| i have been doing quite a bit of database development lately in filemaker
pro which runs cross platform on mac, windows, and can work off the web as
well. would be willing to create structure to deal with the information on
opcode status in filemaker that could be easily updated by any number of
people given access to it on web and willing to do the *hard* work of
compiling and entering the "easy to update" information.
disclaimer: i work on mac and have not ever tested on windows though many
people do run filemaker on both (allegedly fast becoming the best selling
database). and have never posted any of my databases on the web. filemaker
list postings indicate that there shouldn't be too many wrinkles.
tolve
>seems no one had answered on that so I dare to:
>
>I think such a database could really improve the usabily of csound (at least
>for me...). And I am willing to collect infos.
>But I think: if it is used as a collection of know problems, the really
>question
>is: how keep such a db up to date?
>
>Torsten Anders
>
>
>On Wed, 01 Sep 1999 you wrote:
>>Hi, I wonder if anyone else thinks that it would be very usefull if
>>there were some easy way to compile some sort of database of the
>>"correctness" of the various csound orc ugens and function generators.
>>In other words, is there a workable way to provide some sort of
>>information as to what ugens are generally known to work, and which ones
>>are either untested or known to be broken?
>>
>>I'm thinking both of the person who is casually trying out csound for
>>the first time, and most likely does not want to immediately venture
>>into source-level debugging, and also of people like me, whose time is
>>limited, and hence might in some circumstances want to avoid using
>>opcodes that have not been verified to work as advertised.
>>
>>I suspect that at any given point in time, probably somewhere between 10
>>and 50 percent of the included opcodes don't work correctly in many
>>common circumstances. This is just a wild guess based on my own
>>experience, and from my reading of this mailing list.
>>
>>So, it would be helpfull to know which of these ugens are heavilly used
>>and generally assumed to work, and which are the ones that haven't been
>>really exercised by anyone who knows enough to tell that they don't work
>>right.
>>
>>Of course, the problem here is how to compile this sort of information.
>>I don't have an answer for this. Maybe for starters, if people would
>>post to the list when they have used a certain ugen and found no
>>problems? Of course, this would add a lot of traffic to this list, but
>>maybe it would be worth it.
>>
>>Larry Troxler
|