, 2009, at 9:39 PM, Victor Lazzarini wrote:
>>>
>>>> I like the rhythmic manipulations and the overall concept, but  
>>>> somehow this is despite the nature of the medium in which the  
>>>> piece is written. In other words, after a little while, I started  
>>>> to hear the piece as a demo of what could be if this was played  
>>>> by (definitely more than one) pianos with real pianists behind  
>>>> them (disregarding whether this is possible or not).
>>>> That feeling seems to be inevitable to me (but it's personal).
>>>>
>>>> Victor
>>>>
>>>> On 10 Dec 2009, at 12:19, Stéphane Rollandin wrote:
>>>>
>>>>> Victor Lazzarini a écrit :
>>>>>> Where can I listen to it?
>>>>>
>>>>> directly there (Ogg format):
>>>>> http://www.zogotounga.net/TGG/zik/var/06-Derive.ogg
>>>>>
>>>>> Stef
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Send bugs reports to this list.
>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>>> "unsubscribe csound"
>>>>
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>> "unsubscribe csound"
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>> "unsubscribe csound"
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

What is PWGL?
On 10 Dec 2009, at 14:18, Peiman Khosravi wrote:

> Yes OM does it. I've also done this:
>
> export partial tracks from SPEAR in txt format and import them in  
> PWGL. After that it's easy to convert the data to midi.
>
> http://www2.siba.fi/pwgl/
>
> Many of the openmusic libraries have been ported to PWGL.
>
> Best
>
> Peiman
>
> On 10 Dec 2009, at 14:13, Victor Lazzarini wrote:
>
>> I thought things like OpenMusic did something of the kind. There  
>> must be something like this around. Otherwise it is a good tool to  
>> be developed.
>> On 10 Dec 2009, at 13:09, DavidW wrote:
>>
>>>
>>> On 10/12/2009, at 11:01 PM, Victor Lazzarini wrote:
>>>
>>>> Because, I guess this is quite a hard thing to do. One thing is  
>>>> the FFT, the other is how we perceive sound and a third one is  
>>>> how things are notated. Not a straight one-to-one mapping, I'm  
>>>> afraid. Like polyphonic pitch-tracking and instrument  
>>>> segregation, these are very hard things. Also for us: consider  
>>>> how many years of training it takes to transcribe down a 4-part  
>>>> Bach chorale from a recording. I also have the impression that  
>>>> each one of us hears the combination of tones/instrument/timbre  
>>>> in a complex mix slightly differently and objects do not all get  
>>>> segregated in the same way in each of our heads. I might be wrong  
>>>> of course.
>>>>
>>> I understand that what you describe is complex. But it wasn't what  
>>> I was asking. I don't want the software to make its decisions for  
>>> me. I want algorithmic access to the parameter space so I can  
>>> determine, say how many partials to render to as chords in the m/ 
>>> s, how many different dynamic levels etc.
>>>
>>> Also, I'm not sure the Bach chorale eg is a good one. How many  
>>> years of training would it take to one calculate 1 million decimal  
>>> places of pi. And how many seconds for a computer?
>>>> Last week, I heard of a mobile phone applet that can tell you  
>>>> what a piece of music is if it is played to it. Not having a  
>>>> mobile phone myself, I can't say it's true or otherwise. That  
>>>> seems to me to at least partially negate what I have just said  
>>>> above, so it just shows how much I know about these things... ;)
>>>>
>>> me too.
>>>> Victor
>>>>
>>>> On 10 Dec 2009, at 10:39, DavidW wrote:
>>>>
>>>>> I can't quiet believe, for example, that there isn't a public  
>>>>> domain API solution to the task of FFT to score notation with  
>>>>> the filter parameters under algorithmic control. Or there is and  
>>>>> I've missed it perhaps? Even a half-decent FFT to MIDI would  
>>>>> fill a gap.
>>>>
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>> "unsubscribe csound"
>>>
>>> ________________________________________________
>>> Dr David Worrall.
>>> - Experimental Polymedia:	  worrall.avatar.com.au
>>> - Sonification: www.sonifiction.com.au
>>> - Education for Financial Independence: www.mindthemarkets.com.au
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>> "unsubscribe csound"
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

In my view, there is much to do -- very much to do.

To begin with, computer music sounds are not comparable to
instrumental music sounds. Generally, frankly, they are not as good.
This is not due to lack of potential, which is far greater for the
computer, but rather to lack of (a) understanding and (b) traditional
evolution towards quality.

Next, without quality sounds, quality performance is (a) less
rewarding, and (b) harder to improve through practice.

However, I believe the future is very rosy.

CD-quality audio covers a subset of human perception - a large subset,
but with less than the full audible dynamic range and frequency
response. Using this level of quality means that the sheer precision
of the sound is inherently inferior to what you get off a fiddle,
piano, or horn on stage. This is where computer music has been stuck
until very recently.

High-resolution audio (float samples at 96 KHz, say), which is now
standard in professional audio, covers the full range of human
perception, so is no longer at a loss compared to acoustical
instruments. If you have good enough speakers.

Synthesis algorithms continue to evolve. They evolve faster than the
acoustical instruments have evolved. Ergo, they will equal or surpass
the synthesis beauty of acoustical instruments in the future, probably
the near future.

Robotics and human user interface design principles also are
continuing to evolve and afford the possibility of musical performance
interfaces that exploit the full speed, precision and number of
degrees of freedom afforded by trained human performers (which, as I
am sure we all understand, is phenomenally beyond the capacity of the
most advanced robots today).

Indeed, I expect new interfaces to surpass acoustical instruments in
exploiting the potential of the body for expression. We can exploit
not only pressure and impact, but also orientation, speed of motion,
wireless sensing, the whole body, you name it.

Again, these performance interfaces are evolving rapidly.

When the "musical payoff" of the combination of synthesis beauty,
acoustical precision, and (potential) performance precision and
bandwidth hits a "sweet spot," performers will have an incentive to
practice. Not before. Where new musical possibilities are afforded,
for some performers these spots are already being exploited, e.g. Pat
Metheny and the MIDI guitar. More of these spots will appear -- at
least, in potential.

I would advise designers of computer performance system to view
synthesis beauty and performance interfaces as unified systems, and
not to work on one side while neglecting the other side. Otherwise,
the sweetness of the work will not be heard and nobody will feel
motivated to practice enough to get the music out of the system.

Regards,
Mike

On 12/10/09, DavidW  wrote:
>
> On 10/12/2009, at 11:20 PM, Victor Lazzarini wrote:
>
>> That brings us to an interesting question: Need computer music
>> instruments (software or hardware) be 'easy to play'? This is
>> something I have always asked myself. Often we hear about how
>> something is either hard or easy, and whether in people's opinions
>> this makes it good or bad.
>>
>> In the case of traditional music instruments you don't seem to see
>> the same things. OK, players can complain some music is hard to
>> play, students complain that their instrument is difficult to
>> master, etc. But you don't see people going to redesign a violin to
>> make it easier to play; or attempts to do something like this seemed
>> to have taken away so much of the expressive possibilities that they
>> are disregarded as serious.
>>
> My experience with acoustic instruments, also backed-up by far better
> players than I, is that good instruments are harder to play than el-
> cheapo ones because they are more responsive, had a greater tonal
> range etc; these characteristics make them difficult for a beginner to
> play, not less.
> So the issue fro me is not ease of play, it is a sort of ratio of
> potentiality / skill. As skill develops software potentiality can be
> increased for more expressive power.
> What I don't find interesting, except in a perverse party-trick kind
> of way,  is a piece of software written and locked off by an anonymous
> 3rd party that generates a "composition" for you on the click of
> mouse. Or, at the other extreme, the sort of software that makes tasks
> harder than they should be, given the required level of skill;
> requiring dogged determination because of the way the SW forces you to
> deal with it.
>
>> This links to another question: should we not be regarding 'computer
>> music' as a field of study and development that is every bit as
>> complex and with long-term-development goals as standard musical
>> instruction?
>>
> Do you mean the standard music performance instruction? If so, I'm
> sure there's as much work involved - probably more - but it is
> different class of work - very little concerned with learning through
> embodied memory, for example.
>
>> Victor
>>
>>
>> On 10 Dec 2009, at 12:06, Stéphane Rollandin wrote:
>>
>>> same here. by ease, I meant practicability. some concepts can
>>> really be played with only when the software supports them
>>> comprehensively, else it gets very tedious. we (as composers and
>>> software developer) have to reify the lower structural aspects of
>>> our composition in order to use them effectively; and while these
>>> are lower in the view of the composition, they are very high-level
>>> in terms of software engineering.
>>
>>
> Yes, I think that's what I was meaning above re brute force.
>
> David
> ________________________________________________
> Dr David Worrall.
> - Experimental Polymedia:	  worrall.avatar.com.au
> - Sonification: www.sonifiction.com.au
> - Education for Financial Independence: www.mindthemarkets.com.au
>
>
>
>
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"


-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Sorry I saw your link. Oh, yes I've seen this elsewehere in a couple  
of conferences. Looks interesting. Yet Another Language, though.
On 10 Dec 2009, at 14:29, Victor Lazzarini wrote:

> What is PWGL?
> On 10 Dec 2009, at 14:18, Peiman Khosravi wrote:
>
>> Yes OM does it. I've also done this:
>>
>> export partial tracks from SPEAR in txt format and import them in  
>> PWGL. After that it's easy to convert the data to midi.
>>
>> http://www2.siba.fi/pwgl/
>>
>> Many of the openmusic libraries have been ported to PWGL.
>>
>> Best
>>
>> Peiman
>>
>> On 10 Dec 2009, at 14:13, Victor Lazzarini wrote:
>>
>>> I thought things like OpenMusic did something of the kind. There  
>>> must be something like this around. Otherwise it is a good tool to  
>>> be developed.
>>> On 10 Dec 2009, at 13:09, DavidW wrote:
>>>
>>>>
>>>> On 10/12/2009, at 11:01 PM, Victor Lazzarini wrote:
>>>>
>>>>> Because, I guess this is quite a hard thing to do. One thing is  
>>>>> the FFT, the other is how we perceive sound and a third one is  
>>>>> how things are notated. Not a straight one-to-one mapping, I'm  
>>>>> afraid. Like polyphonic pitch-tracking and instrument  
>>>>> segregation, these are very hard things. Also for us: consider  
>>>>> how many years of training it takes to transcribe down a 4-part  
>>>>> Bach chorale from a recording. I also have the impression that  
>>>>> each one of us hears the combination of tones/instrument/timbre  
>>>>> in a complex mix slightly differently and objects do not all get  
>>>>> segregated in the same way in each of our heads. I might be  
>>>>> wrong of course.
>>>>>
>>>> I understand that what you describe is complex. But it wasn't  
>>>> what I was asking. I don't want the software to make its  
>>>> decisions for me. I want algorithmic access to the parameter  
>>>> space so I can determine, say how many partials to render to as  
>>>> chords in the m/s, how many different dynamic levels etc.
>>>>
>>>> Also, I'm not sure the Bach chorale eg is a good one. How many  
>>>> years of training would it take to one calculate 1 million  
>>>> decimal places of pi. And how many seconds for a computer?
>>>>> Last week, I heard of a mobile phone applet that can tell you  
>>>>> what a piece of music is if it is played to it. Not having a  
>>>>> mobile phone myself, I can't say it's true or otherwise. That  
>>>>> seems to me to at least partially negate what I have just said  
>>>>> above, so it just shows how much I know about these things... ;)
>>>>>
>>>> me too.
>>>>> Victor
>>>>>
>>>>> On 10 Dec 2009, at 10:39, DavidW wrote:
>>>>>
>>>>>> I can't quiet believe, for example, that there isn't a public  
>>>>>> domain API solution to the task of FFT to score notation with  
>>>>>> the filter parameters under algorithmic control. Or there is  
>>>>>> and I've missed it perhaps? Even a half-decent FFT to MIDI  
>>>>>> would fill a gap.
>>>>>
>>>>>
>>>>>
>>>>> Send bugs reports to this list.
>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>>> "unsubscribe csound"
>>>>
>>>> ________________________________________________
>>>> Dr David Worrall.
>>>> - Experimental Polymedia:	  worrall.avatar.com.au
>>>> - Sonification: www.sonifiction.com.au
>>>> - Education for Financial Independence: www.mindthemarkets.com.au
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>> "unsubscribe csound"
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>> "unsubscribe csound"
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Most MIDI implementations I'm aware of incorporate amplitude or some  
other qualities outside of pitch and duration, right?
Greg

On Dec 10, 2009, at 11:28 PM, Victor Lazzarini wrote:

> I think you are misunderstanding me. I mean pitch-duration lattice  
> in terms of music that primarily concentrates on the  discrete  
> instances of these two parameters, disregarding all other possible  
> elements that can make up music. Think of a MIDI rendition of 'My  
> favourite things' against John Coltrane's (to demonstrate an  
> obvious idea). I am using MIDI as an example because it's an easy  
> one, but the issues go beyond MIDI.
>
> Victor
>
>
> On 10 Dec 2009, at 14:13, gmschroeder wrote:
>
>> I'm having a hard time following your argument somewhere that  
>> leads to a conclusion that allows digital manipulation and music  
>> to overlap.
>> After all, you *are* ultimately just shifting around ones and  
>> zeroes in "pitch-duration-instrument lattices" whenever you write  
>> music with a computer.
>> Or am I ultimately misunderstanding you here?
>> Greg
>>
>> On Dec 10, 2009, at 11:08 PM, Victor Lazzarini wrote:
>>
>>> Not at all, but it makes not quite the final product.
>>>
>>> I think I have to qualify what I said earlier in relation to  
>>> pitch-duration-instrument lattices. This does not really stop  
>>> music-making, perhaps I should have said that, but makes it  
>>> difficult for a complete musical experience. And for some  
>>> composers, like myself, it really makes it hard to write music  
>>> for, because some of the elements we are used to manipulate are  
>>> absent.
>>>
>>> Victor
>>>
>>>
>>> On 10 Dec 2009, at 12:41, gmschroeder wrote:
>>>
>>>> So your opinion renders that not music?
>>>>
>>>> On Dec 10, 2009, at 9:39 PM, Victor Lazzarini wrote:
>>>>
>>>>> I like the rhythmic manipulations and the overall concept, but  
>>>>> somehow this is despite the nature of the medium in which the  
>>>>> piece is written. In other words, after a little while, I  
>>>>> started to hear the piece as a demo of what could be if this  
>>>>> was played by (definitely more than one) pianos with real  
>>>>> pianists behind them (disregarding whether this is possible or  
>>>>> not).
>>>>> That feeling seems to be inevitable to me (but it's personal).
>>>>>
>>>>> Victor
>>>>>
>>>>> On 10 Dec 2009, at 12:19, Stéphane Rollandin wrote:
>>>>>
>>>>>> Victor Lazzarini a écrit :
>>>>>>> Where can I listen to it?
>>>>>>
>>>>>> directly there (Ogg format):
>>>>>> http://www.zogotounga.net/TGG/zik/var/06-Derive.ogg
>>>>>>
>>>>>> Stef
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> Send bugs reports to this list.
>>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>>>> "unsubscribe csound"
>>>>>
>>>>>
>>>>>
>>>>> Send bugs reports to this list.
>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>>> "unsubscribe csound"
>>>>
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>> "unsubscribe csound"
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>> "unsubscribe csound"
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I agree with you that if I cannot physically hear the idea or its
unique result, there is no music. (I think that's what you are saying,
anyway).

But I don't think I made my point about ideas and the ear very clear.
What I meant is that by working purely with ideas and algorithms,
without any "pre-hearing" if you will, it is possible to make music
that sounds good (and that may be difficult if not impossible to make
any other way).

I think John Cage did that, Xenakis and others did that, and some of
us would indeed be comparatively helpless musically if we could not do
that.

Regards,
Mike

On 12/10/09, Peiman Khosravi  wrote:
> Yes I see this is certainly the case. However I would like to mention
> that there are semi-algorithmic approaches (which by the way I
> wouldn't categorize as a musical "style") that do heavily rely on the
> ear (Murail and OpenMusic, Xenakis, Grisey). Algorithmic in the sense
> of using an automated algorithm to generate certain material: either
> the algorithm itself or the selective process that follows the
> generation can be guided heavily by the ear.
>
> Where music stops to be audible (audible in the sense of the mind's
> ear if you like) I draw the line. I think that's a residue of an
> unfortunate fixation on western notation.
>
> But then that's just me.
>
> Best
> P
>
> On 10 Dec 2009, at 13:45, Michael Gogins wrote:
>
>> In many styles of music the ears are indeed the most important tool --
>> ears understood metaphorically as including some kind of musical
>> perception and taste that surely occurs in the brain not the ears.
>>
>> However, in other styles the mind is the most important tool. I would
>> include process music and algorithmic composition. In these cases
>> without the ideas there is nothing for the ears. The ideas have a
>> distinctly musical component that is heard, if at all, only in the
>> mind. It may not even be heard at all, only "understood" in a musical
>> sense.
>>
>> Regards,
>> Mike
>>
>> On 12/10/09, Peiman Khosravi  wrote:
>>> No but your ears are the tools that you rely on more than anything
>>> else during the process of composition. So they are the most
>>> important
>>> part of your compositional tools. As long as you have them you
>>> needn't
>>> worry about software, or pencil sharpeners. Of course we all want
>>> better software environments that suite our particular creative
>>> patterns and preferences. But my point was that you are still guided
>>> by your ears when actually using the software. We sure don't need
>>> software to compose good music, whatever that is.
>>>
>>> I am thinking of Grisey who created the most spectrally rich works
>>> with no spectral analyzers or FFT to MIDI convertors (on that note
>>> you
>>> should take a look at this:
>>> http://recherche.ircam.fr/equipes/repmus/carpentier/orchidee.html)
>>> .
>>>
>>> P
>>>
>>> On 10 Dec 2009, at 13:13, Stéphane Rollandin wrote:
>>>
>>>>>> but we sorely lack tools for musical composition proper.
>>>>> We still have our ears.
>>>>
>>>> I don't compose with my ears, just as I don't paint with my eyes :)
>>>>
>>>> Stef
>>>>
>>>>
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>>> "unsubscribe csound"
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>> "unsubscribe
>>> csound"
>>
>>
>> --
>> Michael Gogins
>> Irreducible Productions
>> http://www.michael-gogins.com
>> Michael dot Gogins at gmail dot com
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>> "unsubscribe csound"
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"


-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

It's basically a free (not sure about open source) visual language  
(lisp based) similar to OpenMusic. I think it is based on Patch-work  
the predecessor of OM. So I find the two almost identical, but PWGL is  
free!

It also includes a synthesis engine and a tricky to use but much  
better than OM's music notation.
See here for more detail.

http://www2.siba.fi/pwgl/

Best

Peiman

Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"


On 10 Dec 2009, at 14:29, Victor Lazzarini wrote:

> What is PWGL?
> On 10 Dec 2009, at 14:18, Peiman Khosravi wrote:
>
>> Yes OM does it. I've also done this:
>>
>> export partial tracks from SPEAR in txt format and import them in  
>> PWGL. After that it's easy to convert the data to midi.
>>
>> http://www2.siba.fi/pwgl/
>>
>> Many of the openmusic libraries have been ported to PWGL.
>>
>> Best
>>
>> Peiman
>>
>> On 10 Dec 2009, at 14:13, Victor Lazzarini wrote:
>>
>>> I thought things like OpenMusic did something of the kind. There  
>>> must be something like this around. Otherwise it is a good tool to  
>>> be developed.
>>> On 10 Dec 2009, at 13:09, DavidW wrote:
>>>
>>>>
>>>> On 10/12/2009, at 11:01 PM, Victor Lazzarini wrote:
>>>>
>>>>> Because, I guess this is quite a hard thing to do. One thing is  
>>>>> the FFT, the other is how we perceive sound and a third one is  
>>>>> how things are notated. Not a straight one-to-one mapping, I'm  
>>>>> afraid. Like polyphonic pitch-tracking and instrument  
>>>>> segregation, these are very hard things. Also for us: consider  
>>>>> how many years of training it takes to transcribe down a 4-part  
>>>>> Bach chorale from a recording. I also have the impression that  
>>>>> each one of us hears the combination of tones/instrument/timbre  
>>>>> in a complex mix slightly differently and objects do not all get  
>>>>> segregated in the same way in each of our heads. I might be  
>>>>> wrong of course.
>>>>>
>>>> I understand that what you describe is complex. But it wasn't  
>>>> what I was asking. I don't want the software to make its  
>>>> decisions for me. I want algorithmic access to the parameter  
>>>> space so I can determine, say how many partials to render to as  
>>>> chords in the m/s, how many different dynamic levels etc.
>>>>
>>>> Also, I'm not sure the Bach chorale eg is a good one. How many  
>>>> years of training would it take to one calculate 1 million  
>>>> decimal places of pi. And how many seconds for a computer?
>>>>> Last week, I heard of a mobile phone applet that can tell you  
>>>>> what a piece of music is if it is played to it. Not having a  
>>>>> mobile phone myself, I can't say it's true or otherwise. That  
>>>>> seems to me to at least partially negate what I have just said  
>>>>> above, so it just shows how much I know about these things... ;)
>>>>>
>>>> me too.
>>>>> Victor
>>>>>
>>>>> On 10 Dec 2009, at 10:39, DavidW wrote:
>>>>>
>>>>>> I can't quiet believe, for example, that there isn't a public  
>>>>>> domain API solution to the task of FFT to score notation with  
>>>>>> the filter parameters under algorithmic control. Or there is  
>>>>>> and I've missed it perhaps? Even a half-decent FFT to MIDI  
>>>>>> would fill a gap.
>>>>>
>>>>>
>>>>>
>>>>> Send bugs reports to this list.
>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>>> "unsubscribe csound"
>>>>
>>>> ________________________________________________
>>>> Dr David Worrall.
>>>> - Experimental Polymedia:	  worrall.avatar.com.au
>>>> - Sonification: www.sonifiction.com.au
>>>> - Education for Financial Independence: www.mindthemarkets.com.au
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>> "unsubscribe csound"
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>> "unsubscribe csound"
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"

If you are familiar with OM then this feels like home. I only used it  
once in the past to convert partial-tracks to notation for an  
unfulfilled project! Ahhh the spectral dream... :-)

On 10 Dec 2009, at 14:31, Victor Lazzarini wrote:

> Sorry I saw your link. Oh, yes I've seen this elsewehere in a couple  
> of conferences. Looks interesting. Yet Another Language, though.
> On 10 Dec 2009, at 14:29, Victor Lazzarini wrote:
>
>> What is PWGL?
>> On 10 Dec 2009, at 14:18, Peiman Khosravi wrote:
>>
>>> Yes OM does it. I've also done this:
>>>
>>> export partial tracks from SPEAR in txt format and import them in  
>>> PWGL. After that it's easy to convert the data to midi.
>>>
>>> http://www2.siba.fi/pwgl/
>>>
>>> Many of the openmusic libraries have been ported to PWGL.
>>>
>>> Best
>>>
>>> Peiman
>>>
>>> On 10 Dec 2009, at 14:13, Victor Lazzarini wrote:
>>>
>>>> I thought things like OpenMusic did something of the kind. There  
>>>> must be something like this around. Otherwise it is a good tool  
>>>> to be developed.
>>>> On 10 Dec 2009, at 13:09, DavidW wrote:
>>>>
>>>>>
>>>>> On 10/12/2009, at 11:01 PM, Victor Lazzarini wrote:
>>>>>
>>>>>> Because, I guess this is quite a hard thing to do. One thing is  
>>>>>> the FFT, the other is how we perceive sound and a third one is  
>>>>>> how things are notated. Not a straight one-to-one mapping, I'm  
>>>>>> afraid. Like polyphonic pitch-tracking and instrument  
>>>>>> segregation, these are very hard things. Also for us: consider  
>>>>>> how many years of training it takes to transcribe down a 4-part  
>>>>>> Bach chorale from a recording. I also have the impression that  
>>>>>> each one of us hears the combination of tones/instrument/timbre  
>>>>>> in a complex mix slightly differently and objects do not all  
>>>>>> get segregated in the same way in each of our heads. I might be  
>>>>>> wrong of course.
>>>>>>
>>>>> I understand that what you describe is complex. But it wasn't  
>>>>> what I was asking. I don't want the software to make its  
>>>>> decisions for me. I want algorithmic access to the parameter  
>>>>> space so I can determine, say how many partials to render to as  
>>>>> chords in the m/s, how many different dynamic levels etc.
>>>>>
>>>>> Also, I'm not sure the Bach chorale eg is a good one. How many  
>>>>> years of training would it take to one calculate 1 million  
>>>>> decimal places of pi. And how many seconds for a computer?
>>>>>> Last week, I heard of a mobile phone applet that can tell you  
>>>>>> what a piece of music is if it is played to it. Not having a  
>>>>>> mobile phone myself, I can't say it's true or otherwise. That  
>>>>>> seems to me to at least partially negate what I have just said  
>>>>>> above, so it just shows how much I know about these things... ;)
>>>>>>
>>>>> me too.
>>>>>> Victor
>>>>>>
>>>>>> On 10 Dec 2009, at 10:39, DavidW wrote:
>>>>>>
>>>>>>> I can't quiet believe, for example, that there isn't a public  
>>>>>>> domain API solution to the task of FFT to score notation with  
>>>>>>> the filter parameters under algorithmic control. Or there is  
>>>>>>> and I've missed it perhaps? Even a half-decent FFT to MIDI  
>>>>>>> would fill a gap.
>>>>>>
>>>>>>
>>>>>>
>>>>>> Send bugs reports to this list.
>>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>>>> "unsubscribe csound"
>>>>>
>>>>> ________________________________________________
>>>>> Dr David Worrall.
>>>>> - Experimental Polymedia:	  worrall.avatar.com.au
>>>>> - Sonification: www.sonifiction.com.au
>>>>> - Education for Financial Independence: www.mindthemarkets.com.au
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Send bugs reports to this list.
>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>>> "unsubscribe csound"
>>>>
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>> "unsubscribe csound"
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>> "unsubscribe csound"
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I thought Openmusic was part of the IRCAM suite and thus didn't  
qualify accord. to my criteria.
Has it been released, or is this a different beast under another name?

I know Spear - beaut piece of SW.  and I just found  PWGL today.
Now if we can script them into 'doing their thing' and pipe between  
them we have the beginnings of algorithmic control of the task.

On 11/12/2009, at 1:41 AM, Peiman Khosravi wrote:

> If you are familiar with OM then this feels like home. I only used  
> it once in the past to convert partial-tracks to notation for an  
> unfulfilled project! Ahhh the spectral dream... :-)
>
clouds of clouds :-)



David
...
________________________________________________
Dr David Worrall.
- Experimental Polymedia:	  worrall.avatar.com.au
- Sonification: www.sonifiction.com.au
- Education for Financial Independence: www.mindthemarkets.com.au








Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Well the actual source of open-Music is I think available but you need  
the compiler to compile it (which you need to buy!)

But this is a different software that follows up on the predecessor of  
OM (before it was called OM). They are very similar.

Yeah certainly possible to hook SPEARS and PWGL up together (not in  
real-time). I've done it. You could even read data from SPEAR via text  
files, import them into PWGL. Mess them up and send them in real-time  
to Csound via OSC.

P


On 10 Dec 2009, at 15:13, DavidW wrote:

> I thought Openmusic was part of the IRCAM suite and thus didn't  
> qualify accord. to my criteria.
> Has it been released, or is this a different beast under another name?
>
> I know Spear - beaut piece of SW.  and I just found  PWGL today.
> Now if we can script them into 'doing their thing' and pipe between  
> them we have the beginnings of algorithmic control of the task.
>
> On 11/12/2009, at 1:41 AM, Peiman Khosravi wrote:
>
>> If you are familiar with OM then this feels like home. I only used  
>> it once in the past to convert partial-tracks to notation for an  
>> unfulfilled project! Ahhh the spectral dream... :-)
>>
> clouds of clouds :-)
>
>
>
> David
> ...
> ________________________________________________
> Dr David Worrall.
> - Experimental Polymedia:	  worrall.avatar.com.au
> - Sonification: www.sonifiction.com.au
> - Education for Financial Independence: www.mindthemarkets.com.au
>
>
>
>
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On Thu, 2009-12-10 at 12:01 +0000, Victor Lazzarini wrote:
> Last week, I heard of a mobile phone applet that can tell you what a  
> piece of music is if it is played to it. Not having a mobile phone  
> myself, I can't say it's true or otherwise. That seems to me to at  
> least partially negate what I have just said above, so it just shows  
> how much I know about these things... ;)

My phone has this... although it is not an applet. I call a number, put
the phone mic close to the speakers, and I'll get a message with the
name of the artist, track and album of the song.

However this is VERY different from FFT-to-score. This thing probably
calculates some sort of checksum on the music (like
musicbrainz/last.fm), and then queries a database.


-- 
Saludos,
Felipe Sateler

Greetings,

A few years ago OpenMusic was available for Linux in an open-source (not 
GPL, IIRC) version but to my knowledge it has not been maintained. Too 
bad, but then IRCAM hasn't been what I'd call "friendly" towards open 
source.

Last I checked, the author of PWGL doesn't seem to be interested in an 
open-source version of his software.

No-one has so far mentioned Rick Taube's Common Music, one the very few 
pieces of software specifically designed for non-denominational music 
composition, not synthesis. In his book Computer Music: Synthesis, 
Composition, And Performance the composer Charles Dodge praised CM as 
"... a truly general computer language for music composition".

Just FYI.

dp



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Can you explain how you think it works?

Victor
On 10 Dec 2009, at 15:25, Felipe Sateler wrote:

> On Thu, 2009-12-10 at 12:01 +0000, Victor Lazzarini wrote:
>> Last week, I heard of a mobile phone applet that can tell you what a
>> piece of music is if it is played to it. Not having a mobile phone
>> myself, I can't say it's true or otherwise. That seems to me to at
>> least partially negate what I have just said above, so it just shows
>> how much I know about these things... ;)
>
> My phone has this... although it is not an applet. I call a number,  
> put
> the phone mic close to the speakers, and I'll get a message with the
> name of the artist, track and album of the song.
>
> However this is VERY different from FFT-to-score. This thing probably
> calculates some sort of checksum on the music (like
> musicbrainz/last.fm), and then queries a database.
>
>
> -- 
> Saludos,
> Felipe Sateler



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On 11/12/2009, at 1:12 AM, Peiman Khosravi wrote:

> For the record I love those late Liszt works!!
>
me too. And they remain playable with having to practice 24/7 for a  
decade!

> I see your point. I would agree with you as far as you mention the  
> use of MIDI-based instrumental timbres. I just wasn't sure if you  
> were referring to more spectromorphological listening attitudes as  
> well.
>
well yes, Like Victor, I luv them works too. My point was really was  
that I find the bag of techniques for making them very narrow and I'm  
interested in composing work with other motivations, such as a  
stronger sense of in being embodied i.e. perceived vestibularly for  
example.

I'm not setting up the music as sound or notation argument (I  
suggesting it is bigger than, and encompasses both). I'm not  
criticising the music, just not wanting to write like that any more  
and once one move away from cloud-land, the existing software support  
gets pretty thing. Unless you want to interactive work al la Chadabe  
et al. Which, though interesting in itself, sort of avoids the issue.

David

> Best
>
> Peiman
>
> On 10 Dec 2009, at 14:03, DavidW wrote:
>
>> On 11/12/2009, at 12:05 AM, Peiman Khosravi wrote:
>>
>>> I agree with most of what you say. But can you clarify this point?  
>>> I have never heard of timbre-fetishism.
>>>
>>> P
>>>
>>> On 10 Dec 2009, at 04:53, DavidW wrote:
>>>
>>>> most computer music composers seem to be still more enchanted by  
>>>> timbre-fetishisms than music per se.
>>>> Sigh,
>>
>> A term I invented a few years ago to describe a kind of musical  
>> practice in which the effort and thus resulting interest is almost  
>> exclusively timbral.
>> For example, an enormous effort has gone into the building and  
>> tweaking of physical models to produce the sounds musical  
>> instruments make when activated by some dumb activator, but little  
>> effort in the building and tweaking of physical models to a  
>> gesturally rich sequence of activations of such models.
>> The underlying assumption is that the sound of the instrument is  
>> more important that the way it is played. Well that isn't the case  
>> in the acoustic instrumental music world, where an accomplished  
>> musician is evaluated primarily according to their "musicianship",  
>> not the complexity of their instrument.
>>
>> It is this concentration on the synthesis of sounds synthesised  
>> from such models, and the almost exclusive ignoring of the  
>> temporal, gestural means of activating them , that I label "timbre- 
>> fetishism".
>>
>> It's historical origins are in the European practices of dissolving  
>> melody and rhythm into "resonances" (late piano works of Liszt, 2nd  
>> Viennese School and the serialists etc.
>>
...
________________________________________________
Dr David Worrall.
- Experimental Polymedia:	  worrall.avatar.com.au
- Sonification: www.sonifiction.com.au
- Education for Financial Independence: www.mindthemarkets.com.au








Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Thanks Dave, yeah I'd forgotten CM. Will check it out There are some  
good options for MIDI to notation, its the FFT - to MIDI, or some  
higher level of aggregation than a matrix of FFT components, that I  
was pondering.

D.

On 11/12/2009, at 2:31 AM, Dave Phillips wrote:

> Greetings,
>
> A few years ago OpenMusic was available for Linux in an open-source  
> (not GPL, IIRC) version but to my knowledge it has not been  
> maintained. Too bad, but then IRCAM hasn't been what I'd call  
> "friendly" towards open source.
>
> Last I checked, the author of PWGL doesn't seem to be interested in  
> an open-source version of his software.
>
> No-one has so far mentioned Rick Taube's Common Music, one the very  
> few pieces of software specifically designed for non-denominational  
> music composition, not synthesis. In his book Computer Music:  
> Synthesis, Composition, And Performance the composer Charles Dodge  
> praised CM as "... a truly general computer language for music  
> composition".
>
> Just FYI.
>
> dp
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"

________________________________________________
Dr David Worrall.
- Experimental Polymedia:	  worrall.avatar.com.au
- Sonification: www.sonifiction.com.au
- Education for Financial Independence: www.mindthemarkets.com.au








Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Dave Phillips wrote:
>
> Last I checked, the author of PWGL doesn't seem to be interested in an 
> open-source version of his software.
>
But a Google search reveals that a version for Linux was planned for 
2009. I checked the PWGL site, still nothing for Linux.

Only 15 more days to Christmas ! Maybe it will show up under the tree...

Best,

dp





Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I would like to weigh in on open source and music software. This has
several parts.

(1) Any academic institution, and any institution funded with public
money (like IRCAM), should by default return the public's investment
in their work to the public in the form of open source software.

IRCAM should be reprimanded for not doing this.

(2) Over time, music software that is open in some sense (I would
include documented soundfile headers, VST plugins, MIDI, and such even
though their licenses are not public) will survive, and non-open
software will die out. Period. For obvious reasons.

(3) Composers, curators, librarians, and musicologists have a huge
interest in the long-term maintainability of scores and software. The
maintainability any software must stop when the owner goes broke or
goes out of business. Open source software is owned, in effect, by the
public who will never either go broke or go out of business. Any
composer who entrusts his or her scores to closed software is telling
the future "hey, I don't care if you can't read me or study me."

Regards,
Mike







On 12/10/09, Dave Phillips  wrote:
> Greetings,
>
> A few years ago OpenMusic was available for Linux in an open-source (not
> GPL, IIRC) version but to my knowledge it has not been maintained. Too
> bad, but then IRCAM hasn't been what I'd call "friendly" towards open
> source.
>
> Last I checked, the author of PWGL doesn't seem to be interested in an
> open-source version of his software.
>
> No-one has so far mentioned Rick Taube's Common Music, one the very few
> pieces of software specifically designed for non-denominational music
> composition, not synthesis. In his book Computer Music: Synthesis,
> Composition, And Performance the composer Charles Dodge praised CM as
> "... a truly general computer language for music composition".
>
> Just FYI.
>
> dp
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>

I do not know precisely how those work, except that they very likely
don't create an internal "score" of the song. And there are probably
lots of ways to do this.
fdmf is an open source tool that does this (finds "duplicate" tracks in
a MP3 collection, without reading tags or file hashes). The algorithm is
as follows (from the README file):

HOW IT WORKS:

1) Build a list of all music files in music_dir.

FOR EACH FILE:
2) Decode/decompress file to raw binary data.
3) Calculate the energy in 4 frequency bands for each 250 msec chunk.
 3.3) Calculate the sum of the 4 bands for each chunk.
 3.5) Calculate the (b2 + b3):(b0 + b1) ratio for each chunk.
 3.7) Calculate the (b0 + b2):(b1 + b3) ratio for each chunk.
4) Calculate the power spectrum of steps 3.3, 3.5, and 3.7.
5) Spline fit power spectra to a fixed set of frequency points.
6) Quantize the result of step 5 to one bit using median as threshold. 
 6.5) Store the result of step 6 in a database.

FOR EACH POSSIBLE PAIR OF FILES:
7) Count the number of bits that are the same in the two spectra.
8) If the result of 7 exceeds a threshold, and both files still exist,
	print the filenames. 


On Thu, 2009-12-10 at 15:31 +0000, Victor Lazzarini wrote:
> Can you explain how you think it works?
> 
> Victor
> On 10 Dec 2009, at 15:25, Felipe Sateler wrote:
> 
> > On Thu, 2009-12-10 at 12:01 +0000, Victor Lazzarini wrote:
> >> Last week, I heard of a mobile phone applet that can tell you what a
> >> piece of music is if it is played to it. Not having a mobile phone
> >> myself, I can't say it's true or otherwise. That seems to me to at
> >> least partially negate what I have just said above, so it just shows
> >> how much I know about these things... ;)
> >
> > My phone has this... although it is not an applet. I call a number,  
> > put
> > the phone mic close to the speakers, and I'll get a message with the
> > name of the artist, track and album of the song.
> >
> > However this is VERY different from FFT-to-score. This thing probably
> > calculates some sort of checksum on the music (like
> > musicbrainz/last.fm), and then queries a database.
> >
> >
> > -- 
> > Saludos,
> > Felipe Sateler
> 
> 
> 
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"


-- 
Saludos,
Felipe Sateler

Michael Gogins wrote:
> (1) Any academic institution, and any institution funded with public
> money (like IRCAM), should by default return the public's investment
> in their work to the public in the form of open source software.
>
> IRCAM should be reprimanded for not doing this.
>   

At various times various employees of IRCAM have tried to explain to me 
how the institute operates. They have supported some open-source 
projects (jMax, OM, some others), but their subscription stuff is the 
meat on their software table. IIRC France has funding channels set up 
that have no counterparts here in the US. Thus, a research institute 
such as GRAME would not be so likely here in the States, though GRAME 
certainly operates under specific constraints. Alas, my current 
knowledge on this matter is dim. Any IRCAM members on the list who can 
explain its funding and specific responsibilities to the public ?

I'm not slagging IRCAM. It's a cool place, a lot of good music has come 
from it, and I wish it a long healthy existence. However, as you point 
out, open-source is its most logical ally against its software's 
obsolesence.

Btw, I think PWGL, is "PatchWorksGL", a rewrite of IRCAM's famous (and 
proprietary closed-source) PatchWorks.
> (2) Over time, music software that is open in some sense (I would
> include documented soundfile headers, VST plugins, MIDI, and such even
> though their licenses are not public) will survive, and non-open
> software will die out. Period. For obvious reasons.
>   
Agree++

> (3) ... Any composer who entrusts his or her scores to closed software is telling
> the future "hey, I don't care if you can't read me or study me."
>   

Ouch, that's harsh. Workable alternatives may not be available or as 
fully featured, and I've always disagreed with Stallman about this 
topic. I have no beef against using closed-source software, but it is a 
fact that entrusting data to proprietary formats is an invitation to the 
train wreck of "discontinued and unsupported".

Best,

dp



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I think I may suffer a bit from synth engine-fetishism, as I spend a
majority of my time dissecting the Csound language itself, and much
less effort into making interesting music, or even timbres. I'm in the
process of trying to correct this.  :)

Best,
Jake
-- 
The Csound Blog - http://csound.noisepages.com/

On Thu, Dec 10, 2009 at 6:03 AM, DavidW  wrote:
> On 11/12/2009, at 12:05 AM, Peiman Khosravi wrote:
>
>> I agree with most of what you say. But can you clarify this point? I have
>> never heard of timbre-fetishism.
>>
>> P
>>
>> On 10 Dec 2009, at 04:53, DavidW wrote:
>>
>>> most computer music composers seem to be still more enchanted by
>>> timbre-fetishisms than music per se.
>>> Sigh,
>
> A term I invented a few years ago to describe a kind of musical practice in
> which the effort and thus resulting interest is almost exclusively timbral.
> For example, an enormous effort has gone into the building and tweaking of
> physical models to produce the sounds musical instruments make when
> activated by some dumb activator, but little effort in the building and
> tweaking of physical models to a gesturally rich sequence of activations of
> such models.
> The underlying assumption is that the sound of the instrument is more
> important that the way it is played. Well that isn't the case in the
> acoustic instrumental music world, where an accomplished musician is
> evaluated primarily according to their "musicianship", not the complexity of
> their instrument.
>
> It is this concentration on the synthesis of sounds synthesised from such
> models, and the almost exclusive ignoring of the temporal, gestural means of
> activating them , that I label "timbre-fetishism".
>
> It's historical origins are in the European practices of dissolving melody
> and rhythm into "resonances" (late piano works of Liszt, 2nd Viennese School
> and the serialists etc.
>
>
> David
> ________________________________________________
> Dr David Worrall.
> - Experimental Polymedia:         worrall.avatar.com.au
> - Sonification: www.sonifiction.com.au
> - Education for Financial Independence: www.mindthemarkets.com.au
>
>
>
>
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Thanks for your comments.

About the first point, I did know about IRCAM's mixed bag model of
software support. My point is merely that it shouldn't be that way. If
IRCAM is funded by the public, the public should get what IRCAM makes.
Otherwise, in my view, IRCAM is ripping off the public.

About the last point, of course commercial software (sequencers such
as Logic or Cubase spring to mind) can be far more capable than any
current open source software. One then has a nasty choice... but It's
not THAT nasty because commercial sequencers export MIDI and may
export MusicXML, both of which are "open" formats in my definition.
You composers: if you want your computer files usable in future
archive MIDI and/or MusicXML as well as the proprietary formats.

I also disagree with Stallman in that I prefer and use LGPL over GPL,
i.e. I prefer and use "open source" over "free software."

Nothing I say should be construed as saying I don't like commercial
software and don't respect such truly excellent products as Sibelius
or the Pianoteq plugin. I'm happy to pay for stuff like this. Sibelius
of course imports and exports MusicXML.

At this point, the only commercial software that I truly need and
regularly use is the Pianoteq. If I did more notation, I would add
Sibelius. If I did more overdubbing, I would add a commercial
sequencer.

I do find myself relying more and more on open source software,
working more on Linux, and so on. This is partly driven by upset over
old computer music pieces of mine that I couldn't re-realize 10 years
after they were made because the original commercial software I used
is just gone, and partly driven by free tools on Linux.

Regards,
Mike

On 12/10/09, Dave Phillips  wrote:
> Michael Gogins wrote:
>> (1) Any academic institution, and any institution funded with public
>> money (like IRCAM), should by default return the public's investment
>> in their work to the public in the form of open source software.
>>
>> IRCAM should be reprimanded for not doing this.
>>
>
> At various times various employees of IRCAM have tried to explain to me
> how the institute operates. They have supported some open-source
> projects (jMax, OM, some others), but their subscription stuff is the
> meat on their software table. IIRC France has funding channels set up
> that have no counterparts here in the US. Thus, a research institute
> such as GRAME would not be so likely here in the States, though GRAME
> certainly operates under specific constraints. Alas, my current
> knowledge on this matter is dim. Any IRCAM members on the list who can
> explain its funding and specific responsibilities to the public ?
>
> I'm not slagging IRCAM. It's a cool place, a lot of good music has come
> from it, and I wish it a long healthy existence. However, as you point
> out, open-source is its most logical ally against its software's
> obsolesence.
>
> Btw, I think PWGL, is "PatchWorksGL", a rewrite of IRCAM's famous (and
> proprietary closed-source) PatchWorks.
>> (2) Over time, music software that is open in some sense (I would
>> include documented soundfile headers, VST plugins, MIDI, and such even
>> though their licenses are not public) will survive, and non-open
>> software will die out. Period. For obvious reasons.
>>
> Agree++
>
>> (3) ... Any composer who entrusts his or her scores to closed software is
>> telling
>> the future "hey, I don't care if you can't read me or study me."
>>
>
> Ouch, that's harsh. Workable alternatives may not be available or as
> fully featured, and I've always disagreed with Stallman about this
> topic. I have no beef against using closed-source software, but it is a
> fact that entrusting data to proprietary formats is an invitation to the
> train wreck of "discontinued and unsupported".
>
> Best,
>
> dp
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>

> Well, not for everyone. If your music is based on things that cannot be 
> capture in that medium, then it can't be a good benchmark.

ok, point taken. it's just a benchmark, among many others.

Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

But that is still another dimension of the lattice. The principle does not 
change.

Victor
----- Original Message ----- 
From: "gmschroeder" 
To: 
Sent: Thursday, December 10, 2009 2:32 PM
Subject: [Csnd] Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: 
[OT] Human speech is music to out ears


Most MIDI implementations I'm aware of incorporate amplitude or some
other qualities outside of pitch and duration, right?
Greg

On Dec 10, 2009, at 11:28 PM, Victor Lazzarini wrote:

> I think you are misunderstanding me. I mean pitch-duration lattice  in 
> terms of music that primarily concentrates on the  discrete  instances of 
> these two parameters, disregarding all other possible  elements that can 
> make up music. Think of a MIDI rendition of 'My  favourite things' against 
> John Coltrane's (to demonstrate an  obvious idea). I am using MIDI as an 
> example because it's an easy  one, but the issues go beyond MIDI.
>
> Victor
>
>
> On 10 Dec 2009, at 14:13, gmschroeder wrote:
>
>> I'm having a hard time following your argument somewhere that  leads to a 
>> conclusion that allows digital manipulation and music  to overlap.
>> After all, you *are* ultimately just shifting around ones and  zeroes in 
>> "pitch-duration-instrument lattices" whenever you write  music with a 
>> computer.
>> Or am I ultimately misunderstanding you here?
>> Greg
>>
>> On Dec 10, 2009, at 11:08 PM, Victor Lazzarini wrote:
>>
>>> Not at all, but it makes not quite the final product.
>>>
>>> I think I have to qualify what I said earlier in relation to 
>>> pitch-duration-instrument lattices. This does not really stop 
>>> music-making, perhaps I should have said that, but makes it  difficult 
>>> for a complete musical experience. And for some  composers, like myself, 
>>> it really makes it hard to write music  for, because some of the 
>>> elements we are used to manipulate are  absent.
>>>
>>> Victor
>>>
>>>
>>> On 10 Dec 2009, at 12:41, gmschroeder wrote:
>>>
>>>> So your opinion renders that not music?
>>>>
>>>> On Dec 10, 2009, at 9:39 PM, Victor Lazzarini wrote:
>>>>
>>>>> I like the rhythmic manipulations and the overall concept, but 
>>>>> somehow this is despite the nature of the medium in which the  piece 
>>>>> is written. In other words, after a little while, I  started to hear 
>>>>> the piece as a demo of what could be if this  was played by 
>>>>> (definitely more than one) pianos with real  pianists behind them 
>>>>> (disregarding whether this is possible or  not).
>>>>> That feeling seems to be inevitable to me (but it's personal).
>>>>>
>>>>> Victor
>>>>>
>>>>> On 10 Dec 2009, at 12:19, Stéphane Rollandin wrote:
>>>>>
>>>>>> Victor Lazzarini a écrit :
>>>>>>> Where can I listen to it?
>>>>>>
>>>>>> directly there (Ogg format):
>>>>>> http://www.zogotounga.net/TGG/zik/var/06-Derive.ogg
>>>>>>
>>>>>> Stef
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> Send bugs reports to this list.
>>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body 
>>>>>> "unsubscribe csound"
>>>>>
>>>>>
>>>>>
>>>>> Send bugs reports to this list.
>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body 
>>>>> "unsubscribe csound"
>>>>
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body 
>>>> "unsubscribe csound"
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body 
>>> "unsubscribe csound"
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  "unsubscribe 
>> csound"
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  "unsubscribe 
> csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe 
csound"= 



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

on 2009-12-10 at 12:04 Victor Lazzarini wrote:

>But my point is that as far as I am concerned, I can't see that the  
>environment you suggested allows for much musicality to be expressed.
>That's only my opinion.

if i understand correctly what stéphane proposes, it describes very
well the complete volumes of studies for player piano by conlon
nancarrow. which means the bulk of the production of one of the
most important composers of the 20th century. (but that also is "just
my opinion"). one timbre, one dynamic, the musical structure is just
the pitches and times. as i see it, not such a bad idea after all. i'm
not sure i understand what is "musicality" and how it can be
"expressed", but to my ears nancarrow sounds very "musical". (not to
mention others that could be described in a similar way).


>On 10 Dec 2009, at 11:58, Stéphane Rollandin wrote:
>>
>> no, what I meant is much simpler: by cornering the composition  
>> activity into a very limited form, MIDI notes (so, a crude  
>> representation) played by a piano soundfont (not a pianist,  
>> precisely), then we can better appreciate the sheer musicality of  
>> the compositions.




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

gmschroeder wrote:
> Most MIDI implementations I'm aware of incorporate amplitude or some 
> other qualities outside of pitch and duration, right?

Velocity values are typically associated with amplitude, but of course 
they could be routed to anything on the receiver (assuming such a 
routing is available, as in Csound).

The MIDI note-on event include channel, pitch level (note number), and a 
velocity value. Any further modulation of the event comes from 
controllers (pitch bend, aftertouch, mod wheel, etc), which are not 
components of the note-on event.

[OT]: rerererererererererere - Jeez those things are annoying.

Best,

dp



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Hi Mike,
Not to pick an argument for  the sake of it, but in order to clarify  
what I'm saying, which you're perfectly at liberty to disagree with.

On 11/12/2009, at 12:45 AM, Michael Gogins wrote:
> In many styles of music the ears are indeed the most important tool --
> ears understood metaphorically as including some kind of musical
> perception and taste that surely occurs in the brain not the ears.

Strictly speaking not a metaphor but a synecdoche. And not necessarily  
just he brain. I'd be happier with "the neurone system"

> However, in other styles the mind is the most important tool. I would
> include process music and algorithmic composition. In these cases
> without the ideas there is nothing for the ears. The ideas have a
> distinctly musical component that is heard, if at all, only in the
> mind. It may not even be heard at all, only "understood" in a musical
> sense.

> Regards,
> Mike

I understand this as the traditional explanation. From my perspective,  
it has a few problems, related to my "the neurone system" example.
Firstly, you're using "the mind" in two different ways, neither of  
which work for me, but that is besides the point in the first instance.
1. 'The mind is a tool': This 'mind' thing is not like any other tool,  
or even a physical organ (which are also not tools), so for me, the  
idea of what it is in this context is not clear. The majority of post  
Kantian philosophers would want to argue with you about it. Especially  
Heidegger, Merleau-Ponty etc.

2. "Heard only in the mind": Is this mind thing included under the ear/ 
hearing synecdoche we were talking about earlier? There is so much in  
the concept of musical "understanding" that is opaque, at least to me.

David
...
________________________________________________
Dr David Worrall.
- Experimental Polymedia:	  worrall.avatar.com.au
- Sonification: www.sonifiction.com.au
- Education for Financial Independence: www.mindthemarkets.com.au








Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

When I used to run a university centre, we were signed up members to  
the IRCAM suite. And we had the same issue as reported by Francois.  
And now I'm freelance, I resist paying for such tools, not just  
because I'm a Scrooge, but for all the reasons Mike has outlined in  
favour of Open Source (OS).

A couple of other points wrt this thread.
* OS doesn't guarantee against obsolescence, but it does mean that you  
have the (sometimes very inconvenient) possibility of recovery if  
you're prepared to hack it. Personally I have a set of criteria over  
and above the OS one. There are projects out there, for example, some  
quite mature, which use their OSness as a 'lure' to the unsuspecting,  
in the expectation that once you're (your department) is committed and  
you've worked out that you've committed to a nightmare, you'll engage  
the 'authors' as consultants.

* I agree. w. Mike that having the data in a generic format is a wise  
safety step. The same applies to other SW too, like word processors  
etc. I've started archiving everything that can be archived in ascii,  
in ascii, as a matter of precaution, storage space not being a problem  
these days. Notation is another storage format when applicable, and is  
unlikely to become obsolescent any time soon.

* The economics of closed-source make it an increasingly risky  
enterprise, especially for big projects. The Aust. National Library  
has completely converted to open source software and the improvement  
is remarkable. They now put their considerable resources into  
continued development as do similar institutions, and so there is no  
way a stand-a-lone app. can compete. They apply the same model to  
massive amounts of OCR human eyeball OCR error correction with  
stunning results.

* The API/library SW model is another example of the 'mitigation  
against obsolescence' approach.

David

On 10/12/2009, at 10:40 AM, Fran=8Dcois_Roux wrote:

> Michael Gogins wrote:
>> Thanks for your comments.
>>
>> About the first point, I did know about IRCAM's mixed bag model of
>> software support. My point is merely that it shouldn't be that way.  
>> If
>> IRCAM is funded by the public, the public should get what IRCAM  
>> makes.
>> Otherwise, in my view, IRCAM is ripping off the public.
>>
>
> That why here in our "composition departement" of
> CNSMD-Lyon (Conservatoire National Supérieur de Musique de Lyon,  
> France).
> we use PWGL !!
> We have some OM software that our public french state governement  
> paid for us,
> but which interest for students when they go away from here.
>
> -- 
> oxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxox
> Francois Roux
> 9 rue Vaudrey - 69003 - Lyon
> Tel : 09 51 77 30 39 / 06 67 28 05 63
> E-mail : xrfrx@free.fr
> oxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxox


________________________________________________
Dr David Worrall.
- Experimental Polymedia:	  worrall.avatar.com.au
- Sonification: www.sonifiction.com.au
- Education for Financial Independence: www.mindthemarkets.com.au








Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I did not know Conlon Nancarrow;
thanks for the pointer,

Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On 10 Dec 2009, at 13:26, Peiman Khosravi wrote:

> on that note you should take a look at this: http:// 
> recherche.ircam.fr/equipes/repmus/carpentier/orchidee.html

Interesting - especially as I was at the première of Speakings

I presume the results were selected to be "non-trivial", but can't  
help wondering whether one of the possible solutions to sound example  
1.1 was "solo bassoon"? (And that was calculated WITHOUT recourse to  
computer - there must be something in this formal music study after  
all!  ;-) )

Suppose it's one way to keep musicians in employment

Seriously, this does look like a fascinating tool - if you want to  
know how much of a headache this can be for composers, take a look at  
the preface to Stockhausen's Inori (and that was just for dynamics!)

Only problem is that it's another IRCAM product - prepare to be fleeced…

Joe




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Hi Michael Gogins and others,

Algorithmic music and more importantly, non-algorithmic music--this is an
area of enormous interest to me, both philosophically and practically.

Human consciousness is largely a non-algorithmic process, otherwise, we'd
have a hard time noticing patterns which are 'stuck' or 'cliches'....most
music produced in history has been non-algorithmic, and most certainly, the
non-algorithmic universe is infinitely vaster (more interesting, too?) than
the algorithmic universe.  (I mean this in the literal, mathematical sense
of the transfinite numbers of Georg Cantor (see below) as well as
poetically/figuratively)

Think about it from a purely mathematical point-of-view: there are an
infinite number of computable irrational numbers (sqrt of 2 for instance).
Imagine that a given irrational number codes in a one-to-one way a given
score or soundfile for a piece of music. The (strict) algorithmic or
process-oriented composer's focus is on these numbers whose properties are
potentially deterministic: i.e., the piece can be thought of as the outcome
of a more-or-less 'fixed' process.

Now think about the numbers which correspond to 'tweaked' computable
irrationals. For instance, the square root of 2 with fifty of its digits
changed, perhaps the first 50 digits in prime numbered 'slots' in its
decimal representation. Many composers of algorithmic music do such
things---they 'tweak the output'....

Now think about the greater infinity--in the sense of Cantor's number
theory, where there are 'transfinite sets'---the set of irrational numbers
which are non-computable in a *direct* sense by any algorithm. IOW, they
require a conscious being to creatively search them out. There are
infinitely more of these numbers, literally. IOW, this infinity is GREATER
than the infinity of algorithmic, computable numbers. For those without
background, I encourage you to read about Georg Cantor, and as it relates to
programs and computing, a book called "The Computational Beauty of Nature"

Of course, many of these numbers can be thought of as random strings,
indeed, most of them are. However, there is an important subset of them that
are NOT random, IOW, they contain a clear structure, but a structure which
is provably non-algorithmically generated in the easy way that random
numbers can be. My conjecture is that such numbers would/could be
code-strings for a piece of music which has clear structure, but also has
deep cohesiveness on multiple levels, and also a rich set of surprising and
enjoyable details. Clearly, there is a sense of motives, repetition, and
self-reference at play, just like one has in most traditionally-written
music that gives a sense of 'organic cohesion'. Such enjoyable and
surprising stray details in a traditional masterwork might be possibly
simulated by randomness, but most appear to be directed by an irreducible
sense that a conscious being's choices are at work. David Cope's work has
come close to this, but mostly because he does an awful lot of massaging of
data which was pre-coded, like formal structures and Markov chain
statistics. None of these things would by any stretch likely arise by pure
chance. The more interesting question is "what forces intrisic or extrinsic
to the composer produce such structures" Surely cultural history and
training play a part, as well as pure chance....but they combine in such a
way that I think may forever belie analysis. It might be too complex a thing
for the human mind to actually know itself, just as a mouse will never
fathom what a prime number is.

So, anyway, imagine an irrational number that codes Stravinsky's "Rite of
Spring" or Beethoven's "Eroica" symphony....certainly these pieces are not
'computable' in any meaningful way...they go beyond syntax and into
semantics---meaning, they require a conscious being to create them. No
algorithm would simply spit out the "Eroica"...unless the program were
simply to print out the precoded representation of the "Eroica", which is
not computing it in the sense I mean, but merely printing stored data.

A simple example suffices to show the limits of a strictly, or even largely,
algorithmic approach to composition: the choices an algorithm makes have to
be set by parameters made available to the algorithm beforehand, by the
author: for instance--no algorithm will start spitting out triplets because
it is feeling 'bored' or 'playful'---it will spit out triplets because the
algorithm's author decided to include them as a parameter choice---and
algorithmic music will always be limited to the parameter choices which are
built-in to the algorithm. Further, the algorithm wouldn't ever consider
such a choice to be a 'meaninful gesture' in any semantic sense, whereas in
traditional music one sees meaning in these gestures in an extra-musical
sense *all the time*---think about the music for cartoons or films...all the
humorous gagging going on in Carl Stalling's Looney Tunes scores defy any
attempt at an algorithmic description.

No sense of 'escaping the system' in a *conscious* way can ever come out of
such a system unless it's hard-coded in, which, is neither a 'true' escape,
and, at least right now, requires a conscious being like a human to fulfill.
Think about all the "fractal music" you've heard. How much of it ever has
anything remotely structured like a scale passage or a simple arpeggiated
triad? The larger the scale passage, the less likely it will appear by
chance. And so much great music uses such extremely non-random materials as
long scale passages and broken chords, to give two examples.

So, I think it will always come back to a human's aesthetic judgment and the
fact that we are conscious pilots who want to fly through the terrain of
musical possibilities.

All this being said, I am personally interested in the potential of
algorithmic music, and some of it's results. In fact, I've written some
myself, and it can be interesting and satisfying to do so. Mostly, when the
effort is collaborative with a highly critical composer who *listens* to the
results and is not a-priori 'detached' from the results. Meaning they can
jump in and change and edit the output of a program to fit their taste.
Otherwise, I mostly find purely algorithmic music dreadfully cold and boring
to hear. However, I do find algorithms a convenient way to spur creativity,
and to escape one's own predilections by introducing constraints and random
choices one wouldn't have made. This leads to I think an expanded palette of
creative motifs that I think of as a dialogical process.

All of this, of course, echoes the really pretentious extremes of scientific
reductionism. I'd love to be proven wrong, and I think it'd be damn great to
have "equation music" on a par with say Bach or Beethoven or the Beatles,
but I remain very skeptical.

What are your thoughts?

Best,
AKJ


Michael Gogins-2 wrote:
> 
> I agree with you that if I cannot physically hear the idea or its
> unique result, there is no music. (I think that's what you are saying,
> anyway).
> 
> But I don't think I made my point about ideas and the ear very clear.
> What I meant is that by working purely with ideas and algorithms,
> without any "pre-hearing" if you will, it is possible to make music
> that sounds good (and that may be difficult if not impossible to make
> any other way).
> 
> I think John Cage did that, Xenakis and others did that, and some of
> us would indeed be comparatively helpless musically if we could not do
> that.
> 
> Regards,
> Mike
> 
> On 12/10/09, Peiman Khosravi  wrote:
>> Yes I see this is certainly the case. However I would like to mention
>> that there are semi-algorithmic approaches (which by the way I
>> wouldn't categorize as a musical "style") that do heavily rely on the
>> ear (Murail and OpenMusic, Xenakis, Grisey). Algorithmic in the sense
>> of using an automated algorithm to generate certain material: either
>> the algorithm itself or the selective process that follows the
>> generation can be guided heavily by the ear.
>>
>> Where music stops to be audible (audible in the sense of the mind's
>> ear if you like) I draw the line. I think that's a residue of an
>> unfortunate fixation on western notation.
>>
>> But then that's just me.
>>
>> Best
>> P
>>
>> On 10 Dec 2009, at 13:45, Michael Gogins wrote:
>>
>>> In many styles of music the ears are indeed the most important tool --
>>> ears understood metaphorically as including some kind of musical
>>> perception and taste that surely occurs in the brain not the ears.
>>>
>>> However, in other styles the mind is the most important tool. I would
>>> include process music and algorithmic composition. In these cases
>>> without the ideas there is nothing for the ears. The ideas have a
>>> distinctly musical component that is heard, if at all, only in the
>>> mind. It may not even be heard at all, only "understood" in a musical
>>> sense.
>>>
>>> Regards,
>>> Mike
>>>
>>> On 12/10/09, Peiman Khosravi  wrote:
>>>> No but your ears are the tools that you rely on more than anything
>>>> else during the process of composition. So they are the most
>>>> important
>>>> part of your compositional tools. As long as you have them you
>>>> needn't
>>>> worry about software, or pencil sharpeners. Of course we all want
>>>> better software environments that suite our particular creative
>>>> patterns and preferences. But my point was that you are still guided
>>>> by your ears when actually using the software. We sure don't need
>>>> software to compose good music, whatever that is.
>>>>
>>>> I am thinking of Grisey who created the most spectrally rich works
>>>> with no spectral analyzers or FFT to MIDI convertors (on that note
>>>> you
>>>> should take a look at this:
>>>> http://recherche.ircam.fr/equipes/repmus/carpentier/orchidee.html)
>>>> .
>>>>
>>>> P
>>>>
>>>> On 10 Dec 2009, at 13:13, Stéphane Rollandin wrote:
>>>>
>>>>>>> but we sorely lack tools for musical composition proper.
>>>>>> We still have our ears.
>>>>>
>>>>> I don't compose with my ears, just as I don't paint with my eyes :)
>>>>>
>>>>> Stef
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Send bugs reports to this list.
>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>>>> "unsubscribe csound"
>>>>
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>>> "unsubscribe
>>>> csound"
>>>
>>>
>>> --
>>> Michael Gogins
>>> Irreducible Productions
>>> http://www.michael-gogins.com
>>> Michael dot Gogins at gmail dot com
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>> "unsubscribe csound"
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
> 
> 
> -- 
> Michael Gogins
> Irreducible Productions
> http://www.michael-gogins.com
> Michael dot Gogins at gmail dot com
> 
> 
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
> 

-- 
View this message in context: http://old.nabble.com/-OT--Human-speech-is-music-to-out-ears-tp26629182p26770920.html
Sent from the Csound - General mailing list archive at Nabble.com.



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I've given a good deal of thought to these issues. "A Tutorial
Introduction to CsoundAc" contains some of these thoughts. I quote the
relevant portion below. If I'm not mistaken, we have a somewhat
similar basic viewpoint.

However, I think you are too pessimistic about what can be done with
finitary means which, if the world has a beginning and an end and a
horizon, which it definitely seems to do, is all we have. In other
words, I think that the transcendental reason applies itself to choice
of finitary means, i.e. generative algorithms, just as it applies
itself directly to writing scores or poems or whatever. In other
words, creating the algorithm is just as creative as creating the
score. It's just that in my view you can do many things by creating
algorithms that you simply can't do any other way.

Quotation follows...

Regards,
Mike

Computability is a metaphysical question: Is an entity definable,
knowable, finite, or not? – So, any question concerning the
computability of music is a question of the metaphysics of music.

My considerations are grounded in several obvious and unassailable
facts, from which far-reaching inferences can be drawn.

In the first place, works of music are physical phenomena – sounds.
Each work of music is a sound of finite tessitura, dynamic range, and
duration, which persists without gaps until it ends. Human hearing,
too, has limited resolution. Therefore, each work of music can be
considered to be completely represented by a digital recording of
sufficient bandwidth: a certain finite series of 0s and 1s. Or again,
a work of music is a sequence of finite complexity. So of course all
such sequences are computable, though not necessarily in a reasonable
period of time.

But works of music have no meaning unless they are heard, and hearing
music is an inward, subjective experience. It would be a mistake to
suppose that the experience of music is not caused by the physical
sound, yet it would equally be a mistake to suppose that the physical
sound has any meaning without the experience of hearing it. Forgetting
the physical causation of music leads to the error of cultural
relativism sometimes found in postmodernism, and forgetting the inward
hearing of music leads to the error of objectifying music typical of
scientism.

Now although any given work of music is a finite object, the art of
music consists of a sequence of such works, each distinct from all
others. Mathematically speaking, this series could continue
indefinitely. The art of music can thus be considered to be an
infinite series of 0s and 1s.

In other words, considered in the concrete, as a completed series,
from the viewpoint of God if you will, the art of music is a series of
countably infinite length and countably infinite complexity. But
considered in the abstract, as an uncompleted series, from the
viewpoint of creatures if you will, the art of music is an infinite
set of possible series, each of countably infinite length. This set of
possible series can be diagonalized (exactly as in the halting
theorem) to prove that it is of uncountable infinity and uncountable
complexity. Since the series has not been completed and the art of
music has not been finished, the reality of our situation is that the
art of music is effectively uncomputable.

The most important implication of this, for us, is that it would be
absurd to hope for computers to generate the art of music as a whole.
But as I have tried to show, they can be very useful indeed in
computing works of music.




On 12/13/09, Aaron Krister Johnson  wrote:
>
> Hi Michael Gogins and others,
>
> Algorithmic music and more importantly, non-algorithmic music--this is an
> area of enormous interest to me, both philosophically and practically.
>
> Human consciousness is largely a non-algorithmic process, otherwise, we'd
> have a hard time noticing patterns which are 'stuck' or 'cliches'....most
> music produced in history has been non-algorithmic, and most certainly, the
> non-algorithmic universe is infinitely vaster (more interesting, too?) than
> the algorithmic universe.  (I mean this in the literal, mathematical sense
> of the transfinite numbers of Georg Cantor (see below) as well as
> poetically/figuratively)
>
> Think about it from a purely mathematical point-of-view: there are an
> infinite number of computable irrational numbers (sqrt of 2 for instance).
> Imagine that a given irrational number codes in a one-to-one way a given
> score or soundfile for a piece of music. The (strict) algorithmic or
> process-oriented composer's focus is on these numbers whose properties are
> potentially deterministic: i.e., the piece can be thought of as the outcome
> of a more-or-less 'fixed' process.
>
> Now think about the numbers which correspond to 'tweaked' computable
> irrationals. For instance, the square root of 2 with fifty of its digits
> changed, perhaps the first 50 digits in prime numbered 'slots' in its
> decimal representation. Many composers of algorithmic music do such
> things---they 'tweak the output'....
>
> Now think about the greater infinity--in the sense of Cantor's number
> theory, where there are 'transfinite sets'---the set of irrational numbers
> which are non-computable in a *direct* sense by any algorithm. IOW, they
> require a conscious being to creatively search them out. There are
> infinitely more of these numbers, literally. IOW, this infinity is GREATER
> than the infinity of algorithmic, computable numbers. For those without
> background, I encourage you to read about Georg Cantor, and as it relates to
> programs and computing, a book called "The Computational Beauty of Nature"
>
> Of course, many of these numbers can be thought of as random strings,
> indeed, most of them are. However, there is an important subset of them that
> are NOT random, IOW, they contain a clear structure, but a structure which
> is provably non-algorithmically generated in the easy way that random
> numbers can be. My conjecture is that such numbers would/could be
> code-strings for a piece of music which has clear structure, but also has
> deep cohesiveness on multiple levels, and also a rich set of surprising and
> enjoyable details. Clearly, there is a sense of motives, repetition, and
> self-reference at play, just like one has in most traditionally-written
> music that gives a sense of 'organic cohesion'. Such enjoyable and
> surprising stray details in a traditional masterwork might be possibly
> simulated by randomness, but most appear to be directed by an irreducible
> sense that a conscious being's choices are at work. David Cope's work has
> come close to this, but mostly because he does an awful lot of massaging of
> data which was pre-coded, like formal structures and Markov chain
> statistics. None of these things would by any stretch likely arise by pure
> chance. The more interesting question is "what forces intrisic or extrinsic
> to the composer produce such structures" Surely cultural history and
> training play a part, as well as pure chance....but they combine in such a
> way that I think may forever belie analysis. It might be too complex a thing
> for the human mind to actually know itself, just as a mouse will never
> fathom what a prime number is.
>
> So, anyway, imagine an irrational number that codes Stravinsky's "Rite of
> Spring" or Beethoven's "Eroica" symphony....certainly these pieces are not
> 'computable' in any meaningful way...they go beyond syntax and into
> semantics---meaning, they require a conscious being to create them. No
> algorithm would simply spit out the "Eroica"...unless the program were
> simply to print out the precoded representation of the "Eroica", which is
> not computing it in the sense I mean, but merely printing stored data.
>
> A simple example suffices to show the limits of a strictly, or even largely,
> algorithmic approach to composition: the choices an algorithm makes have to
> be set by parameters made available to the algorithm beforehand, by the
> author: for instance--no algorithm will start spitting out triplets because
> it is feeling 'bored' or 'playful'---it will spit out triplets because the
> algorithm's author decided to include them as a parameter choice---and
> algorithmic music will always be limited to the parameter choices which are
> built-in to the algorithm. Further, the algorithm wouldn't ever consider
> such a choice to be a 'meaninful gesture' in any semantic sense, whereas in
> traditional music one sees meaning in these gestures in an extra-musical
> sense *all the time*---think about the music for cartoons or films...all the
> humorous gagging going on in Carl Stalling's Looney Tunes scores defy any
> attempt at an algorithmic description.
>
> No sense of 'escaping the system' in a *conscious* way can ever come out of
> such a system unless it's hard-coded in, which, is neither a 'true' escape,
> and, at least right now, requires a conscious being like a human to fulfill.
> Think about all the "fractal music" you've heard. How much of it ever has
> anything remotely structured like a scale passage or a simple arpeggiated
> triad? The larger the scale passage, the less likely it will appear by
> chance. And so much great music uses such extremely non-random materials as
> long scale passages and broken chords, to give two examples.
>
> So, I think it will always come back to a human's aesthetic judgment and the
> fact that we are conscious pilots who want to fly through the terrain of
> musical possibilities.
>
> All this being said, I am personally interested in the potential of
> algorithmic music, and some of it's results. In fact, I've written some
> myself, and it can be interesting and satisfying to do so. Mostly, when the
> effort is collaborative with a highly critical composer who *listens* to the
> results and is not a-priori 'detached' from the results. Meaning they can
> jump in and change and edit the output of a program to fit their taste.
> Otherwise, I mostly find purely algorithmic music dreadfully cold and boring
> to hear. However, I do find algorithms a convenient way to spur creativity,
> and to escape one's own predilections by introducing constraints and random
> choices one wouldn't have made. This leads to I think an expanded palette of
> creative motifs that I think of as a dialogical process.
>
> All of this, of course, echoes the really pretentious extremes of scientific
> reductionism. I'd love to be proven wrong, and I think it'd be damn great to
> have "equation music" on a par with say Bach or Beethoven or the Beatles,
> but I remain very skeptical.
>
> What are your thoughts?
>
> Best,
> AKJ
>
>
> Michael Gogins-2 wrote:
>>
>> I agree with you that if I cannot physically hear the idea or its
>> unique result, there is no music. (I think that's what you are saying,
>> anyway).
>>
>> But I don't think I made my point about ideas and the ear very clear.
>> What I meant is that by working purely with ideas and algorithms,
>> without any "pre-hearing" if you will, it is possible to make music
>> that sounds good (and that may be difficult if not impossible to make
>> any other way).
>>
>> I think John Cage did that, Xenakis and others did that, and some of
>> us would indeed be comparatively helpless musically if we could not do
>> that.
>>
>> Regards,
>> Mike
>>
>> On 12/10/09, Peiman Khosravi  wrote:
>>> Yes I see this is certainly the case. However I would like to mention
>>> that there are semi-algorithmic approaches (which by the way I
>>> wouldn't categorize as a musical "style") that do heavily rely on the
>>> ear (Murail and OpenMusic, Xenakis, Grisey). Algorithmic in the sense
>>> of using an automated algorithm to generate certain material: either
>>> the algorithm itself or the selective process that follows the
>>> generation can be guided heavily by the ear.
>>>
>>> Where music stops to be audible (audible in the sense of the mind's
>>> ear if you like) I draw the line. I think that's a residue of an
>>> unfortunate fixation on western notation.
>>>
>>> But then that's just me.
>>>
>>> Best
>>> P
>>>
>>> On 10 Dec 2009, at 13:45, Michael Gogins wrote:
>>>
>>>> In many styles of music the ears are indeed the most important tool --
>>>> ears understood metaphorically as including some kind of musical
>>>> perception and taste that surely occurs in the brain not the ears.
>>>>
>>>> However, in other styles the mind is the most important tool. I would
>>>> include process music and algorithmic composition. In these cases
>>>> without the ideas there is nothing for the ears. The ideas have a
>>>> distinctly musical component that is heard, if at all, only in the
>>>> mind. It may not even be heard at all, only "understood" in a musical
>>>> sense.
>>>>
>>>> Regards,
>>>> Mike
>>>>
>>>> On 12/10/09, Peiman Khosravi  wrote:
>>>>> No but your ears are the tools that you rely on more than anything
>>>>> else during the process of composition. So they are the most
>>>>> important
>>>>> part of your compositional tools. As long as you have them you
>>>>> needn't
>>>>> worry about software, or pencil sharpeners. Of course we all want
>>>>> better software environments that suite our particular creative
>>>>> patterns and preferences. But my point was that you are still guided
>>>>> by your ears when actually using the software. We sure don't need
>>>>> software to compose good music, whatever that is.
>>>>>
>>>>> I am thinking of Grisey who created the most spectrally rich works
>>>>> with no spectral analyzers or FFT to MIDI convertors (on that note
>>>>> you
>>>>> should take a look at this:
>>>>> http://recherche.ircam.fr/equipes/repmus/carpentier/orchidee.html)
>>>>> .
>>>>>
>>>>> P
>>>>>
>>>>> On 10 Dec 2009, at 13:13, Stéphane Rollandin wrote:
>>>>>
>>>>>>>> but we sorely lack tools for musical composition proper.
>>>>>>> We still have our ears.
>>>>>>
>>>>>> I don't compose with my ears, just as I don't paint with my eyes :)
>>>>>>
>>>>>> Stef
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> Send bugs reports to this list.
>>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>>>>> "unsubscribe csound"
>>>>>
>>>>>
>>>>>
>>>>> Send bugs reports to this list.
>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>>>> "unsubscribe
>>>>> csound"
>>>>
>>>>
>>>> --
>>>> Michael Gogins
>>>> Irreducible Productions
>>>> http://www.michael-gogins.com
>>>> Michael dot Gogins at gmail dot com
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>>> "unsubscribe csound"
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>>> csound"
>>
>>
>> --
>> Michael Gogins
>> Irreducible Productions
>> http://www.michael-gogins.com
>> Michael dot Gogins at gmail dot com
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>>
>
> --
> View this message in context:
> http://old.nabble.com/-OT--Human-speech-is-music-to-out-ears-tp26629182p26770920.html
> Sent from the Csound - General mailing list archive at Nabble.com.
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"


-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I think you are confusing "algorithmability" with computability. There
is a well defined algotithm for calculating Pi, yet it is not computable
(because it is irrational).

Also, your analogy breaks down rather quickly because irrational numbers
are by definition infinite. Musical pieces are done in finite time, and
a finite amount of parameters are manipulated in a finite number of
ways. Therefore, a musical piece is a finite set of information. There
are a few infinite precision parameters (like duration of a particular
note/sound). However, perception is not continuous but discrete, and
therefore even those parameters are perceived as discrete -> describable
in a finite number of bits.

On Sun, 2009-12-13 at 15:29 -0800, Aaron Krister Johnson wrote:
> Hi Michael Gogins and others,
> 
> Algorithmic music and more importantly, non-algorithmic music--this is an
> area of enormous interest to me, both philosophically and practically.
> 
> Human consciousness is largely a non-algorithmic process, otherwise, we'd
> have a hard time noticing patterns which are 'stuck' or 'cliches'....most
> music produced in history has been non-algorithmic, and most certainly, the
> non-algorithmic universe is infinitely vaster (more interesting, too?) than
> the algorithmic universe.  (I mean this in the literal, mathematical sense
> of the transfinite numbers of Georg Cantor (see below) as well as
> poetically/figuratively)
> 
> Think about it from a purely mathematical point-of-view: there are an
> infinite number of computable irrational numbers (sqrt of 2 for instance).
> Imagine that a given irrational number codes in a one-to-one way a given
> score or soundfile for a piece of music. The (strict) algorithmic or
> process-oriented composer's focus is on these numbers whose properties are
> potentially deterministic: i.e., the piece can be thought of as the outcome
> of a more-or-less 'fixed' process.
> 
> Now think about the numbers which correspond to 'tweaked' computable
> irrationals. For instance, the square root of 2 with fifty of its digits
> changed, perhaps the first 50 digits in prime numbered 'slots' in its
> decimal representation. Many composers of algorithmic music do such
> things---they 'tweak the output'....
> 
> Now think about the greater infinity--in the sense of Cantor's number
> theory, where there are 'transfinite sets'---the set of irrational numbers
> which are non-computable in a *direct* sense by any algorithm. IOW, they
> require a conscious being to creatively search them out. There are
> infinitely more of these numbers, literally. IOW, this infinity is GREATER
> than the infinity of algorithmic, computable numbers. For those without
> background, I encourage you to read about Georg Cantor, and as it relates to
> programs and computing, a book called "The Computational Beauty of Nature"
> 
> Of course, many of these numbers can be thought of as random strings,
> indeed, most of them are. However, there is an important subset of them that
> are NOT random, IOW, they contain a clear structure, but a structure which
> is provably non-algorithmically generated in the easy way that random
> numbers can be. My conjecture is that such numbers would/could be
> code-strings for a piece of music which has clear structure, but also has
> deep cohesiveness on multiple levels, and also a rich set of surprising and
> enjoyable details. Clearly, there is a sense of motives, repetition, and
> self-reference at play, just like one has in most traditionally-written
> music that gives a sense of 'organic cohesion'. Such enjoyable and
> surprising stray details in a traditional masterwork might be possibly
> simulated by randomness, but most appear to be directed by an irreducible
> sense that a conscious being's choices are at work. David Cope's work has
> come close to this, but mostly because he does an awful lot of massaging of
> data which was pre-coded, like formal structures and Markov chain
> statistics. None of these things would by any stretch likely arise by pure
> chance. The more interesting question is "what forces intrisic or extrinsic
> to the composer produce such structures" Surely cultural history and
> training play a part, as well as pure chance....but they combine in such a
> way that I think may forever belie analysis. It might be too complex a thing
> for the human mind to actually know itself, just as a mouse will never
> fathom what a prime number is.
> 
> So, anyway, imagine an irrational number that codes Stravinsky's "Rite of
> Spring" or Beethoven's "Eroica" symphony....certainly these pieces are not
> 'computable' in any meaningful way...they go beyond syntax and into
> semantics---meaning, they require a conscious being to create them. No
> algorithm would simply spit out the "Eroica"...unless the program were
> simply to print out the precoded representation of the "Eroica", which is
> not computing it in the sense I mean, but merely printing stored data.
> 
> A simple example suffices to show the limits of a strictly, or even largely,
> algorithmic approach to composition: the choices an algorithm makes have to
> be set by parameters made available to the algorithm beforehand, by the
> author: for instance--no algorithm will start spitting out triplets because
> it is feeling 'bored' or 'playful'---it will spit out triplets because the
> algorithm's author decided to include them as a parameter choice---and
> algorithmic music will always be limited to the parameter choices which are
> built-in to the algorithm. Further, the algorithm wouldn't ever consider
> such a choice to be a 'meaninful gesture' in any semantic sense, whereas in
> traditional music one sees meaning in these gestures in an extra-musical
> sense *all the time*---think about the music for cartoons or films...all the
> humorous gagging going on in Carl Stalling's Looney Tunes scores defy any
> attempt at an algorithmic description.
> 
> No sense of 'escaping the system' in a *conscious* way can ever come out of
> such a system unless it's hard-coded in, which, is neither a 'true' escape,
> and, at least right now, requires a conscious being like a human to fulfill.
> Think about all the "fractal music" you've heard. How much of it ever has
> anything remotely structured like a scale passage or a simple arpeggiated
> triad? The larger the scale passage, the less likely it will appear by
> chance. And so much great music uses such extremely non-random materials as
> long scale passages and broken chords, to give two examples.
> 
> So, I think it will always come back to a human's aesthetic judgment and the
> fact that we are conscious pilots who want to fly through the terrain of
> musical possibilities.
> 
> All this being said, I am personally interested in the potential of
> algorithmic music, and some of it's results. In fact, I've written some
> myself, and it can be interesting and satisfying to do so. Mostly, when the
> effort is collaborative with a highly critical composer who *listens* to the
> results and is not a-priori 'detached' from the results. Meaning they can
> jump in and change and edit the output of a program to fit their taste.
> Otherwise, I mostly find purely algorithmic music dreadfully cold and boring
> to hear. However, I do find algorithms a convenient way to spur creativity,
> and to escape one's own predilections by introducing constraints and random
> choices one wouldn't have made. This leads to I think an expanded palette of
> creative motifs that I think of as a dialogical process.
> 
> All of this, of course, echoes the really pretentious extremes of scientific
> reductionism. I'd love to be proven wrong, and I think it'd be damn great to
> have "equation music" on a par with say Bach or Beethoven or the Beatles,
> but I remain very skeptical.
> 
> What are your thoughts?
> 
> Best,
> AKJ
> 
> 
> Michael Gogins-2 wrote:
> > 
> > I agree with you that if I cannot physically hear the idea or its
> > unique result, there is no music. (I think that's what you are saying,
> > anyway).
> > 
> > But I don't think I made my point about ideas and the ear very clear.
> > What I meant is that by working purely with ideas and algorithms,
> > without any "pre-hearing" if you will, it is possible to make music
> > that sounds good (and that may be difficult if not impossible to make
> > any other way).
> > 
> > I think John Cage did that, Xenakis and others did that, and some of
> > us would indeed be comparatively helpless musically if we could not do
> > that.
> > 
> > Regards,
> > Mike
> > 
> > On 12/10/09, Peiman Khosravi  wrote:
> >> Yes I see this is certainly the case. However I would like to mention
> >> that there are semi-algorithmic approaches (which by the way I
> >> wouldn't categorize as a musical "style") that do heavily rely on the
> >> ear (Murail and OpenMusic, Xenakis, Grisey). Algorithmic in the sense
> >> of using an automated algorithm to generate certain material: either
> >> the algorithm itself or the selective process that follows the
> >> generation can be guided heavily by the ear.
> >>
> >> Where music stops to be audible (audible in the sense of the mind's
> >> ear if you like) I draw the line. I think that's a residue of an
> >> unfortunate fixation on western notation.
> >>
> >> But then that's just me.
> >>
> >> Best
> >> P
> >>
> >> On 10 Dec 2009, at 13:45, Michael Gogins wrote:
> >>
> >>> In many styles of music the ears are indeed the most important tool --
> >>> ears understood metaphorically as including some kind of musical
> >>> perception and taste that surely occurs in the brain not the ears.
> >>>
> >>> However, in other styles the mind is the most important tool. I would
> >>> include process music and algorithmic composition. In these cases
> >>> without the ideas there is nothing for the ears. The ideas have a
> >>> distinctly musical component that is heard, if at all, only in the
> >>> mind. It may not even be heard at all, only "understood" in a musical
> >>> sense.
> >>>
> >>> Regards,
> >>> Mike
> >>>
> >>> On 12/10/09, Peiman Khosravi  wrote:
> >>>> No but your ears are the tools that you rely on more than anything
> >>>> else during the process of composition. So they are the most
> >>>> important
> >>>> part of your compositional tools. As long as you have them you
> >>>> needn't
> >>>> worry about software, or pencil sharpeners. Of course we all want
> >>>> better software environments that suite our particular creative
> >>>> patterns and preferences. But my point was that you are still guided
> >>>> by your ears when actually using the software. We sure don't need
> >>>> software to compose good music, whatever that is.
> >>>>
> >>>> I am thinking of Grisey who created the most spectrally rich works
> >>>> with no spectral analyzers or FFT to MIDI convertors (on that note
> >>>> you
> >>>> should take a look at this:
> >>>> http://recherche.ircam.fr/equipes/repmus/carpentier/orchidee.html)
> >>>> .
> >>>>
> >>>> P
> >>>>
> >>>> On 10 Dec 2009, at 13:13, Stéphane Rollandin wrote:
> >>>>
> >>>>>>> but we sorely lack tools for musical composition proper.
> >>>>>> We still have our ears.
> >>>>>
> >>>>> I don't compose with my ears, just as I don't paint with my eyes :)
> >>>>>
> >>>>> Stef
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>> Send bugs reports to this list.
> >>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
> >>>>> "unsubscribe csound"
> >>>>
> >>>>
> >>>>
> >>>> Send bugs reports to this list.
> >>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
> >>>> "unsubscribe
> >>>> csound"
> >>>
> >>>
> >>> --
> >>> Michael Gogins
> >>> Irreducible Productions
> >>> http://www.michael-gogins.com
> >>> Michael dot Gogins at gmail dot com
> >>>
> >>>
> >>> Send bugs reports to this list.
> >>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
> >>> "unsubscribe csound"
> >>
> >>
> >> Send bugs reports to this list.
> >> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> >> csound"
> > 
> > 
> > -- 
> > Michael Gogins
> > Irreducible Productions
> > http://www.michael-gogins.com
> > Michael dot Gogins at gmail dot com
> > 
> > 
> > Send bugs reports to this list.
> > To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> > csound"
> > 
> 
> -- 
> View this message in context: http://old.nabble.com/-OT--Human-speech-is-music-to-out-ears-tp26629182p26770920.html
> Sent from the Csound - General mailing list archive at Nabble.com.
> 
> 
> 
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"


-- 
Saludos,
Felipe Sateler

That's an interesting question. If I may add some thoughts.

There is certainly something to be said about the ease of performing  
certain tasks with certain tools. A good example is doing multichannel  
work in a DAW. I was recently trying to get Ardour to run some  
multichannel sessions (not ambiosonic however, at which Ardour  
apparently excels). I soon realized that with the inbuilt multichannel  
panner it is impossible to automate panning movements in Ardour. This  
does not make it impossible to do multichannel works with ardour, it  
simply means that one has to have a single track for each speaker and  
then do all the panning manually by crossfading between, say, 8  
tracks. You get the picture. One would be less likely to do many  
complex spatial patterns as a result or at least try to avoid them if  
possible.

The listening imagination in my experience has a direct link with the  
ease of performing certain tasks, a kind of feedback loop relationship  
with the software tools. Similarly one could do all filtering and  
EQing by just using sample delays and phase manipulation in a DAW, but  
how cumbersome would that be? I for one would be likely to minimize  
all filtering in such a case and try to do without it.

The main factor that makes it difficult to compare software with  
conventional instruments is that software is not by definition an  
instrument. Instruments take years to master, not to mention the  
centuries that went into developing both the instruments and the  
musical language. The piano is designed and developed to perform tonal  
note orientated music (tonal in the broader sense of the word); its  
performance technique and design has been perfected for the musical  
language and conditioned by it (centuries of it). The same can hardly  
be true of computer software, or at least not on such a global scale.

Best,

Peiman


On 10 Dec 2009, at 12:20, Victor Lazzarini wrote:

> That brings us to an interesting question: Need computer music  
> instruments (software or hardware) be 'easy to play'? This is  
> something I have always asked myself. Often we hear about how  
> something is either hard or easy, and whether in people's opinions  
> this makes it good or bad.
>
> In the case of traditional music instruments you don't seem to see  
> the same things. OK, players can complain some music is hard to  
> play, students complain that their instrument is difficult to  
> master, etc. But you don't see people going to redesign a violin to  
> make it easier to play; or attempts to do something like this seemed  
> to have taken away so much of the expressive possibilities that they  
> are disregarded as serious.
>
> This links to another question: should we not be regarding 'computer  
> music' as a field of study and development that is every bit as  
> complex and with long-term-development goals as standard musical  
> instruction?
>
> Victor
>
>
> On 10 Dec 2009, at 12:06, Stéphane Rollandin wrote:
>
>> same here. by ease, I meant practicability. some concepts can  
>> really be played with only when the software supports them  
>> comprehensively, else it gets very tedious. we (as composers and  
>> software developer) have to reify the lower structural aspects of  
>> our composition in order to use them effectively; and while these  
>> are lower in the view of the composition, they are very high-level  
>> in terms of software engineering.
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

> What are your thoughts?

Although all you say is meaningful and possibly deep, it has no 
relationship whatsoever with music composition as I see it. To me, 
algorithms are just tools, mecanical/logical rules made operational by 
being performed by a computer. Composing with algorithms is thus 
composing rules; it is never reduced to a plain mapping from a 
mathematical object to a musical piece. Such mappings are called 
sonification; that's hardly music. As for looking for some object which, 
once the correct mapping is applied, would yield the Rite of Spring, it 
reminds me of Borges' library of Babel which contains all possible 
books. Don't know if it makes much sense on the practical side. Plus, 
once you get the score for the Rite of Spring, you need to have an 
orchestra perform it. Or do you need to find another mathematical object 
which, once decoded, gives you the symphony waveform ?

As we say in French, "c'est de la masturbation intellectuelle"...

I'm more interested in music :)


Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

put "[OT]" (out of topic) in the header!!



----- Original Message ----
From: Aaron Krister Johnson 
To: csound@lists.bath.ac.uk
Sent: Mon, December 14, 2009 1:29:22 AM
Subject: [Csnd]  non-algorithmic music (was: Re: [OT] Human speech is music to out ears)


Hi Michael Gogins and others,

Algorithmic music and more importantly, non-algorithmic music--this is an
area of enormous interest to me, both philosophically and practically.

Human consciousness is largely a non-algorithmic process, otherwise, we'd
have a hard time noticing patterns which are 'stuck' or 'cliches'....most
music produced in history has been non-algorithmic, and most certainly, the
non-algorithmic universe is infinitely vaster (more interesting, too?) than
the algorithmic universe.  (I mean this in the literal, mathematical sense
of the transfinite numbers of Georg Cantor (see below) as well as
poetically/figuratively)

Think about it from a purely mathematical point-of-view: there are an
infinite number of computable irrational numbers (sqrt of 2 for instance).
Imagine that a given irrational number codes in a one-to-one way a given
score or soundfile for a piece of music. The (strict) algorithmic or
process-oriented composer's focus is on these numbers whose properties are
potentially deterministic: i.e., the piece can be thought of as the outcome
of a more-or-less 'fixed' process.

Now think about the numbers which correspond to 'tweaked' computable
irrationals. For instance, the square root of 2 with fifty of its digits
changed, perhaps the first 50 digits in prime numbered 'slots' in its
decimal representation. Many composers of algorithmic music do such
things---they 'tweak the output'....

Now think about the greater infinity--in the sense of Cantor's number
theory, where there are 'transfinite sets'---the set of irrational numbers
which are non-computable in a *direct* sense by any algorithm. IOW, they
require a conscious being to creatively search them out. There are
infinitely more of these numbers, literally. IOW, this infinity is GREATER
than the infinity of algorithmic, computable numbers. For those without
background, I encourage you to read about Georg Cantor, and as it relates to
programs and computing, a book called "The Computational Beauty of Nature"

Of course, many of these numbers can be thought of as random strings,
indeed, most of them are. However, there is an important subset of them that
are NOT random, IOW, they contain a clear structure, but a structure which
is provably non-algorithmically generated in the easy way that random
numbers can be. My conjecture is that such numbers would/could be
code-strings for a piece of music which has clear structure, but also has
deep cohesiveness on multiple levels, and also a rich set of surprising and
enjoyable details. Clearly, there is a sense of motives, repetition, and
self-reference at play, just like one has in most traditionally-written
music that gives a sense of 'organic cohesion'. Such enjoyable and
surprising stray details in a traditional masterwork might be possibly
simulated by randomness, but most appear to be directed by an irreducible
sense that a conscious being's choices are at work. David Cope's work has
come close to this, but mostly because he does an awful lot of massaging of
data which was pre-coded, like formal structures and Markov chain
statistics. None of these things would by any stretch likely arise by pure
chance. The more interesting question is "what forces intrisic or extrinsic
to the composer produce such structures" Surely cultural history and
training play a part, as well as pure chance....but they combine in such a
way that I think may forever belie analysis. It might be too complex a thing
for the human mind to actually know itself, just as a mouse will never
fathom what a prime number is.

So, anyway, imagine an irrational number that codes Stravinsky's "Rite of
Spring" or Beethoven's "Eroica" symphony....certainly these pieces are not
'computable' in any meaningful way...they go beyond syntax and into
semantics---meaning, they require a conscious being to create them. No
algorithm would simply spit out the "Eroica"...unless the program were
simply to print out the precoded representation of the "Eroica", which is
not computing it in the sense I mean, but merely printing stored data.

A simple example suffices to show the limits of a strictly, or even largely,
algorithmic approach to composition: the choices an algorithm makes have to
be set by parameters made available to the algorithm beforehand, by the
author: for instance--no algorithm will start spitting out triplets because
it is feeling 'bored' or 'playful'---it will spit out triplets because the
algorithm's author decided to include them as a parameter choice---and
algorithmic music will always be limited to the parameter choices which are
built-in to the algorithm. Further, the algorithm wouldn't ever consider
such a choice to be a 'meaninful gesture' in any semantic sense, whereas in
traditional music one sees meaning in these gestures in an extra-musical
sense *all the time*---think about the music for cartoons or films...all the
humorous gagging going on in Carl Stalling's Looney Tunes scores defy any
attempt at an algorithmic description.

No sense of 'escaping the system' in a *conscious* way can ever come out of
such a system unless it's hard-coded in, which, is neither a 'true' escape,
and, at least right now, requires a conscious being like a human to fulfill.
Think about all the "fractal music" you've heard. How much of it ever has
anything remotely structured like a scale passage or a simple arpeggiated
triad? The larger the scale passage, the less likely it will appear by
chance. And so much great music uses such extremely non-random materials as
long scale passages and broken chords, to give two examples.

So, I think it will always come back to a human's aesthetic judgment and the
fact that we are conscious pilots who want to fly through the terrain of
musical possibilities.

All this being said, I am personally interested in the potential of
algorithmic music, and some of it's results. In fact, I've written some
myself, and it can be interesting and satisfying to do so. Mostly, when the
effort is collaborative with a highly critical composer who *listens* to the
results and is not a-priori 'detached' from the results. Meaning they can
jump in and change and edit the output of a program to fit their taste.
Otherwise, I mostly find purely algorithmic music dreadfully cold and boring
to hear. However, I do find algorithms a convenient way to spur creativity,
and to escape one's own predilections by introducing constraints and random
choices one wouldn't have made. This leads to I think an expanded palette of
creative motifs that I think of as a dialogical process.

All of this, of course, echoes the really pretentious extremes of scientific
reductionism. I'd love to be proven wrong, and I think it'd be damn great to
have "equation music" on a par with say Bach or Beethoven or the Beatles,
but I remain very skeptical.

What are your thoughts?

Best,
AKJ


Michael Gogins-2 wrote:
> 
> I agree with you that if I cannot physically hear the idea or its
> unique result, there is no music. (I think that's what you are saying,
> anyway).
> 
> But I don't think I made my point about ideas and the ear very clear.
> What I meant is that by working purely with ideas and algorithms,
> without any "pre-hearing" if you will, it is possible to make music
> that sounds good (and that may be difficult if not impossible to make
> any other way).
> 
> I think John Cage did that, Xenakis and others did that, and some of
> us would indeed be comparatively helpless musically if we could not do
> that.
> 
> Regards,
> Mike
> 
> On 12/10/09, Peiman Khosravi  wrote:
>> Yes I see this is certainly the case. However I would like to mention
>> that there are semi-algorithmic approaches (which by the way I
>> wouldn't categorize as a musical "style") that do heavily rely on the
>> ear (Murail and OpenMusic, Xenakis, Grisey). Algorithmic in the sense
>> of using an automated algorithm to generate certain material: either
>> the algorithm itself or the selective process that follows the
>> generation can be guided heavily by the ear.
>>
>> Where music stops to be audible (audible in the sense of the mind's
>> ear if you like) I draw the line. I think that's a residue of an
>> unfortunate fixation on western notation.
>>
>> But then that's just me.
>>
>> Best
>> P
>>
>> On 10 Dec 2009, at 13:45, Michael Gogins wrote:
>>
>>> In many styles of music the ears are indeed the most important tool --
>>> ears understood metaphorically as including some kind of musical
>>> perception and taste that surely occurs in the brain not the ears.
>>>
>>> However, in other styles the mind is the most important tool. I would
>>> include process music and algorithmic composition. In these cases
>>> without the ideas there is nothing for the ears. The ideas have a
>>> distinctly musical component that is heard, if at all, only in the
>>> mind. It may not even be heard at all, only "understood" in a musical
>>> sense.
>>>
>>> Regards,
>>> Mike
>>>
>>> On 12/10/09, Peiman Khosravi  wrote:
>>>> No but your ears are the tools that you rely on more than anything
>>>> else during the process of composition. So they are the most
>>>> important
>>>> part of your compositional tools. As long as you have them you
>>>> needn't
>>>> worry about software, or pencil sharpeners. Of course we all want
>>>> better software environments that suite our particular creative
>>>> patterns and preferences. But my point was that you are still guided
>>>> by your ears when actually using the software. We sure don't need
>>>> software to compose good music, whatever that is.
>>>>
>>>> I am thinking of Grisey who created the most spectrally rich works
>>>> with no spectral analyzers or FFT to MIDI convertors (on that note
>>>> you
>>>> should take a look at this:
>>>> http://recherche.ircam.fr/equipes/repmus/carpentier/orchidee.html)
>>>> .
>>>>
>>>> P
>>>>
>>>> On 10 Dec 2009, at 13:13, Stéphane Rollandin wrote:
>>>>
>>>>>>> but we sorely lack tools for musical composition proper.
>>>>>> We still have our ears.
>>>>>
>>>>> I don't compose with my ears, just as I don't paint with my eyes :)
>>>>>
>>>>> Stef
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Send bugs reports to this list.
>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>>>> "unsubscribe csound"
>>>>
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>>> "unsubscribe
>>>> csound"
>>>
>>>
>>> --
>>> Michael Gogins
>>> Irreducible Productions
>>> http://www.michael-gogins.com
>>> Michael dot Gogins at gmail dot com
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>> "unsubscribe csound"
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
> 
> 
> -- 
> Michael Gogins
> Irreducible Productions
> http://www.michael-gogins.com
> Michael dot Gogins at gmail dot com
> 
> 
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
> 

-- 
View this message in context: http://old.nabble.com/-OT--Human-speech-is-music-to-out-ears-tp26629182p26770920.html
Sent from the Csound - General mailing list archive at Nabble.com.



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"





Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Hi Felipe,


Felipe Sateler wrote:
> 
> I think you are confusing "algorithmability" with computability. There
> is a well defined algotithm for calculating Pi, yet it is not computable
> (because it is irrational).
> 
> Aaron:
> Actually, this is your misunderstanding. PI is both algorithmic and
> computable. It is computable to any precision one should desire. One just
> has to wait an infinitely long time for an infinitely accurate
> representation. So there's 'practical computability' and 'computability in
> principal'. Yes, infinitely long numbers are not 'practically
> computable'--there we agree. See the author Gary Flake (The computational
> beauty of Nature) for the discussion on recursively enumerable sets and
> their ilk for more.
> 
> Felipe:
> Also, your analogy breaks down rather quickly because irrational numbers
> are by definition infinite. Musical pieces are done in finite time, and
> a finite amount of parameters are manipulated in a finite number of
> ways. Therefore, a musical piece is a finite set of information. There
> are a few infinite precision parameters (like duration of a particular
> note/sound).
> 
> Aaron:
> Let's put this in practical terms. A piece of music has no bounds on its
> length. Secondly, consider for a moment all the pieces of one-minute
> length. The number of such pieces is certainly astronomically large,
> considering all the factors such as timbre, pitch, articulation, etc. Of
> course, we'd have cases analogous to "different interpretations of the
> same pieces" but these would be the vast minority of any such pieces.
> My point was that the vast majority of such pieces would not be
> algorithmic in nature.
> 
> Felipe: 
> However, perception is not continuous but discrete, and
> therefore even those parameters are perceived as discrete -> describable
> in a finite number of bits.
> 
> Aaron:
> Says who? This seems to me to be a bias of a certain digital age
> worldview, that has a utility for us, but is not the fabric of reality,
> per se. There are always infinitely more details that could be mined from
> any phenomenon, which would exhaust for instance a 64-bit representation.
> And, just because a given n-bit representation of a phenomenon like an
> audio recording may "appear to be practically the same" to a given human
> subject, in no way implies that it is.
> 
> On Sun, 2009-12-13 at 15:29 -0800, Aaron Krister Johnson wrote:
>> Hi Michael Gogins and others,
>> 
>> Algorithmic music and more importantly, non-algorithmic music--this is an
>> area of enormous interest to me, both philosophically and practically.
>> 
>> Human consciousness is largely a non-algorithmic process, otherwise, we'd
>> have a hard time noticing patterns which are 'stuck' or 'cliches'....most
>> music produced in history has been non-algorithmic, and most certainly,
>> the
>> non-algorithmic universe is infinitely vaster (more interesting, too?)
>> than
>> the algorithmic universe.  (I mean this in the literal, mathematical
>> sense
>> of the transfinite numbers of Georg Cantor (see below) as well as
>> poetically/figuratively)
>> 
>> Think about it from a purely mathematical point-of-view: there are an
>> infinite number of computable irrational numbers (sqrt of 2 for
>> instance).
>> Imagine that a given irrational number codes in a one-to-one way a given
>> score or soundfile for a piece of music. The (strict) algorithmic or
>> process-oriented composer's focus is on these numbers whose properties
>> are
>> potentially deterministic: i.e., the piece can be thought of as the
>> outcome
>> of a more-or-less 'fixed' process.
>> 
>> Now think about the numbers which correspond to 'tweaked' computable
>> irrationals. For instance, the square root of 2 with fifty of its digits
>> changed, perhaps the first 50 digits in prime numbered 'slots' in its
>> decimal representation. Many composers of algorithmic music do such
>> things---they 'tweak the output'....
>> 
>> Now think about the greater infinity--in the sense of Cantor's number
>> theory, where there are 'transfinite sets'---the set of irrational
>> numbers
>> which are non-computable in a *direct* sense by any algorithm. IOW, they
>> require a conscious being to creatively search them out. There are
>> infinitely more of these numbers, literally. IOW, this infinity is
>> GREATER
>> than the infinity of algorithmic, computable numbers. For those without
>> background, I encourage you to read about Georg Cantor, and as it relates
>> to
>> programs and computing, a book called "The Computational Beauty of
>> Nature"
>> 
>> Of course, many of these numbers can be thought of as random strings,
>> indeed, most of them are. However, there is an important subset of them
>> that
>> are NOT random, IOW, they contain a clear structure, but a structure
>> which
>> is provably non-algorithmically generated in the easy way that random
>> numbers can be. My conjecture is that such numbers would/could be
>> code-strings for a piece of music which has clear structure, but also has
>> deep cohesiveness on multiple levels, and also a rich set of surprising
>> and
>> enjoyable details. Clearly, there is a sense of motives, repetition, and
>> self-reference at play, just like one has in most traditionally-written
>> music that gives a sense of 'organic cohesion'. Such enjoyable and
>> surprising stray details in a traditional masterwork might be possibly
>> simulated by randomness, but most appear to be directed by an irreducible
>> sense that a conscious being's choices are at work. David Cope's work has
>> come close to this, but mostly because he does an awful lot of massaging
>> of
>> data which was pre-coded, like formal structures and Markov chain
>> statistics. None of these things would by any stretch likely arise by
>> pure
>> chance. The more interesting question is "what forces intrisic or
>> extrinsic
>> to the composer produce such structures" Surely cultural history and
>> training play a part, as well as pure chance....but they combine in such
>> a
>> way that I think may forever belie analysis. It might be too complex a
>> thing
>> for the human mind to actually know itself, just as a mouse will never
>> fathom what a prime number is.
>> 
>> So, anyway, imagine an irrational number that codes Stravinsky's "Rite of
>> Spring" or Beethoven's "Eroica" symphony....certainly these pieces are
>> not
>> 'computable' in any meaningful way...they go beyond syntax and into
>> semantics---meaning, they require a conscious being to create them. No
>> algorithm would simply spit out the "Eroica"...unless the program were
>> simply to print out the precoded representation of the "Eroica", which is
>> not computing it in the sense I mean, but merely printing stored data.
>> 
>> A simple example suffices to show the limits of a strictly, or even
>> largely,
>> algorithmic approach to composition: the choices an algorithm makes have
>> to
>> be set by parameters made available to the algorithm beforehand, by the
>> author: for instance--no algorithm will start spitting out triplets
>> because
>> it is feeling 'bored' or 'playful'---it will spit out triplets because
>> the
>> algorithm's author decided to include them as a parameter choice---and
>> algorithmic music will always be limited to the parameter choices which
>> are
>> built-in to the algorithm. Further, the algorithm wouldn't ever consider
>> such a choice to be a 'meaninful gesture' in any semantic sense, whereas
>> in
>> traditional music one sees meaning in these gestures in an extra-musical
>> sense *all the time*---think about the music for cartoons or films...all
>> the
>> humorous gagging going on in Carl Stalling's Looney Tunes scores defy any
>> attempt at an algorithmic description.
>> 
>> No sense of 'escaping the system' in a *conscious* way can ever come out
>> of
>> such a system unless it's hard-coded in, which, is neither a 'true'
>> escape,
>> and, at least right now, requires a conscious being like a human to
>> fulfill.
>> Think about all the "fractal music" you've heard. How much of it ever has
>> anything remotely structured like a scale passage or a simple arpeggiated
>> triad? The larger the scale passage, the less likely it will appear by
>> chance. And so much great music uses such extremely non-random materials
>> as
>> long scale passages and broken chords, to give two examples.
>> 
>> So, I think it will always come back to a human's aesthetic judgment and
>> the
>> fact that we are conscious pilots who want to fly through the terrain of
>> musical possibilities.
>> 
>> All this being said, I am personally interested in the potential of
>> algorithmic music, and some of it's results. In fact, I've written some
>> myself, and it can be interesting and satisfying to do so. Mostly, when
>> the
>> effort is collaborative with a highly critical composer who *listens* to
>> the
>> results and is not a-priori 'detached' from the results. Meaning they can
>> jump in and change and edit the output of a program to fit their taste.
>> Otherwise, I mostly find purely algorithmic music dreadfully cold and
>> boring
>> to hear. However, I do find algorithms a convenient way to spur
>> creativity,
>> and to escape one's own predilections by introducing constraints and
>> random
>> choices one wouldn't have made. This leads to I think an expanded palette
>> of
>> creative motifs that I think of as a dialogical process.
>> 
>> All of this, of course, echoes the really pretentious extremes of
>> scientific
>> reductionism. I'd love to be proven wrong, and I think it'd be damn great
>> to
>> have "equation music" on a par with say Bach or Beethoven or the Beatles,
>> but I remain very skeptical.
>> 
>> What are your thoughts?
>> 
>> Best,
>> AKJ
>> 
>> 
>> Michael Gogins-2 wrote:
>> > 
>> > I agree with you that if I cannot physically hear the idea or its
>> > unique result, there is no music. (I think that's what you are saying,
>> > anyway).
>> > 
>> > But I don't think I made my point about ideas and the ear very clear.
>> > What I meant is that by working purely with ideas and algorithms,
>> > without any "pre-hearing" if you will, it is possible to make music
>> > that sounds good (and that may be difficult if not impossible to make
>> > any other way).
>> > 
>> > I think John Cage did that, Xenakis and others did that, and some of
>> > us would indeed be comparatively helpless musically if we could not do
>> > that.
>> > 
>> > Regards,
>> > Mike
>> > 
>> > On 12/10/09, Peiman Khosravi  wrote:
>> >> Yes I see this is certainly the case. However I would like to mention
>> >> that there are semi-algorithmic approaches (which by the way I
>> >> wouldn't categorize as a musical "style") that do heavily rely on the
>> >> ear (Murail and OpenMusic, Xenakis, Grisey). Algorithmic in the sense
>> >> of using an automated algorithm to generate certain material: either
>> >> the algorithm itself or the selective process that follows the
>> >> generation can be guided heavily by the ear.
>> >>
>> >> Where music stops to be audible (audible in the sense of the mind's
>> >> ear if you like) I draw the line. I think that's a residue of an
>> >> unfortunate fixation on western notation.
>> >>
>> >> But then that's just me.
>> >>
>> >> Best
>> >> P
>> >>
>> >> On 10 Dec 2009, at 13:45, Michael Gogins wrote:
>> >>
>> >>> In many styles of music the ears are indeed the most important tool
>> --
>> >>> ears understood metaphorically as including some kind of musical
>> >>> perception and taste that surely occurs in the brain not the ears.
>> >>>
>> >>> However, in other styles the mind is the most important tool. I would
>> >>> include process music and algorithmic composition. In these cases
>> >>> without the ideas there is nothing for the ears. The ideas have a
>> >>> distinctly musical component that is heard, if at all, only in the
>> >>> mind. It may not even be heard at all, only "understood" in a musical
>> >>> sense.
>> >>>
>> >>> Regards,
>> >>> Mike
>> >>>
>> >>> On 12/10/09, Peiman Khosravi  wrote:
>> >>>> No but your ears are the tools that you rely on more than anything
>> >>>> else during the process of composition. So they are the most
>> >>>> important
>> >>>> part of your compositional tools. As long as you have them you
>> >>>> needn't
>> >>>> worry about software, or pencil sharpeners. Of course we all want
>> >>>> better software environments that suite our particular creative
>> >>>> patterns and preferences. But my point was that you are still guided
>> >>>> by your ears when actually using the software. We sure don't need
>> >>>> software to compose good music, whatever that is.
>> >>>>
>> >>>> I am thinking of Grisey who created the most spectrally rich works
>> >>>> with no spectral analyzers or FFT to MIDI convertors (on that note
>> >>>> you
>> >>>> should take a look at this:
>> >>>> http://recherche.ircam.fr/equipes/repmus/carpentier/orchidee.html)
>> >>>> .
>> >>>>
>> >>>> P
>> >>>>
>> >>>> On 10 Dec 2009, at 13:13, Stéphane Rollandin wrote:
>> >>>>
>> >>>>>>> but we sorely lack tools for musical composition proper.
>> >>>>>> We still have our ears.
>> >>>>>
>> >>>>> I don't compose with my ears, just as I don't paint with my eyes :)
>> >>>>>
>> >>>>> Stef
>> >>>>>
>> >>>>>
>> >>>>>
>> >>>>>
>> >>>>> Send bugs reports to this list.
>> >>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>> >>>>> "unsubscribe csound"
>> >>>>
>> >>>>
>> >>>>
>> >>>> Send bugs reports to this list.
>> >>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>> >>>> "unsubscribe
>> >>>> csound"
>> >>>
>> >>>
>> >>> --
>> >>> Michael Gogins
>> >>> Irreducible Productions
>> >>> http://www.michael-gogins.com
>> >>> Michael dot Gogins at gmail dot com
>> >>>
>> >>>
>> >>> Send bugs reports to this list.
>> >>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>> >>> "unsubscribe csound"
>> >>
>> >>
>> >> Send bugs reports to this list.
>> >> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>> "unsubscribe
>> >> csound"
>> > 
>> > 
>> > -- 
>> > Michael Gogins
>> > Irreducible Productions
>> > http://www.michael-gogins.com
>> > Michael dot Gogins at gmail dot com
>> > 
>> > 
>> > Send bugs reports to this list.
>> > To unsubscribe, send email sympa@lists.bath.ac.uk with body
>> "unsubscribe
>> > csound"
>> > 
>> 
>> -- 
>> View this message in context:
>> http://old.nabble.com/-OT--Human-speech-is-music-to-out-ears-tp26629182p26770920.html
>> Sent from the Csound - General mailing list archive at Nabble.com.
>> 
>> 
>> 
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
> 
> 
> -- 
> Saludos,
> Felipe Sateler
> 
>  
> 

-- 
View this message in context: http://old.nabble.com/-OT--Human-speech-is-music-to-out-ears-tp26629182p26779753.html
Sent from the Csound - General mailing list archive at Nabble.com.



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Stéphane Rollandin wrote:
> 
>> What are your thoughts?
> 
> Although all you say is meaningful and possibly deep, it has no 
> relationship whatsoever with music composition as I see it. To me, 
> algorithms are just tools, mecanical/logical rules made operational by 
> being performed by a computer. Composing with algorithms is thus 
> composing rules; it is never reduced to a plain mapping from a 
> mathematical object to a musical piece.

I agree that it shouldn't be called music, but that's my point. Many people
do it, and call it such.


Stéphane Rollandin wrote:
> 
> Such mappings are called 
> sonification; that's hardly music.

Yup. I think, as a composer, it's *enormously* interesting and relevent to
think about exactly *why* that is the case.


Stéphane Rollandin wrote:
>  for looking for some object which, 
> once the correct mapping is applied, would yield the Rite of Spring, it 
> reminds me of Borges' library of Babel which contains all possible 
> books. Don't know if it makes much sense on the practical side. Plus, 
> once you get the score for the Rite of Spring, you need to have an 
> orchestra perform it. Or do you need to find another mathematical object 
> which, once decoded, gives you the symphony waveform ?

I was thinking more of the score, in this case. Either way, it's moot. But
the simpler example would just be, say, an ascii representation of the
score.


Stéphane Rollandin wrote:
> 
> As we say in French, "c'est de la masturbation intellectuelle"...

I've always enjoyed intellectual things, and masturbation. Sure, not as good
as real sex, but certainly useful to relieve tension, and it can be
satisfying. I'm sorry the French are so repressed and so low-brow
simultaneously to come up with such a derogatory phrase. :) Besides---c'mon,
Jacques Derrida? About as 'masturbatory' as it gets, not to mention utterly
opaque.....


Stéphane Rollandin wrote:
> 
> I'm more interested in music :) 

Yes, me too.

Would you define the algorithmic thing you posted earlier as 'music'? 

AKJ

Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
csound"

-- 
View this message in context: http://old.nabble.com/-OT--Human-speech-is-music-to-out-ears-tp26629182p26779898.html
Sent from the Csound - General mailing list archive at Nabble.com.



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

>> Such mappings are called 
>> sonification; that's hardly music.
> 
> Yup. I think, as a composer, it's *enormously* interesting and relevent to
> think about exactly *why* that is the case.

Well, in my view, to be musically interesting a mapping must be quite 
complex. The art of composing is then shifted from the manipulation of 
some elementary musical representations (as it is traditionnally: notes 
chords, motifs, etc.) to the manipulation of the mapping itself. At this 
point we arrive to my conception of algorithmic composition: composition 
of rules/procedures. Composing the mapping is composing the music; only 
at another level.


>> As we say in French, "c'est de la masturbation intellectuelle"...
> 
> I've always enjoyed intellectual things, and masturbation. Sure, not as good
> as real sex, but certainly useful to relieve tension, and it can be
> satisfying.

Agreed, although I think the meaning of masturbatory here is that the 
intellectual process is not fruitful.


> I'm sorry the French are so repressed and so low-brow
> simultaneously to come up with such a derogatory phrase. :) Besides---c'mon,
> Jacques Derrida? About as 'masturbatory' as it gets, not to mention utterly
> opaque.....

Oh yes. Why do you think we have such a saying ? It's very much a French 
thing... so-called "intellectuals" are a plague in this country (ok, 
some of them are good)


> Would you define the algorithmic thing you posted earlier as 'music'? 

Yes. I would not call it "my music", though. It's an experiment in 
musicality, so to say. Hopefully one day I will produce music that 
really express something "felt inside", but at the moment this is just a 
far away, long-term, goal. I'm only studying.


Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

In my own discussion of this issue, I carefully distinguished between
the finitude and computability of any given piece of music, and the
infinitude and uncomputability of music as an art, which must be
presumed to have no determinate end.

To be perfectly precise, if we accept current cosmology the universe
as we experience it has a beginning and end in time and a finite
horizon in space, therefore is a finite entity. But, from within, we
cannot determine the possibilities of music within this horizon in
advance. Music like any other form of thought is computationally
irreducible.

This is a critical issue, which I will stress and elucidate. A
phenomenon may be computable in the abstract and yet uncomputable by
any observer. This is a technically rigorous statement of our
situation.

As usual in this kind of more or less formalized epistemological
situation there is a hierarchy:

Music without reference to horizons, music in all universes, the
proper class of music if you will, music from God's point of view --
formally uncomputable because of uncountably infinite complexity.

Music with reference to our horizon, music as all pieces that might be
realized within the horizon of our cosmos -- formally computable
because finite, but uncomputable by any observer in our cosmos due to
lack of time and resources, to being part of the whole.

The important point is that although the art of music within the
horizons of our cosmos is uncomputable by any observer, each work of
music is nevertheless created by an observer.

Two consequences: our attitude in composing is implicitly that of God
or the gods, since we cannot compose as though the art of music is
computable by us since we know it is not; and the question whether we
are in fact so composing, or only act and feel as though we are but in
reality draw only upon chance, is itself formally undecidable.

Regards,
Mike




This is why I named my studio "Irreducible Productions", by the way.

On 12/14/09, Aaron Krister Johnson  wrote:
>
> Hi Felipe,
>
>
> Felipe Sateler wrote:
>>
>> I think you are confusing "algorithmability" with computability. There
>> is a well defined algotithm for calculating Pi, yet it is not computable
>> (because it is irrational).
>>
>> Aaron:
>> Actually, this is your misunderstanding. PI is both algorithmic and
>> computable. It is computable to any precision one should desire. One just
>> has to wait an infinitely long time for an infinitely accurate
>> representation. So there's 'practical computability' and 'computability in
>> principal'. Yes, infinitely long numbers are not 'practically
>> computable'--there we agree. See the author Gary Flake (The computational
>> beauty of Nature) for the discussion on recursively enumerable sets and
>> their ilk for more.
>>
>> Felipe:
>> Also, your analogy breaks down rather quickly because irrational numbers
>> are by definition infinite. Musical pieces are done in finite time, and
>> a finite amount of parameters are manipulated in a finite number of
>> ways. Therefore, a musical piece is a finite set of information. There
>> are a few infinite precision parameters (like duration of a particular
>> note/sound).
>>
>> Aaron:
>> Let's put this in practical terms. A piece of music has no bounds on its
>> length. Secondly, consider for a moment all the pieces of one-minute
>> length. The number of such pieces is certainly astronomically large,
>> considering all the factors such as timbre, pitch, articulation, etc. Of
>> course, we'd have cases analogous to "different interpretations of the
>> same pieces" but these would be the vast minority of any such pieces.
>> My point was that the vast majority of such pieces would not be
>> algorithmic in nature.
>>
>> Felipe:
>> However, perception is not continuous but discrete, and
>> therefore even those parameters are perceived as discrete -> describable
>> in a finite number of bits.
>>
>> Aaron:
>> Says who? This seems to me to be a bias of a certain digital age
>> worldview, that has a utility for us, but is not the fabric of reality,
>> per se. There are always infinitely more details that could be mined from
>> any phenomenon, which would exhaust for instance a 64-bit representation.
>> And, just because a given n-bit representation of a phenomenon like an
>> audio recording may "appear to be practically the same" to a given human
>> subject, in no way implies that it is.
>>
>> On Sun, 2009-12-13 at 15:29 -0800, Aaron Krister Johnson wrote:
>>> Hi Michael Gogins and others,
>>>
>>> Algorithmic music and more importantly, non-algorithmic music--this is an
>>> area of enormous interest to me, both philosophically and practically.
>>>
>>> Human consciousness is largely a non-algorithmic process, otherwise, we'd
>>> have a hard time noticing patterns which are 'stuck' or 'cliches'....most
>>> music produced in history has been non-algorithmic, and most certainly,
>>> the
>>> non-algorithmic universe is infinitely vaster (more interesting, too?)
>>> than
>>> the algorithmic universe.  (I mean this in the literal, mathematical
>>> sense
>>> of the transfinite numbers of Georg Cantor (see below) as well as
>>> poetically/figuratively)
>>>
>>> Think about it from a purely mathematical point-of-view: there are an
>>> infinite number of computable irrational numbers (sqrt of 2 for
>>> instance).
>>> Imagine that a given irrational number codes in a one-to-one way a given
>>> score or soundfile for a piece of music. The (strict) algorithmic or
>>> process-oriented composer's focus is on these numbers whose properties
>>> are
>>> potentially deterministic: i.e., the piece can be thought of as the
>>> outcome
>>> of a more-or-less 'fixed' process.
>>>
>>> Now think about the numbers which correspond to 'tweaked' computable
>>> irrationals. For instance, the square root of 2 with fifty of its digits
>>> changed, perhaps the first 50 digits in prime numbered 'slots' in its
>>> decimal representation. Many composers of algorithmic music do such
>>> things---they 'tweak the output'....
>>>
>>> Now think about the greater infinity--in the sense of Cantor's number
>>> theory, where there are 'transfinite sets'---the set of irrational
>>> numbers
>>> which are non-computable in a *direct* sense by any algorithm. IOW, they
>>> require a conscious being to creatively search them out. There are
>>> infinitely more of these numbers, literally. IOW, this infinity is
>>> GREATER
>>> than the infinity of algorithmic, computable numbers. For those without
>>> background, I encourage you to read about Georg Cantor, and as it relates
>>> to
>>> programs and computing, a book called "The Computational Beauty of
>>> Nature"
>>>
>>> Of course, many of these numbers can be thought of as random strings,
>>> indeed, most of them are. However, there is an important subset of them
>>> that
>>> are NOT random, IOW, they contain a clear structure, but a structure
>>> which
>>> is provably non-algorithmically generated in the easy way that random
>>> numbers can be. My conjecture is that such numbers would/could be
>>> code-strings for a piece of music which has clear structure, but also has
>>> deep cohesiveness on multiple levels, and also a rich set of surprising
>>> and
>>> enjoyable details. Clearly, there is a sense of motives, repetition, and
>>> self-reference at play, just like one has in most traditionally-written
>>> music that gives a sense of 'organic cohesion'. Such enjoyable and
>>> surprising stray details in a traditional masterwork might be possibly
>>> simulated by randomness, but most appear to be directed by an irreducible
>>> sense that a conscious being's choices are at work. David Cope's work has
>>> come close to this, but mostly because he does an awful lot of massaging
>>> of
>>> data which was pre-coded, like formal structures and Markov chain
>>> statistics. None of these things would by any stretch likely arise by
>>> pure
>>> chance. The more interesting question is "what forces intrisic or
>>> extrinsic
>>> to the composer produce such structures" Surely cultural history and
>>> training play a part, as well as pure chance....but they combine in such
>>> a
>>> way that I think may forever belie analysis. It might be too complex a
>>> thing
>>> for the human mind to actually know itself, just as a mouse will never
>>> fathom what a prime number is.
>>>
>>> So, anyway, imagine an irrational number that codes Stravinsky's "Rite of
>>> Spring" or Beethoven's "Eroica" symphony....certainly these pieces are
>>> not
>>> 'computable' in any meaningful way...they go beyond syntax and into
>>> semantics---meaning, they require a conscious being to create them. No
>>> algorithm would simply spit out the "Eroica"...unless the program were
>>> simply to print out the precoded representation of the "Eroica", which is
>>> not computing it in the sense I mean, but merely printing stored data.
>>>
>>> A simple example suffices to show the limits of a strictly, or even
>>> largely,
>>> algorithmic approach to composition: the choices an algorithm makes have
>>> to
>>> be set by parameters made available to the algorithm beforehand, by the
>>> author: for instance--no algorithm will start spitting out triplets
>>> because
>>> it is feeling 'bored' or 'playful'---it will spit out triplets because
>>> the
>>> algorithm's author decided to include them as a parameter choice---and
>>> algorithmic music will always be limited to the parameter choices which
>>> are
>>> built-in to the algorithm. Further, the algorithm wouldn't ever consider
>>> such a choice to be a 'meaninful gesture' in any semantic sense, whereas
>>> in
>>> traditional music one sees meaning in these gestures in an extra-musical
>>> sense *all the time*---think about the music for cartoons or films...all
>>> the
>>> humorous gagging going on in Carl Stalling's Looney Tunes scores defy any
>>> attempt at an algorithmic description.
>>>
>>> No sense of 'escaping the system' in a *conscious* way can ever come out
>>> of
>>> such a system unless it's hard-coded in, which, is neither a 'true'
>>> escape,
>>> and, at least right now, requires a conscious being like a human to
>>> fulfill.
>>> Think about all the "fractal music" you've heard. How much of it ever has
>>> anything remotely structured like a scale passage or a simple arpeggiated
>>> triad? The larger the scale passage, the less likely it will appear by
>>> chance. And so much great music uses such extremely non-random materials
>>> as
>>> long scale passages and broken chords, to give two examples.
>>>
>>> So, I think it will always come back to a human's aesthetic judgment and
>>> the
>>> fact that we are conscious pilots who want to fly through the terrain of
>>> musical possibilities.
>>>
>>> All this being said, I am personally interested in the potential of
>>> algorithmic music, and some of it's results. In fact, I've written some
>>> myself, and it can be interesting and satisfying to do so. Mostly, when
>>> the
>>> effort is collaborative with a highly critical composer who *listens* to
>>> the
>>> results and is not a-priori 'detached' from the results. Meaning they can
>>> jump in and change and edit the output of a program to fit their taste.
>>> Otherwise, I mostly find purely algorithmic music dreadfully cold and
>>> boring
>>> to hear. However, I do find algorithms a convenient way to spur
>>> creativity,
>>> and to escape one's own predilections by introducing constraints and
>>> random
>>> choices one wouldn't have made. This leads to I think an expanded palette
>>> of
>>> creative motifs that I think of as a dialogical process.
>>>
>>> All of this, of course, echoes the really pretentious extremes of
>>> scientific
>>> reductionism. I'd love to be proven wrong, and I think it'd be damn great
>>> to
>>> have "equation music" on a par with say Bach or Beethoven or the Beatles,
>>> but I remain very skeptical.
>>>
>>> What are your thoughts?
>>>
>>> Best,
>>> AKJ
>>>
>>>
>>> Michael Gogins-2 wrote:
>>> >
>>> > I agree with you that if I cannot physically hear the idea or its
>>> > unique result, there is no music. (I think that's what you are saying,
>>> > anyway).
>>> >
>>> > But I don't think I made my point about ideas and the ear very clear.
>>> > What I meant is that by working purely with ideas and algorithms,
>>> > without any "pre-hearing" if you will, it is possible to make music
>>> > that sounds good (and that may be difficult if not impossible to make
>>> > any other way).
>>> >
>>> > I think John Cage did that, Xenakis and others did that, and some of
>>> > us would indeed be comparatively helpless musically if we could not do
>>> > that.
>>> >
>>> > Regards,
>>> > Mike
>>> >
>>> > On 12/10/09, Peiman Khosravi  wrote:
>>> >> Yes I see this is certainly the case. However I would like to mention
>>> >> that there are semi-algorithmic approaches (which by the way I
>>> >> wouldn't categorize as a musical "style") that do heavily rely on the
>>> >> ear (Murail and OpenMusic, Xenakis, Grisey). Algorithmic in the sense
>>> >> of using an automated algorithm to generate certain material: either
>>> >> the algorithm itself or the selective process that follows the
>>> >> generation can be guided heavily by the ear.
>>> >>
>>> >> Where music stops to be audible (audible in the sense of the mind's
>>> >> ear if you like) I draw the line. I think that's a residue of an
>>> >> unfortunate fixation on western notation.
>>> >>
>>> >> But then that's just me.
>>> >>
>>> >> Best
>>> >> P
>>> >>
>>> >> On 10 Dec 2009, at 13:45, Michael Gogins wrote:
>>> >>
>>> >>> In many styles of music the ears are indeed the most important tool
>>> --
>>> >>> ears understood metaphorically as including some kind of musical
>>> >>> perception and taste that surely occurs in the brain not the ears.
>>> >>>
>>> >>> However, in other styles the mind is the most important tool. I would
>>> >>> include process music and algorithmic composition. In these cases
>>> >>> without the ideas there is nothing for the ears. The ideas have a
>>> >>> distinctly musical component that is heard, if at all, only in the
>>> >>> mind. It may not even be heard at all, only "understood" in a musical
>>> >>> sense.
>>> >>>
>>> >>> Regards,
>>> >>> Mike
>>> >>>
>>> >>> On 12/10/09, Peiman Khosravi  wrote:
>>> >>>> No but your ears are the tools that you rely on more than anything
>>> >>>> else during the process of composition. So they are the most
>>> >>>> important
>>> >>>> part of your compositional tools. As long as you have them you
>>> >>>> needn't
>>> >>>> worry about software, or pencil sharpeners. Of course we all want
>>> >>>> better software environments that suite our particular creative
>>> >>>> patterns and preferences. But my point was that you are still guided
>>> >>>> by your ears when actually using the software. We sure don't need
>>> >>>> software to compose good music, whatever that is.
>>> >>>>
>>> >>>> I am thinking of Grisey who created the most spectrally rich works
>>> >>>> with no spectral analyzers or FFT to MIDI convertors (on that note
>>> >>>> you
>>> >>>> should take a look at this:
>>> >>>> http://recherche.ircam.fr/equipes/repmus/carpentier/orchidee.html)
>>> >>>> .
>>> >>>>
>>> >>>> P
>>> >>>>
>>> >>>> On 10 Dec 2009, at 13:13, Stéphane Rollandin wrote:
>>> >>>>
>>> >>>>>>> but we sorely lack tools for musical composition proper.
>>> >>>>>> We still have our ears.
>>> >>>>>
>>> >>>>> I don't compose with my ears, just as I don't paint with my eyes :)
>>> >>>>>
>>> >>>>> Stef
>>> >>>>>
>>> >>>>>
>>> >>>>>
>>> >>>>>
>>> >>>>> Send bugs reports to this list.
>>> >>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>> >>>>> "unsubscribe csound"
>>> >>>>
>>> >>>>
>>> >>>>
>>> >>>> Send bugs reports to this list.
>>> >>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>> >>>> "unsubscribe
>>> >>>> csound"
>>> >>>
>>> >>>
>>> >>> --
>>> >>> Michael Gogins
>>> >>> Irreducible Productions
>>> >>> http://www.michael-gogins.com
>>> >>> Michael dot Gogins at gmail dot com
>>> >>>
>>> >>>
>>> >>> Send bugs reports to this list.
>>> >>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>> >>> "unsubscribe csound"
>>> >>
>>> >>
>>> >> Send bugs reports to this list.
>>> >> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>> "unsubscribe
>>> >> csound"
>>> >
>>> >
>>> > --
>>> > Michael Gogins
>>> > Irreducible Productions
>>> > http://www.michael-gogins.com
>>> > Michael dot Gogins at gmail dot com
>>> >
>>> >
>>> > Send bugs reports to this list.
>>> > To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>> "unsubscribe
>>> > csound"
>>> >
>>>
>>> --
>>> View this message in context:
>>> http://old.nabble.com/-OT--Human-speech-is-music-to-out-ears-tp26629182p26770920.html
>>> Sent from the Csound - General mailing list archive at Nabble.com.
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>>> csound"
>>
>>
>> --
>> Saludos,
>> Felipe Sateler
>>
>>
>>
>
> --
> View this message in context:
> http://old.nabble.com/-OT--Human-speech-is-music-to-out-ears-tp26629182p26779753.html
> Sent from the Csound - General mailing list archive at Nabble.com.
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"


-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On Mon, 2009-12-14 at 07:38 -0800, Aaron Krister Johnson wrote:
> Hi Felipe,
> 
> 
> Felipe Sateler wrote:
> > 
> > I think you are confusing "algorithmability" with computability. There
> > is a well defined algotithm for calculating Pi, yet it is not computable
> > (because it is irrational).
> > 
> > Aaron:
> > Actually, this is your misunderstanding. PI is both algorithmic and
> > computable. It is computable to any precision one should desire. One just
> > has to wait an infinitely long time for an infinitely accurate
> > representation. So there's 'practical computability' and 'computability in
> > principal'. Yes, infinitely long numbers are not 'practically
> > computable'--there we agree. See the author Gary Flake (The computational
> > beauty of Nature) for the discussion on recursively enumerable sets and
> > their ilk for more.

Then I don't understand what do you mean by non-algorithmic. Algorithmic
means there is an algorithm to do something, but you say irrational
numbers are not algorithmic? Pi, e, sqrt(n) all have well defined
algorithms, which means irrational numbers are not non-algorithmic in
nature.
Or do you mean that the irrational number set is not countable?

> > 
> > Felipe:
> > Also, your analogy breaks down rather quickly because irrational numbers
> > are by definition infinite. Musical pieces are done in finite time, and
> > a finite amount of parameters are manipulated in a finite number of
> > ways. Therefore, a musical piece is a finite set of information. There
> > are a few infinite precision parameters (like duration of a particular
> > note/sound).
> > 
> > Aaron:
> > Let's put this in practical terms. A piece of music has no bounds on its
> > length. Secondly, consider for a moment all the pieces of one-minute
> > length. The number of such pieces is certainly astronomically large,
> > considering all the factors such as timbre, pitch, articulation, etc. Of
> > course, we'd have cases analogous to "different interpretations of the
> > same pieces" but these would be the vast minority of any such pieces.
> > My point was that the vast majority of such pieces would not be
> > algorithmic in nature.

There is no theoretical bound on length, but there is a practical one.
No piece can be of infinite length.


> > 
> > Felipe: 
> > However, perception is not continuous but discrete, and
> > therefore even those parameters are perceived as discrete -> describable
> > in a finite number of bits.
> > 
> > Aaron:
> > Says who? This seems to me to be a bias of a certain digital age
> > worldview, that has a utility for us, but is not the fabric of reality,
> > per se. There are always infinitely more details that could be mined from
> > any phenomenon, which would exhaust for instance a 64-bit representation.
> > And, just because a given n-bit representation of a phenomenon like an
> > audio recording may "appear to be practically the same" to a given human
> > subject, in no way implies that it is.

There are 2 points here.

1. Is perception discrete? My statement does not come from a digital age
worldview[1] (although I can't claim I'm not biased, I got raised with
computers), but rather from certain scientific facts:
  - Energy is not continuous, but discrete (and a music piece is an
    energetic wave).
  - The brain does function binarily. Not like computers, of course, but
    a given neuron either fires or doesn't fire. There is no intensity
    in the firing of neurons.
  - "Classic" physics break down when applied at really small scales,
    which suggests to me that there is something really wrong with the
    assumptions on continuity of time and space.

2. Do you really care about "what it really is"? To me, a piece of music
matters because of the value I perceive in it. Given that I can't
perceive certain things, those things are therefore out of the
assessment of a given piece of music, and thus have no place in the set
of possible music pieces. All the possible variations on the parameters
I can't detect are reduced to a single node in the set.



[1] The opposite claim can be made too: your rejection of the idea can
be a bias from the continuum case worldview.


-- 
Saludos,
Felipe Sateler

Felipe Sateler  wrote:

> There is no theoretical bound on length, but there is a
> practical one. No piece can be of infinite length.

Why not?  What limit does the Csound API enforce?

> 1. Is perception discrete? My statement does not come
> from a digital age worldview[1] (although I can't claim
> I'm not biased, I got raised with computers), but rather
> from certain scientific facts:
>   - Energy is not continuous, but discrete (and a music
> piece is an energetic wave).



>   - "Classic" physics break down when applied at really
> small scales, which suggests to me that there is
> something really wrong with the assumptions on continuity
> of time and space.

These two facts are contradictory.  Yes, time and space are
probably not continuous, but working out the details is
quite difficult.  At least, quantum mechanics tells us that
energy is discrete in practical situations.  (You can write
music for free waves in infinite space if you want.)


                           Graham


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On Tue, 2009-12-15 at 08:44 +0800, Graham Breed wrote:
> Felipe Sateler  wrote:
> 
> > There is no theoretical bound on length, but there is a
> > practical one. No piece can be of infinite length.
> 
> Why not?  What limit does the Csound API enforce?

Not the csound API. But the people composing/performing, providing
electricity and so on, up to the end of human life.

> 
> > 1. Is perception discrete? My statement does not come
> > from a digital age worldview[1] (although I can't claim
> > I'm not biased, I got raised with computers), but rather
> > from certain scientific facts:
> >   - Energy is not continuous, but discrete (and a music
> > piece is an energetic wave).
> 
> 
> 
> >   - "Classic" physics break down when applied at really
> > small scales, which suggests to me that there is
> > something really wrong with the assumptions on continuity
> > of time and space.
> 
> These two facts are contradictory.  Yes, time and space are
> probably not continuous, but working out the details is
> quite difficult.  At least, quantum mechanics tells us that
> energy is discrete in practical situations.  (You can write
> music for free waves in infinite space if you want.)

I'm not quite sure what you mean here. Why are they contradictory? I
wrote "classic" precisely because I meant non-quantum theories.


-- 
Saludos,
Felipe Sateler

on 2009-12-14 at 07:38 Aaron Krister Johnson wrote:

>> Felipe: 
>> However, perception is not continuous but discrete, 

>> Aaron:
>> Says who? 

every book on psychoacoustics, where you can read about just noticeable
differences (JND), just noticeable variations (JNV), critical bands, etc




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Michael,

This argument is spot on. Exactly what I'm getting at......and more
precisely put, too.

I often wonder, too....is the length of a program that could produce
Beethoven's Eroica, for example, equivalent in length to "print [Beethoven
Eroica data]"....one of the offshoots of algorithmic complexity theory was
by a guy named Gregory Chaitin, who talked about the length of a program
that produced a certain result being a rough estimate of that algorithm's
complexity. Total predictable pattern and total randomness (is there such a
thing as total randomness?) are the extremes; in between is where the most
interesting music (from a consensus POV, mind you) seems to lie.

Another way of framing it: how compressible is a certain score---can it be
compactly described?

We find that there is among the classics a sense that there are something
akin to algorithms at work (think of canons, fugues, the recipe for a high
classical sonata-allegro movement, etc.), but that typical computer attempts
lack the fluidity of the human equivalent exercises...perhaps the problem
will be solved with truly conscious machines, if that indeed is an
achievable goal. I'm torn--I tend to think it's possible in principle, but I
have my doubts about it being possible in pracice, and it might just be
something irreducible at work to make a conscious being---i.e.,
consciousness might be a fundamental property of the universe in some way,
like matter and energy are....???

AKJ


Michael Gogins-2 wrote:
> 
> In my own discussion of this issue, I carefully distinguished between
> the finitude and computability of any given piece of music, and the
> infinitude and uncomputability of music as an art, which must be
> presumed to have no determinate end.
> 
> To be perfectly precise, if we accept current cosmology the universe
> as we experience it has a beginning and end in time and a finite
> horizon in space, therefore is a finite entity. But, from within, we
> cannot determine the possibilities of music within this horizon in
> advance. Music like any other form of thought is computationally
> irreducible.
> 
> This is a critical issue, which I will stress and elucidate. A
> phenomenon may be computable in the abstract and yet uncomputable by
> any observer. This is a technically rigorous statement of our
> situation.
> 
> As usual in this kind of more or less formalized epistemological
> situation there is a hierarchy:
> 
> Music without reference to horizons, music in all universes, the
> proper class of music if you will, music from God's point of view --
> formally uncomputable because of uncountably infinite complexity.
> 
> Music with reference to our horizon, music as all pieces that might be
> realized within the horizon of our cosmos -- formally computable
> because finite, but uncomputable by any observer in our cosmos due to
> lack of time and resources, to being part of the whole.
> 
> The important point is that although the art of music within the
> horizons of our cosmos is uncomputable by any observer, each work of
> music is nevertheless created by an observer.
> 
> Two consequences: our attitude in composing is implicitly that of God
> or the gods, since we cannot compose as though the art of music is
> computable by us since we know it is not; and the question whether we
> are in fact so composing, or only act and feel as though we are but in
> reality draw only upon chance, is itself formally undecidable.
> 
> Regards,
> Mike
> 
> 
> This is why I named my studio "Irreducible Productions", by the way. 

Yes, this is a good point....or rather, good points.

AKJ


luis jure wrote:
> 
> 
> on 2009-12-14 at 07:38 Aaron Krister Johnson wrote:
> 
>>> Felipe: 
>>> However, perception is not continuous but discrete, 
> 
>>> Aaron:
>>> Says who? 
> 
> every book on psychoacoustics, where you can read about just noticeable
> differences (JND), just noticeable variations (JNV), critical bands, etc
> 
> 
> 
> 
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
> 
> 

Felipe Sateler  wrote:
> On Tue, 2009-12-15 at 08:44 +0800, Graham Breed wrote:
> > Felipe Sateler  wrote:
> > 
> > > There is no theoretical bound on length, but there is
> > > a practical one. No piece can be of infinite length.
> > 
> > Why not?  What limit does the Csound API enforce?
> 
> Not the csound API. But the people composing/performing,
> providing electricity and so on, up to the end of human
> life.

Why do you care?  You can write an algorithm that produces
an infinitely long piece of music.  You can implement it
using Csound and your programming language of choice.  You
can fast-forward it to hear what it would sound like at some
point in the future. Whether or not anybody listens to the
whole thing, the length of that piece is unbounded.  It's
the same as infinite digits of pi.

As it happens, I think this is an interesting avenue to
explore for algorithmic music.  It's something you couldn't
do otherwise.  But then ideas are cheap -- I don't have an
implementation.

> > > 1. Is perception discrete? My statement does not come
> > > from a digital age worldview[1] (although I can't
> > > claim I'm not biased, I got raised with computers),
> > > but rather from certain scientific facts:
> > >   - Energy is not continuous, but discrete (and a
> > > music piece is an energetic wave).
> > 
> > 
> > 
> > >   - "Classic" physics break down when applied at
> > > really small scales, which suggests to me that there
> > > is something really wrong with the assumptions on
> > > continuity of time and space.
> > 
> > These two facts are contradictory.  Yes, time and space
> > are probably not continuous, but working out the
> > details is quite difficult.  At least, quantum
> > mechanics tells us that energy is discrete in practical
> > situations.  (You can write music for free waves in
> > infinite space if you want.)
> 
> I'm not quite sure what you mean here. Why are they
> contradictory? I wrote "classic" precisely because I
> meant non-quantum theories.

They're contradictory because first you give the classic
interpretation, and then you say it breaks down.  Yes,
non-quantum theories are continuous because it's precisely
quantum theory that brings in the quantization.  As it
happens, quantum theories are closer to reality.  Even if
you ignore the biology and any practical limitations,
perception must be quantized.


                                Graham


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Aaron Krister Johnson  wrote:

> Another way of framing it: how compressible is a certain
> score---can it be compactly described?

I think that's an interesting question.  The simple answer
is to take the score in a suitable digital form, and apply
a good compression algorithm to it.  The simple evasion is
to say that the score isn't the music anyway.

But let's talk about a score, and let's say that we should
understand what the compression algorithm's doing.  We want
to be able to write music with it, after all.  There are
all kinds of ways that a piece of music can be compressed
-- repetitions and near repetitions, notes being more
likely to occur together, some notes more common overall,
some scales dominating different sections.  A good
compression algorithm that works over a whole body of music
would be a codification of music theory.  You could match
its rules up to the standard rules.  The simplest rules
that compress the music most must be the important ones.

The result of this process would be the score of Eroica in
its tersest, human understandable form.  A set of rules
(some ad hoc, some shared with pieces across a whole
culture) and inspired choices made within those rules.  You
could then take it, change some of the rules, and make your
own inspired choices to write a new piece of music.

The advantage of working this way would be that you didn't
have to worry about all the boring details because the
computer could deal with them.  You'd tell the computer
what you wanted in the simplest terms and correct it when
it went wrong. You could also experiment new
harmonic rules (which is something we're both interested
in, at least).

This is how I want algorithmic or computer-aided music to
work.  From what I know of algorithmic music packages, it
isn't the way they do work.  Rather, they expect you to
pick an algorithm and see what comes out.  There are also
tools that handle some simple things like repetition and
transposition but don't go much further. If anybody can
point me in a more useful direction, please do so.


                                  Graham


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On 14/12/2009, at 10:51 AM, Michael Gogins wrote:
...
> Computability is a metaphysical question: Is an entity definable,
> knowable, finite, or not? – So, any question concerning the
> computability of music is a question of the metaphysics of music.
>
> My considerations are grounded in several obvious and unassailable
> facts, from which far-reaching inferences can be drawn.
>
Hello Michael, all.
It is good to have some positions enunciated; Thanks for your thoughts.

[These are very interesting issues and my comments/questions should be  
read not as me being argumentative for the sake of it, but in the  
spirit of discourse.]

As several have already said, this is a deep or involved 'problem' so  
I won't go on a length in this forum; there is, perhaps, better places  
for discussions about Algo. comp (AC) than this; those that afford a  
more in-depth approach. I should preface my remarks by saying I have  
thought of myself as a composer who uses algorithmic processes (AP)   
in making not most of my compositions, but more and more I think of  
music less and less as you describe; i.e. it as discrete audible  
objects so some of these "unassailable facts" as you describe them,  
are not (unassailable facts).
Because much of what you say is premised on this idea, I'll restrict  
myself to that issue; picking through the consequences, being  
laborious in this context.

> In the first place, works of music are physical phenomena – sounds.
> Each work of music is a sound of finite tessitura, dynamic range, and
> duration, which persists without gaps until it ends. Human hearing,
> too, has limited resolution. Therefore, each work of music can be
> considered to be completely represented by a digital recording of
> sufficient bandwidth: a certain finite series of 0s and 1s. Or again,
> a work of music is a sequence of finite complexity. So of course all
> such sequences are computable, though not necessarily in a reasonable
> period of time.

As I understand it, what you're saying,  our use of the phrase "work  
of music"  is distinct from simply "music"  i.e that a  "work of  
music" is not actually "music".  Are (recordings of performances of)  
Beethoven's Op xyz different "work of music"? Is a "musical  
composition" in this context the same thing as a "work of music"?
A "work of music" in the way you describe it is a very recent  
invention (See Lydia Goehr's "The imaginary museum of musical works".)  
There is a position in this debate that contends that "works of music"  
are not (just) sounds; so not physical phenomena.


  If I am not to deduce from your "unassailable fact" an assertion of  
imperialist intent, I have assume you are unaware of that position or  
you dismiss it. If the latter, it would be useful for ignorant people  
such as myself to be informed as to the nature of our delusion, for  
the more I read, think and listen musically, the deeper my delusion  
seems to become.  If the former, i.e. that you re that you are unaware  
of that position, now you're aware that there is another position, the  
"facts" have become not unassailable, which is not quite the same  
thing as assailable, and I'd be happy to enunciate the position further.


> But works of music have no meaning unless they are heard, and hearing
> music is an inward, subjective experience. It would be a mistake to
> suppose that the experience of music is not caused by the physical
> sound, yet it would equally be a mistake to suppose that the physical
> sound has any meaning without the experience of hearing it.

So to clarify, are you saying that "music" is an inner experience  
caused by the hearing of the physical sound of "works of music" and  
that it is this inner experience that creates meaning? That seems  
logically necessary within your definition, but with respect to an  
alternate position (that "works of music" are not (just sounds); so  
not physical phenomena) it is not sufficient.

> Forgetting
> the physical causation of music leads to the error of cultural
> relativism sometimes found in postmodernism, and forgetting the inward
> hearing of music leads to the error of objectifying music typical of
> scientism.

I understand your dialectic if you mean moral relativism. Perhaps you  
really do mean cultural relativism, in which case I'd appreciate an  
explanation of what this error is, unless, again, it is a superiority/  
imperialist one, in which case there remains little interest in the  
matter. Perhaps you'd like to explain nature of that casuation of  
relativism.
I agree with you wrt the "error of objectifying music". Although,  
again the dialectic you pose is not the only position, which seems to  
deal only with the relationship between a sonic object (a "work of  
music") and the individual (objectified, culturally unencumbered)  
individual hearer. So while one can argue that the cultural mileau of  
an individual is "carried" by that individual in the listening, I  
think that is the same kind of category error that lead Margaret  
Thatcher to proclaim that there is no such thing as society, only  
individuals and families.
Furthermore, even from the individualist POV,  in this model, the  
nature of acts of listening seem of vital importance and needs to be  
teased out, especially if one wants, as I do, to admit that there is  
some sense of shared and/or transmitted musical ideas.

> Now although any given work of music is a finite object, the art of
> music consists of a sequence of such works, each distinct from all
> others.
This may defendable as consistent with your definition, but not  
necessarily otherwise. I'm not sure where the "art" part of it comes  
into it, though. I'd like to hear an unpacking of that idea. Do you  
mean, for example "the art of music" as meaning the history of the  
making of "works of music"? Else
  what is the "art"?, an "art object"? If the latter, that is also  
defendable as consistent with your definition, however, the term "art  
of ..." usually refers to a practical skill, such as that applied by  
an artisan, for example (The art of motorcycle maintenance, of playing  
the violin, public speaking etc).

> Mathematically speaking, this series could continue
> indefinitely. The art of music can thus be considered to be an
> infinite series of 0s and 1s.
Is it a series or a sequence? This is not nit-picking, in terms of  
your definition of music, the distinction is very important.
>
> In other words, considered in the concrete, as a completed series,
> from the viewpoint of God if you will, the art of music is a series of
> countably infinite length and countably infinite complexity.
"A" completed series - what is the argument that such a description (a  
sequence, a series or a whatever) is either defensible (as in "the"  
theory of everything") or useful. Why not multiple sequences, series  
or a whatevers?
Also, there is no such "viewpoint of God", other than as a creation  
that may be necessary to defend the absolutist position. Because my  
purpose here was to question your very definition of music, I refrain  
from entering the debate about infinity and  countability.
> But
> considered in the abstract, as an uncompleted series, from the
> viewpoint of creatures if you will, the art of music is an infinite
> set of possible series, each of countably infinite length. This set of
> possible series can be diagonalized (exactly as in the halting
> theorem) to prove that it is of uncountable infinity and uncountable
> complexity. Since the series has not been completed and the art of
> music has not been finished, the reality of our situation is that the
> art of music is effectively uncomputable.
>
> The most important implication of this, for us, is that it would be
> absurd to hope for computers to generate the art of music as a whole.
> But as I have tried to show, they can be very useful indeed in
> computing works of music.
>
...
It seems, I might have missed it though, that there has not been an  
argument put for a different position, on subsequent posts i.e.  that  
"works of music are physical phenomena – sounds". Perhaps such a  
position should be put once Mike clarifies some of the points raised  
herewith.

D.





________________________________________________
Dr David Worrall.
- Experimental Polymedia:	  worrall.avatar.com.au
- Sonification: www.sonifiction.com.au
- Education for Financial Independence: www.mindthemarkets.com.au








Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

> You can write an algorithm that produces
> an infinitely long piece of music.  You can implement it
> using Csound and your programming language of choice.  You
> can fast-forward it to hear what it would sound like at some
> point in the future. Whether or not anybody listens to the
> whole thing, the length of that piece is unbounded.  It's
> the same as infinite digits of pi.
> 
> As it happens, I think this is an interesting avenue to
> explore for algorithmic music.  It's something you couldn't
> do otherwise.  But then ideas are cheap -- I don't have an
> implementation.


I wrote a piece of ambient music, called Rivière (river) which has no 
end. I just cut enough of it to fill a CD. You can get it here:
http://www.zogotounga.net/TGG/zik/riv/02-Riviere.ogg

Stef




Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Graham Breed wrote:
> Felipe Sateler  wrote:
>> On Tue, 2009-12-15 at 08:44 +0800, Graham Breed wrote:
>>> Felipe Sateler  wrote:
>>>
>>>> There is no theoretical bound on length, but there is
>>>> a practical one. No piece can be of infinite length.
>>> Why not?  What limit does the Csound API enforce?
>> Not the csound API. But the people composing/performing,
>> providing electricity and so on, up to the end of human
>> life.
> 
> Why do you care?  You can write an algorithm that produces
> an infinitely long piece of music.  You can implement it
> using Csound and your programming language of choice.  You
> can fast-forward it to hear what it would sound like at some
> point in the future. Whether or not anybody listens to the
> whole thing, the length of that piece is unbounded.  It's
> the same as infinite digits of pi.
> 

Any round, once started (think "Frere Jaques" etc) is infinite; you have 
to impose some termination on it for it to stop.  Which is OK. One 
problem I have with so much contemporary ea is that composers seem 
either unwilling or unable to compose endings. They take a piece so far, 
and then just stop. The first few times it may have some "meaning" by 
subverting expectation, or by rhetorically stomping on the music, but 
when everyine does it all the time, the rule of diminishing returns 
takes over. Some acoustic pieces have done that as well, and (unless the 
conductor is to blame) it is just as annoying for me. I like endings!

Perhaps the issue is that so few composers study classical counterpoint 
(fugue especially) these days that the idea of an algorithm that has no 
ending unless you deliberately compose it in (and which therefore 
becomes a major aspect of the composers' art), has got lost in the 
confusion. Anyone writing a fugue today would probably rely on a 
fade-out, ...


Richard Dobson






Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

> Anyone writing a fugue today would probably rely on a
> fade-out, ...
>
>
> Richard Dobson
>

I remember a symphony concert (Bournemouth SO, conductor Silvestri)
playingf a Variations and Fugue by Reger, when the conductor lais the
baton down and just watched the orchestra go round and round the fugue. 
reger got out of it bu drowning the sound in trumpets/trombones under the
cover of which the fugue disappeared.  So not just EA music!

And then there was the organist at Christchurch Priory who went round a
Bach fugue again becsause he missed the exit ramp...........

==John ff



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

luis jure wrote:
> on 2009-12-14 at 07:38 Aaron Krister Johnson wrote:
> 
>>> Felipe: 
>>> However, perception is not continuous but discrete, 
> 
>>> Aaron:
>>> Says who? 
> 
> every book on psychoacoustics, where you can read about just noticeable
> differences (JND), just noticeable variations (JNV), critical bands, etc
> 

I don't see those phenomena as indicating perception is discrete. We do 
have a tendency towards "categorical" perception, so that a sonic morph 
is too easily perceived as a plain crossfade - we recognise the first 
sound and hold on to that, until we suddenly 'flip' to the second one. 
This is actually useful most of the time. The phenomenon of discrete 
musical pitches establishes and exploits our instinct to focus on 
(contemplate) a fixed point in time/space. Given that, the occasional 
glide from one pitch to the next takes on considerable "meaning" (which 
is one reason I like playing the bansuri so much). But that 
discretization of pitch (say) is therefore symbolical rather than 
literal. We use it because it has meaning, not because it obeys any 
physical or mathematical law. Such laws are always after the fact, since 
for some strange reason we feel more secure if we can find intellectual 
justifications for what we do. We could call that the "grant application 
syndrome".

And we generally do not perceive very slow changes. The only time we are 
normally aware of the motion of the planets minute by minute and second 
by second is during a solar eclipse when it gets to the "diamond ring" 
stage. We imagine a singer is uttering discrete pitches, and only when 
we slow it down (or otherwise analyse it) realise they are really 
including a slide from one tone to the next.

With respect to quantum mechanics, I find incorporating the Planck 
length into  psychoacoustic analysis a little far-fetched. In practice 
we suffer/enjoy persistence (otherwise all films would flicker 
impossibly and all TVs would be a chaos of individually drawn lines); we 
experience the physical world through a fairly massive low-pass 
filter/envelope detector. We generally do not perceive very fast changes 
either, especially if changing. If we didn't have this "problem", FM 
synthesis etc would not work. And knowing that a tsunami is really just 
  gazillions of raindrops (and in particle physics terms comprising 
mostly empty space) does nothing to mitigate its destructive power, or 
enhance our ability to survive it.

Richard Dobson



Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

With respect to the computability, or not, of the art of music it is
important to be precise.

There are degrees of infinity (cardinality) and degrees of
computability (with respect to "oracles" that can decide undecidable
propositions of lower cardinality).

There is a further important distinction between "decidable" and
"recursively enumerable."  "Recursively enumerable" means that all
digits of an infinite sequence can be computed. Pi is recursively
enumerable. The word "computable" is unfortunately used colloquially
to denote both "decideable" and "recursively enumerable."

For theorists, "decidable" is the critical concept. It means you can
figure out whether your theory applies to the evidence because you
have a decision procedure for doing so.

This ambiguity plagues computational music theory (which is more or
less what we have been discussing here) because, I repeat, and this is
of truly fundamental importance, these are the metaphysical or
ultimate logical possibilities:

The "world" (all that is) is of unbounded cardinality. It is neither
decidable nor recursively denumerable. Our physical cosmos is but part
of the "the world." The art of music cannot be reduced to any
algorithm. In current cosmology, "the multiverse" or "chaotic
inflation" might be consistent with this.

The world is of aleph-one cardinality (the number of points on a
line). It is neither decidable nor recursively denumerable. Still the
art of music cannot be reduced to any algorithm. Current notions of
"the multiverse" or "chaotic inflation" may also be consistent with
this.

The world is of aleph-null cardinality (the number of integers). It is
not decidable but it IS recursively denumerable. The art of music can
be reduced to an algorithm, but not even God could decide whether or
not any particular candidate for this algorithm is the real thing.
Current notions of "the multiverse" or "chaotic inflation" are NOT
consistent with this. "Traditional" cosmology and physics, which more
or less ignores what likes beyond our cosmological horizon, MAY OR MAY
NOT be consistent with this, but probably could be, supposing e.g.
that quantum theory (which is finitary and discrete) is an emergent,
effective theory from a more fundamental continuum theory.

The world is of finite cardinality (this is the basic assumption of
the current scientific world-view, though I am convinced it is false).
It is BOTH decidable AND denumerable, but, for logical reasons, NOT
"from within." The art of music can be reduced to an algorithm, and we
can decide whether or not any particular candidate for this algorithm
is closer to, or father away, from the real thing, although (again for
logical reasons of the part not being able to completely represent the
whole) we wouldn't know for certain that we had the real algorithm
even if we did.

To go back and start over, given any of the first THREE metaphysics
above, an infinitely long piece of music cannot represent the art of
music as a whole. There is a countable number of notes, or sound
samples, or any other feature you like in an infinitely long piece of
music. But there is an uncountably infinite number of notes, or sound
samples, in _all possible such infinitely long pieces of music_. This
is because the number of integers is countably infinite, but the
number of subsets of integers is non-countably infinite. That is why I
keep harping on and on about the difference between "pieces of music"
and the "art of music." In a finite world, they have the same
cardinality, finite. But in any non-finite world, even though any
given piece of music has no more than aleph-null cardinality, the art
of music as a whole has aleph-one cardinality.

To make this notion a bit more concrete, assume that we have some tap
into an aleph-one "art of music." Then we open our tap a number of
times and sample "infinite histories of music," an infinite sequence
of possibly infinitely long pieces of music, from this universe. If we
pick our samples at random, every time we do, then with vanishing
improbability we get a _completely different history of music_ even
though each history is infinitely long.

Once again, this is critical to music theory or esthetics of
metaphysics or whatever because, in the case of a finite world, it is
at least logically consistent to suppose that the art of music is at
least recursively denumerable -- that some machine is spitting it out
even if we ourselves are too simple to deduce the internals of this
machine. In all cases of non-finite worlds, it is certain that the art
of music is not recursively denumerable, and so it is simply wrong to
think that it could be automated.

It is obvious that current physics is quite at sea with respect to
whether "the world" is finite, aleph-one, or aleph-null. It follows
that neither current physics, nor the current default set of
metaphysical assumptions, is any help here -- at all.

To sum up: Unless you have VERY GOOD EVIDENCE that "the world" is
finite, it makes no sense to think that music as an art form is
computable.

This, of course, is an utterly different question from whether it
makes sense to use algorithms to compose.

Regards,
Mike



On 12/15/09, Stéphane Rollandin  wrote:
>> You can write an algorithm that produces
>> an infinitely long piece of music.  You can implement it
>> using Csound and your programming language of choice.  You
>> can fast-forward it to hear what it would sound like at some
>> point in the future. Whether or not anybody listens to the
>> whole thing, the length of that piece is unbounded.  It's
>> the same as infinite digits of pi.
>>
>> As it happens, I think this is an interesting avenue to
>> explore for algorithmic music.  It's something you couldn't
>> do otherwise.  But then ideas are cheap -- I don't have an
>> implementation.
>
>
> I wrote a piece of ambient music, called Rivière (river) which has no
> end. I just cut enough of it to fill a CD. You can get it here:
> http://www.zogotounga.net/TGG/zik/riv/02-Riviere.ogg
>
> Stef
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>


-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I'm glad you seem to be pursuing this at the level of at least trying
to elucidate assumptions. Thanks for your responses.

> As I understand it, what you're saying,  our use of the phrase "work
> of music"  is distinct from simply "music"  i.e that a  "work of
> music" is not actually "music".  Are (recordings of performances of)
> Beethoven's Op xyz different "work of music"? Is a "musical
> composition" in this context the same thing as a "work of music"?
> A "work of music" in the way you describe it is a very recent
> invention (See Lydia Goehr's "The imaginary museum of musical works".)
> There is a position in this debate that contends that "works of music"
> are not (just) sounds; so not physical phenomena.

Not quite. A "work of music" is a possibly infinite sequence of
elements of music (notes, sound samples, what have you). "The art of
music" is the set of all possible "works of music." I do not begin to
attempt to decide, given a sequence of notes, sound samples, or what
have you, whether than object is in fact "music." I do not need to
know what "art" is.  I do not need to know this for the purposes of my
argument, hence I don't need the kind of critical theory that you seem
to be appealing to.

What is important to my argument is that you can count all the
elements of music in any work of music, but there are more elements of
music in the set of all works of music than you can count. The former
number is the number of integers, the latter number is the number of
points on a line.

You seem to be confusing my attempt to be objective with "an assertion
of imperialistic intent." Since I am basing my argument on mathematics
and philosophy of science, critical theory is irrelevant to it. A
desire to be object is not a desire to rule the world, it is a desire
to know what is.

>> But works of music have no meaning unless they are heard, and hearing
>> music is an inward, subjective experience. It would be a mistake to
>> suppose that the experience of music is not caused by the physical
>> sound, yet it would equally be a mistake to suppose that the physical
>> sound has any meaning without the experience of hearing it.
>
> So to clarify, are you saying that "music" is an inner experience
> caused by the hearing of the physical sound of "works of music" and
> that it is this inner experience that creates meaning? That seems
> logically necessary within your definition, but with respect to an
> alternate position (that "works of music" are not (just sounds); so
> not physical phenomena) it is not sufficient.

Again, I do not need to know much about the answer to this question to
prove my point.

Although I think the distinction between physical object and inner
experience is useful, I do not depend on much here. It suffices that
SOMETHING in the inner experience is caused by the physical object. I
don't need to know how much, or why.

The only assumption that is required to justify my conclusion is that,
given a different physical object, there will be a different inner
experience hence a different "work of music".

>
>> Forgetting
>> the physical causation of music leads to the error of cultural
>> relativism sometimes found in postmodernism, and forgetting the inward
>> hearing of music leads to the error of objectifying music typical of
>> scientism.
>
> I understand your dialectic if you mean moral relativism. Perhaps you
> really do mean cultural relativism, in which case I'd appreciate an
> explanation of what this error is, unless, again, it is a superiority/
> imperialist one, in which case there remains little interest in the
> matter. Perhaps you'd like to explain nature of that casuation of
> relativism.

I'm not sure I understand you. I meant both moral and cultural
relativism. To clarify my position, I think that some works of art are
better than others. "Better" means a lot of things here. I certainly
do not think my culture has a privileged position and does not need to
justify itself . Similarly I think other cultures need to justify
themselves as well. E.g. I think it is better not to practice slavery
than to practice it, regardless of one's culture. So abolitionists
from some cultures have had a beneficial impact across cultures.
Similarly I think that Western classical music has had a beneficial
impact across cultures. And so has African music or Indian music.

But again, this is irrelevant to the argument at hand.

> I agree with you wrt the "error of objectifying music". Although,
> again the dialectic you pose is not the only position, which seems to
> deal only with the relationship between a sonic object (a "work of
> music") and the individual (objectified, culturally unencumbered)
> individual hearer. So while one can argue that the cultural mileau of
> an individual is "carried" by that individual in the listening, I
> think that is the same kind of category error that lead Margaret
> Thatcher to proclaim that there is no such thing as society, only
> individuals and families.

Yes, but... I don't make the error of Thatcher. I do think society is
worthy of study as an object in itself. The individual is NOT
culturally unencumbered. But as individuals we are in a dialectical
relationship with society. We create it, and it creates us. We fight
it, and it fights us. This is just as true in the arts as it is in the
polity or the economy.

> Furthermore, even from the individualist POV,  in this model, the
> nature of acts of listening seem of vital importance and needs to be
> teased out, especially if one wants, as I do, to admit that there is
> some sense of shared and/or transmitted musical ideas.

I completely agree with this.

>> Now although any given work of music is a finite object, the art of
>> music consists of a sequence of such works, each distinct from all
>> others.
> This may defendable as consistent with your definition, but not
> necessarily otherwise. I'm not sure where the "art" part of it comes
> into it, though. I'd like to hear an unpacking of that idea. Do you
> mean, for example "the art of music" as meaning the history of the
> making of "works of music"? Else
>   what is the "art"?, an "art object"? If the latter, that is also
> defendable as consistent with your definition, however, the term "art
> of ..." usually refers to a practical skill, such as that applied by
> an artisan, for example (The art of motorcycle maintenance, of playing
> the violin, public speaking etc).

As I said I do not care much how "art" is defined here. It looks to me
like my use of this term has misled you. Think "possibilities of
music," not "art of music." For the purposes of my argument regarding
the computability of music, the definition of "art" is not very
important. The number of various quantities or possibilities is what
is critical.

>
>> Mathematically speaking, this series could continue
>> indefinitely. The art of music can thus be considered to be an
>> infinite series of 0s and 1s.
> Is it a series or a sequence? This is not nit-picking, in terms of
> your definition of music, the distinction is very important.
>>

The distinction is not important.

>> In other words, considered in the concrete, as a completed series,
>> from the viewpoint of God if you will, the art of music is a series of
>> countably infinite length and countably infinite complexity.
> "A" completed series - what is the argument that such a description (a
> sequence, a series or a whatever) is either defensible (as in "the"
> theory of everything") or useful. Why not multiple sequences, series
> or a whatevers?

To my argument it makes no difference, multiple or not. Similarly, in
my argument, sequence and series are just different words for the same
thing.

> Also, there is no such "viewpoint of God", other than as a creation
> that may be necessary to defend the absolutist position.

Certainly there is a viewpoint of God. But that is a separate
discussion. I use this idea here only as a metaphor, as a way of
presenting (not justifying) assumptions.

> Because my
> purpose here was to question your very definition of music, I refrain
> from entering the debate about infinity and  countability.

My argument is not a debate, it is an informal statement of a theorem.
The only debate involved is in justifying the assumptions. If you
accept the assumptions, then you accept the conclusion. And my
assumptions are few: (a) a musical object is a countable sequence. (b)
given a different "musical object:" there is a different "musical
experience." (c) The world is not finite. Frankly (c) is the really
critical assumption. From these points, plus accepted results in
computability theory, the rest follows quite logically. It is really
quite elementary.

>> But
>> considered in the abstract, as an uncompleted series, from the
>> viewpoint of creatures if you will, the art of music is an infinite
>> set of possible series, each of countably infinite length. This set of
>> possible series can be diagonalized (exactly as in the halting
>> theorem) to prove that it is of uncountable infinity and uncountable
>> complexity. Since the series has not been completed and the art of
>> music has not been finished, the reality of our situation is that the
>> art of music is effectively uncomputable.
>>
>> The most important implication of this, for us, is that it would be
>> absurd to hope for computers to generate the art of music as a whole.
>> But as I have tried to show, they can be very useful indeed in
>> computing works of music.
>>
> ...
> It seems, I might have missed it though, that there has not been an
> argument put for a different position, on subsequent posts i.e.  that
> "works of music are physical phenomena – sounds". Perhaps such a
> position should be put once Mike clarifies some of the points raised
> herewith.

A "work" of music might be a score, a computer disk, a memorized
tradition, or any of a wide number of means of storing information.
But if there isn't a sound that can be produced for somebody to listen
to that is caused by or represented by this work, there is no music.

Thanks again for your responses.

Mike

>
> D.
>
>
>
>
>
> ________________________________________________
> Dr David Worrall.
> - Experimental Polymedia:	  worrall.avatar.com.au
> - Sonification: www.sonifiction.com.au
> - Education for Financial Independence: www.mindthemarkets.com.au
>
>
>
>
>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"


-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Peiman Khosravi wrote:
> "Further, we need to correct the formulation that hearing occurs in 
> continuity. The ear in fact listens in brief slices, and what it 
> perceives and remembers /already/ consists in short sytheses of two or 
> three seconds of the sounds as it evolves."  -- Michel Chion, Audio-Vision
> 
> 

Hmm, well, I would say both sides of that argument are too simplistic. 
Both aural and visual cognition have to perform substantial degrees of 
information reduction to be manageable (something it appears those 
suffering from autistic-class syndromes find difficult to do);  which is 
part of the reason it is useful to learn stuff (rules, conventions, 
styles, etc so we have a framework onto whoch to map stimuli) )but the 
evidence that this amounts to a simple quantization is insubstantial at 
best.  But I would happily accept it as a description of ~his~ ear since 
he presumably knows very well how he hears! It is already known that 
"the ear" is very good at filling in gaps in a stimulus, to the extent 
that it may not even be aware that there are gaps. Lets perhaps accept 
that hearing (sound cognition) is both non-linear and multi-dimensional, 
and leave it at that?

In NLP terms, we may be classified as an "auditory processor' or a 
"visual processor" I know I am very definitely in the former camp; I can 
be presented with all manner of visual stimuli and I will fail to 
observe/remember vast chunks of it. Multi-media remains a problem for me 
as my strongest instinct is to close my eyes when listening. I imagine 
Chion can't really do that, most of the time, given what his specialism 
is. My experience is that the more we see, the less we hear. ~something~ 
has got to give!


Richard Dobson






Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Graham,

Yes, this description of an ideal software package---who *wouldn't* want
such a tool?

I have a sneaking intuition that if it were that easy, the music wouldn't be
all that great either....there's a concept known as informational
depth---the amount of time a process carries on is something akin to the
quality of its result. There's a reason most symphonies can take months or
years to write. And there's also a reason that a baby takes 9 months to grow
in the womb--and they are not even done when they pop out of the womb!

But, yes, wow---that would be cool....we are on the same page. I like your
amplification of the compressibility concept into the idea of "music theory
= compressibility of music" And it seems like such a software application
would be complex, yet possible. Until then, we are stuck with primitive
tools and our own intelligent ears. Maybe genetic algorithms and neural nets
that might 'learn' an authors tastes have the best future in this regard.

I've found that one thing we like in music is variety. Any one algorithm
would then seem to get old, because it would tend to produce a homogenous
result. One of the things that make music written by humans so nice and
interesting (most of the time) are the complete and sudden changes of say
tempo, texture, mood, etc. Of course, one could argue that that is more
symptomatic of *Western* music in particular, or at least common practice
Western music. Early Western music and Eastern music seem to be ok with a
much more homogenous texture, and I think current algorithmic music comes
much closer to 'modelling' those styles more successfully. Perhaps the
greatest success of algorithmic music has been to convincingly produce
neo-minimalistic music.

AKJ


Graham Breed wrote:
> 
> Aaron Krister Johnson  wrote:
> 
>> Another way of framing it: how compressible is a certain
>> score---can it be compactly described?
> 
> I think that's an interesting question.  The simple answer
> is to take the score in a suitable digital form, and apply
> a good compression algorithm to it.  The simple evasion is
> to say that the score isn't the music anyway.
> 
> But let's talk about a score, and let's say that we should
> understand what the compression algorithm's doing.  We want
> to be able to write music with it, after all.  There are
> all kinds of ways that a piece of music can be compressed
> -- repetitions and near repetitions, notes being more
> likely to occur together, some notes more common overall,
> some scales dominating different sections.  A good
> compression algorithm that works over a whole body of music
> would be a codification of music theory.  You could match
> its rules up to the standard rules.  The simplest rules
> that compress the music most must be the important ones.
> 
> The result of this process would be the score of Eroica in
> its tersest, human understandable form.  A set of rules
> (some ad hoc, some shared with pieces across a whole
> culture) and inspired choices made within those rules.  You
> could then take it, change some of the rules, and make your
> own inspired choices to write a new piece of music.
> 
> The advantage of working this way would be that you didn't
> have to worry about all the boring details because the
> computer could deal with them.  You'd tell the computer
> what you wanted in the simplest terms and correct it when
> it went wrong. You could also experiment new
> harmonic rules (which is something we're both interested
> in, at least).
> 
> This is how I want algorithmic or computer-aided music to
> work.  From what I know of algorithmic music packages, it
> isn't the way they do work.  Rather, they expect you to
> pick an algorithm and see what comes out.  There are also
> tools that handle some simple things like repetition and
> transposition but don't go much further. If anybody can
> point me in a more useful direction, please do so.
> 
> 
>                                   Graham
> 
> 
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
> 
> 

I'm pretty much on the same page with you here.

I don't think, however, there could be a single algorithm that would
boil down all music to one probably fairly complex algorithm. I think
that as music evolves, new styles arise that can't be reduced to the
algorithms of prior styles. A break in the algorithms pretty much
defines a change in the style. That's because I think that the world,
all that is, is infinite and that music can in principle evolve
without end.

But I definitely do use this basic idea in pursuing my own algorithmic
composition. I'm looking for algorithms that generate music in just
this sense. Then I compose music by composing the algorithms,
basically.

There's much more to this. Check out my personal Web page and links
thereon at http://www.michael-gogins.com. I think I have some
fragments of software that are heading in this direction, and I plan
to keep pushing it. I don't begin by reducing a corpus of data, I do
start with procedures, but they are designed to use basic operations
that are "musical" (voice-leadings, root progressions, repetitions,
etc.) and I then compose compose them hierarchically using formal
languages or mathematics. In principle these algorithms could be
treated as genomes that evolve and I plan to try that.

Regards,
Mike

On 12/15/09, Graham Breed  wrote:
> Aaron Krister Johnson  wrote:
>
>> Another way of framing it: how compressible is a certain
>> score---can it be compactly described?
>
> I think that's an interesting question.  The simple answer
> is to take the score in a suitable digital form, and apply
> a good compression algorithm to it.  The simple evasion is
> to say that the score isn't the music anyway.
>
> But let's talk about a score, and let's say that we should
> understand what the compression algorithm's doing.  We want
> to be able to write music with it, after all.  There are
> all kinds of ways that a piece of music can be compressed
> -- repetitions and near repetitions, notes being more
> likely to occur together, some notes more common overall,
> some scales dominating different sections.  A good
> compression algorithm that works over a whole body of music
> would be a codification of music theory.  You could match
> its rules up to the standard rules.  The simplest rules
> that compress the music most must be the important ones.
>
> The result of this process would be the score of Eroica in
> its tersest, human understandable form.  A set of rules
> (some ad hoc, some shared with pieces across a whole
> culture) and inspired choices made within those rules.  You
> could then take it, change some of the rules, and make your
> own inspired choices to write a new piece of music.
>
> The advantage of working this way would be that you didn't
> have to worry about all the boring details because the
> computer could deal with them.  You'd tell the computer
> what you wanted in the simplest terms and correct it when
> it went wrong. You could also experiment new
> harmonic rules (which is something we're both interested
> in, at least).
>
> This is how I want algorithmic or computer-aided music to
> work.  From what I know of algorithmic music packages, it
> isn't the way they do work.  Rather, they expect you to
> pick an algorithm and see what comes out.  There are also
> tools that handle some simple things like repetition and
> transposition but don't go much further. If anybody can
> point me in a more useful direction, please do so.
>
>
>                                   Graham
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>

On Tue, 2009-12-15 at 09:11 -0500, Michael Gogins wrote:
> To sum up: Unless you have VERY GOOD EVIDENCE that "the world" is
> finite, it makes no sense to think that music as an art form is
> computable.
> 
> This, of course, is an utterly different question from whether it
> makes sense to use algorithms to compose.

I'm pretty sure that we are all talking about something different.

Aaron's post suggested the need of a "creative" mind to search the set
of music pieces to find interesting nodes. I (in a somewhat unprecise
way, unfortunately) contested that by noting that the music pieces set
is finite. Therefore, there is no _need_ of a "creative" mind, it can be
simply searched by algorithms.


And I don't understand what do you mean with "music as an art form is
computable". I'm not even sure what do you mean by "music as an art
form"

-- 
Saludos,
Felipe Sateler

Disclaimer: I know very little about this type of algorithmic music,
but hasn't David Cope tried to do something similar with his research?
I"m too lazy to google but his name stands out in my head....

2009/12/15 Aaron Krister Johnson :
>
>
> Graham,
>
> Yes, this description of an ideal software package---who *wouldn't* want
> such a tool?
>
> I have a sneaking intuition that if it were that easy, the music wouldn't be
> all that great either....there's a concept known as informational
> depth---the amount of time a process carries on is something akin to the
> quality of its result. There's a reason most symphonies can take months or
> years to write. And there's also a reason that a baby takes 9 months to grow
> in the womb--and they are not even done when they pop out of the womb!
>
> But, yes, wow---that would be cool....we are on the same page. I like your
> amplification of the compressibility concept into the idea of "music theory
> = compressibility of music" And it seems like such a software application
> would be complex, yet possible. Until then, we are stuck with primitive
> tools and our own intelligent ears. Maybe genetic algorithms and neural nets
> that might 'learn' an authors tastes have the best future in this regard.
>
> I've found that one thing we like in music is variety. Any one algorithm
> would then seem to get old, because it would tend to produce a homogenous
> result. One of the things that make music written by humans so nice and
> interesting (most of the time) are the complete and sudden changes of say
> tempo, texture, mood, etc. Of course, one could argue that that is more
> symptomatic of *Western* music in particular, or at least common practice
> Western music. Early Western music and Eastern music seem to be ok with a
> much more homogenous texture, and I think current algorithmic music comes
> much closer to 'modelling' those styles more successfully. Perhaps the
> greatest success of algorithmic music has been to convincingly produce
> neo-minimalistic music.
>
> AKJ
>
>
> Graham Breed wrote:
>>
>> Aaron Krister Johnson  wrote:
>>
>>> Another way of framing it: how compressible is a certain
>>> score---can it be compactly described?
>>
>> I think that's an interesting question.  The simple answer
>> is to take the score in a suitable digital form, and apply
>> a good compression algorithm to it.  The simple evasion is
>> to say that the score isn't the music anyway.
>>
>> But let's talk about a score, and let's say that we should
>> understand what the compression algorithm's doing.  We want
>> to be able to write music with it, after all.  There are
>> all kinds of ways that a piece of music can be compressed
>> -- repetitions and near repetitions, notes being more
>> likely to occur together, some notes more common overall,
>> some scales dominating different sections.  A good
>> compression algorithm that works over a whole body of music
>> would be a codification of music theory.  You could match
>> its rules up to the standard rules.  The simplest rules
>> that compress the music most must be the important ones.
>>
>> The result of this process would be the score of Eroica in
>> its tersest, human understandable form.  A set of rules
>> (some ad hoc, some shared with pieces across a whole
>> culture) and inspired choices made within those rules.  You
>> could then take it, change some of the rules, and make your
>> own inspired choices to write a new piece of music.
>>
>> The advantage of working this way would be that you didn't
>> have to worry about all the boring details because the
>> computer could deal with them.  You'd tell the computer
>> what you wanted in the simplest terms and correct it when
>> it went wrong. You could also experiment new
>> harmonic rules (which is something we're both interested
>> in, at least).
>>
>> This is how I want algorithmic or computer-aided music to
>> work.  From what I know of algorithmic music packages, it
>> isn't the way they do work.  Rather, they expect you to
>> pick an algorithm and see what comes out.  There are also
>> tools that handle some simple things like repetition and
>> transposition but don't go much further. If anybody can
>> point me in a more useful direction, please do so.
>>
>>
>>                                   Graham
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>>
>>
>
> --
> View this message in context: http://old.nabble.com/-OT--Human-speech-is-music-to-out-ears-tp26629182p26796349.html
> Sent from the Csound - General mailing list archive at Nabble.com.
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Yes. Cope's work is based mainly on pattern matching from a corpus of
examples. Not the same thing as what I am doing, which is pattern
generation based on recursive rules that the user tweaks. In my work
the "compression" that was mentioned is achieved by recursion. Cope's
work seems to have some other "understanding" of musical hierarchical
structure.

Cope's results are very impressive! In terms of imitating or
reproducing examples of styles up to a section or so in length, they
go far beyond anyone else's work.

However, I do not have the computing platform (Macontish Common Lisp)
that is required to run his code, so I haven't been able to experiment
with his software as I would like to. I'm sure I'll look into Cope's
work more at some point. I have almost all Cope's books....

Somewhat as with IRCAM, I think it is damaging to the researcher not
to implement the research in one of the most widely accessible
languages and platforms. Basically, I think a computer music person
has to have a REALLY good reason not to use C, C++, Python, Scheme,
Common Lisp, Lua, or some other language that is very widely used AND
has good open source implementations.

If I had Cope's software running, I am sure I would try to use to
compose based on making up rules, instead of abstracting them from
examples.

Regards,
Mike

On 12/15/09, Rory Walsh  wrote:
> Disclaimer: I know very little about this type of algorithmic music,
> but hasn't David Cope tried to do something similar with his research?
> I"m too lazy to google but his name stands out in my head....
>
> 2009/12/15 Aaron Krister Johnson :
>>
>>
>> Graham,
>>
>> Yes, this description of an ideal software package---who *wouldn't* want
>> such a tool?
>>
>> I have a sneaking intuition that if it were that easy, the music wouldn't
>> be
>> all that great either....there's a concept known as informational
>> depth---the amount of time a process carries on is something akin to the
>> quality of its result. There's a reason most symphonies can take months or
>> years to write. And there's also a reason that a baby takes 9 months to
>> grow
>> in the womb--and they are not even done when they pop out of the womb!
>>
>> But, yes, wow---that would be cool....we are on the same page. I like your
>> amplification of the compressibility concept into the idea of "music
>> theory
>> = compressibility of music" And it seems like such a software application
>> would be complex, yet possible. Until then, we are stuck with primitive
>> tools and our own intelligent ears. Maybe genetic algorithms and neural
>> nets
>> that might 'learn' an authors tastes have the best future in this regard.
>>
>> I've found that one thing we like in music is variety. Any one algorithm
>> would then seem to get old, because it would tend to produce a homogenous
>> result. One of the things that make music written by humans so nice and
>> interesting (most of the time) are the complete and sudden changes of say
>> tempo, texture, mood, etc. Of course, one could argue that that is more
>> symptomatic of *Western* music in particular, or at least common practice
>> Western music. Early Western music and Eastern music seem to be ok with a
>> much more homogenous texture, and I think current algorithmic music comes
>> much closer to 'modelling' those styles more successfully. Perhaps the
>> greatest success of algorithmic music has been to convincingly produce
>> neo-minimalistic music.
>>
>> AKJ
>>
>>
>> Graham Breed wrote:
>>>
>>> Aaron Krister Johnson  wrote:
>>>
>>>> Another way of framing it: how compressible is a certain
>>>> score---can it be compactly described?
>>>
>>> I think that's an interesting question.  The simple answer
>>> is to take the score in a suitable digital form, and apply
>>> a good compression algorithm to it.  The simple evasion is
>>> to say that the score isn't the music anyway.
>>>
>>> But let's talk about a score, and let's say that we should
>>> understand what the compression algorithm's doing.  We want
>>> to be able to write music with it, after all.  There are
>>> all kinds of ways that a piece of music can be compressed
>>> -- repetitions and near repetitions, notes being more
>>> likely to occur together, some notes more common overall,
>>> some scales dominating different sections.  A good
>>> compression algorithm that works over a whole body of music
>>> would be a codification of music theory.  You could match
>>> its rules up to the standard rules.  The simplest rules
>>> that compress the music most must be the important ones.
>>>
>>> The result of this process would be the score of Eroica in
>>> its tersest, human understandable form.  A set of rules
>>> (some ad hoc, some shared with pieces across a whole
>>> culture) and inspired choices made within those rules.  You
>>> could then take it, change some of the rules, and make your
>>> own inspired choices to write a new piece of music.
>>>
>>> The advantage of working this way would be that you didn't
>>> have to worry about all the boring details because the
>>> computer could deal with them.  You'd tell the computer
>>> what you wanted in the simplest terms and correct it when
>>> it went wrong. You could also experiment new
>>> harmonic rules (which is something we're both interested
>>> in, at least).
>>>
>>> This is how I want algorithmic or computer-aided music to
>>> work.  From what I know of algorithmic music packages, it
>>> isn't the way they do work.  Rather, they expect you to
>>> pick an algorithm and see what comes out.  There are also
>>> tools that handle some simple things like repetition and
>>> transposition but don't go much further. If anybody can
>>> point me in a more useful direction, please do so.
>>>
>>>
>>>                                   Graham
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>>> csound"
>>>
>>>
>>
>> --
>> View this message in context:
>> http://old.nabble.com/-OT--Human-speech-is-music-to-out-ears-tp26629182p26796349.html
>> Sent from the Csound - General mailing list archive at Nabble.com.
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"


-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I'd like to add that, after glancing at David Cope's Web site, I have
to say he seems to have made some efforts in the past few years
towards making his software more cross-platform.

Regards,
Mike

On 12/15/09, Michael Gogins  wrote:
> Yes. Cope's work is based mainly on pattern matching from a corpus of
> examples. Not the same thing as what I am doing, which is pattern
> generation based on recursive rules that the user tweaks. In my work
> the "compression" that was mentioned is achieved by recursion. Cope's
> work seems to have some other "understanding" of musical hierarchical
> structure.
>
> Cope's results are very impressive! In terms of imitating or
> reproducing examples of styles up to a section or so in length, they
> go far beyond anyone else's work.
>
> However, I do not have the computing platform (Macontish Common Lisp)
> that is required to run his code, so I haven't been able to experiment
> with his software as I would like to. I'm sure I'll look into Cope's
> work more at some point. I have almost all Cope's books....
>
> Somewhat as with IRCAM, I think it is damaging to the researcher not
> to implement the research in one of the most widely accessible
> languages and platforms. Basically, I think a computer music person
> has to have a REALLY good reason not to use C, C++, Python, Scheme,
> Common Lisp, Lua, or some other language that is very widely used AND
> has good open source implementations.
>
> If I had Cope's software running, I am sure I would try to use to
> compose based on making up rules, instead of abstracting them from
> examples.
>
> Regards,
> Mike
>
> On 12/15/09, Rory Walsh  wrote:
>> Disclaimer: I know very little about this type of algorithmic music,
>> but hasn't David Cope tried to do something similar with his research?
>> I"m too lazy to google but his name stands out in my head....
>>
>> 2009/12/15 Aaron Krister Johnson :
>>>
>>>
>>> Graham,
>>>
>>> Yes, this description of an ideal software package---who *wouldn't* want
>>> such a tool?
>>>
>>> I have a sneaking intuition that if it were that easy, the music
>>> wouldn't
>>> be
>>> all that great either....there's a concept known as informational
>>> depth---the amount of time a process carries on is something akin to the
>>> quality of its result. There's a reason most symphonies can take months
>>> or
>>> years to write. And there's also a reason that a baby takes 9 months to
>>> grow
>>> in the womb--and they are not even done when they pop out of the womb!
>>>
>>> But, yes, wow---that would be cool....we are on the same page. I like
>>> your
>>> amplification of the compressibility concept into the idea of "music
>>> theory
>>> = compressibility of music" And it seems like such a software
>>> application
>>> would be complex, yet possible. Until then, we are stuck with primitive
>>> tools and our own intelligent ears. Maybe genetic algorithms and neural
>>> nets
>>> that might 'learn' an authors tastes have the best future in this
>>> regard.
>>>
>>> I've found that one thing we like in music is variety. Any one algorithm
>>> would then seem to get old, because it would tend to produce a
>>> homogenous
>>> result. One of the things that make music written by humans so nice and
>>> interesting (most of the time) are the complete and sudden changes of
>>> say
>>> tempo, texture, mood, etc. Of course, one could argue that that is more
>>> symptomatic of *Western* music in particular, or at least common
>>> practice
>>> Western music. Early Western music and Eastern music seem to be ok with
>>> a
>>> much more homogenous texture, and I think current algorithmic music
>>> comes
>>> much closer to 'modelling' those styles more successfully. Perhaps the
>>> greatest success of algorithmic music has been to convincingly produce
>>> neo-minimalistic music.
>>>
>>> AKJ
>>>
>>>
>>> Graham Breed wrote:
>>>>
>>>> Aaron Krister Johnson  wrote:
>>>>
>>>>> Another way of framing it: how compressible is a certain
>>>>> score---can it be compactly described?
>>>>
>>>> I think that's an interesting question.  The simple answer
>>>> is to take the score in a suitable digital form, and apply
>>>> a good compression algorithm to it.  The simple evasion is
>>>> to say that the score isn't the music anyway.
>>>>
>>>> But let's talk about a score, and let's say that we should
>>>> understand what the compression algorithm's doing.  We want
>>>> to be able to write music with it, after all.  There are
>>>> all kinds of ways that a piece of music can be compressed
>>>> -- repetitions and near repetitions, notes being more
>>>> likely to occur together, some notes more common overall,
>>>> some scales dominating different sections.  A good
>>>> compression algorithm that works over a whole body of music
>>>> would be a codification of music theory.  You could match
>>>> its rules up to the standard rules.  The simplest rules
>>>> that compress the music most must be the important ones.
>>>>
>>>> The result of this process would be the score of Eroica in
>>>> its tersest, human understandable form.  A set of rules
>>>> (some ad hoc, some shared with pieces across a whole
>>>> culture) and inspired choices made within those rules.  You
>>>> could then take it, change some of the rules, and make your
>>>> own inspired choices to write a new piece of music.
>>>>
>>>> The advantage of working this way would be that you didn't
>>>> have to worry about all the boring details because the
>>>> computer could deal with them.  You'd tell the computer
>>>> what you wanted in the simplest terms and correct it when
>>>> it went wrong. You could also experiment new
>>>> harmonic rules (which is something we're both interested
>>>> in, at least).
>>>>
>>>> This is how I want algorithmic or computer-aided music to
>>>> work.  From what I know of algorithmic music packages, it
>>>> isn't the way they do work.  Rather, they expect you to
>>>> pick an algorithm and see what comes out.  There are also
>>>> tools that handle some simple things like repetition and
>>>> transposition but don't go much further. If anybody can
>>>> point me in a more useful direction, please do so.
>>>>
>>>>
>>>>                                   Graham
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>>> "unsubscribe
>>>> csound"
>>>>
>>>>
>>>
>>> --
>>> View this message in context:
>>> http://old.nabble.com/-OT--Human-speech-is-music-to-out-ears-tp26629182p26796349.html
>>> Sent from the Csound - General mailing list archive at Nabble.com.
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>>> csound"
>>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>
>
> --
> Michael Gogins
> Irreducible Productions
> http://www.michael-gogins.com
> Michael dot Gogins at gmail dot com
>


-- 
Michael Gogins
Irreducible Productions
http://www.michael-gogins.com
Michael dot Gogins at gmail dot com


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Aaron Krister Johnson  wrote:

> Yes, this description of an ideal software package---who
> *wouldn't* want such a tool?

Well, it wouldn't suit people who are gifted improvisers
and don't see a need for the music to be written down.
What they might want is an auto-accompaniment tool, which
is a subtly different thing, and I know there is proprietary
software that tries to help them.

> I have a sneaking intuition that if it were that easy,
> the music wouldn't be all that great either....there's a
> concept known as informational depth---the amount of time
> a process carries on is something akin to the quality of
> its result. There's a reason most symphonies can take
> months or years to write. And there's also a reason that
> a baby takes 9 months to grow in the womb--and they are
> not even done when they pop out of the womb!

It'd be exactly as difficult as it needed to be.  Maybe
music notation is compressed enough that you can't save
much effort.  But I get the feeling that a lot of the work
in getting harmony or counterpoint to fit is mechanical
enough that a computer should be able to help you.  And,
for other styles, why can't the drum and bass lines sort
themselves out?

It's already a lot easier to write loop-based music with
computers.  The trouble is I don't always want loops.
There should be a way of extrapolating material without the
extreme repetition.

Actually writing the software would be very difficult.
I've thought about it, because I'm more inclined to spend a
long time writing code than music, but it's such a big job
that I haven't started.  I reckon there's enough in it for
a PhD, and I did even apply to start one, but I failed, so
there.

I think humanization, that is score->performance, is a
harder job for non-musicians than writing the score.  So
that's a better place to focus the effort.  But the more a
program understands the structure, the better it can make
performance decisions as well, I'd assume.  So the whole
thing could be integrated.

> But, yes, wow---that would be cool....we are on the same
> page. I like your amplification of the compressibility
> concept into the idea of "music theory = compressibility
> of music" And it seems like such a software application
> would be complex, yet possible. Until then, we are stuck
> with primitive tools and our own intelligent ears. Maybe
> genetic algorithms and neural nets that might 'learn' an
> authors tastes have the best future in this regard.

They might work, but it wouldn't be easy to understand what
they're doing.  Maybe for the humanization side but I don't
know I'd trust them for composition.  If it worked, of
course, that might be a different thing.

> I've found that one thing we like in music is variety.
> Any one algorithm would then seem to get old, because it
> would tend to produce a homogenous result. One of the
> things that make music written by humans so nice and
> interesting (most of the time) are the complete and
> sudden changes of say tempo, texture, mood, etc. Of
> course, one could argue that that is more symptomatic of
> *Western* music in particular, or at least common
> practice Western music. Early Western music and Eastern
> music seem to be ok with a much more homogenous texture,
> and I think current algorithmic music comes much closer
> to 'modelling' those styles more successfully. Perhaps
> the greatest success of algorithmic music has been to
> convincingly produce neo-minimalistic music.

Yes, variety's good, and that's something you'd optimize
for.  I wouldn't expect to keep using one set of rules.
Maybe a new set for each album.  And I was thinking of
disco as a good application.  There are a lot of songs that
follow pretty strict structures and the listeners don't
seem to care.  It can even be a good thing that you know
what's going to happen at any moment.

Current algorithmic music maybe nothing at all like what
I'm thinking about ;-)


                        Graham


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Michael Gogins  wrote:
> Yes. Cope's work is based mainly on pattern matching from
> a corpus of examples. Not the same thing as what I am
> doing, which is pattern generation based on recursive
> rules that the user tweaks. In my work the "compression"
> that was mentioned is achieved by recursion. Cope's work
> seems to have some other "understanding" of musical
> hierarchical structure.

I've certainly heard of him, but I don't know the details
of how he does it.  It still seems to be about finding good
rules, and then letting the computer write the music.  What
I want is to do the interesting stuff, and let the computer
fill in the boring details.  But maybe it's adaptable for
that.  And, of course, I'd want to adapt it to new styles
instead of copying old music.

> Somewhat as with IRCAM, I think it is damaging to the
> researcher not to implement the research in one of the
> most widely accessible languages and platforms.
> Basically, I think a computer music person has to have a
> REALLY good reason not to use C, C++, Python, Scheme,
> Common Lisp, Lua, or some other language that is very
> widely used AND has good open source implementations.

It'd be nice if he published something in an open access
journal as well . . .  As you said in another message, the
code's supposed to be standard Common Lisp now, except for
where it needs platform-specific libraries.  But the website
doesn't explain what they do.  And most are in these
proprietary "StuffIt" files that used to be common on
Macs.  All good excuses not to look at it right now.

This one's readable and claims to be standard:

ftp://arts.ucsc.edu/public/cope/gradus

It refers us to a CMJ article for documentation.


                            Graham


Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"