[PD] [ftm] Re: FTM for PD

[Norbert Schnell]

On 6 sept. 10, at 12:12, Thomas Grill wrote:
> Hi Norbert,
>
>> Yes, I think the latest state of the development of FTM for Pd was  
>> already
>> pretty close (since the last adaptations to Max 5 we probably took  
>> back a
>> little distance).
>> But no, we have currently no plans to develop this further.
>>
>> In fact, I didn't have the impression, that Pd "needs" FTM very  
>> much, since
>> there are other approaches for handing "complex" data structures.
>
> Well, the basic built-in possibilities may be there but they haven't
> been developed into a convenient higher-level system. The VASP modular
> system that i have been developing for a while is similar to FTM but i
> don't have the time to maintain it any more and i would rather port
> the higher-level portions to FTM. I think that Pd needs FTM as much as
> Max needs it and i would appreciate patcher-level cross-platform
> compatibility a lot.

Unfortunately, the development of FTM has similar resource issues.
For the moment, we maintain FTM & Co in it's current state and add  
very minor features when they don't add anything that makes the  
maintenance more difficult.

>> For what concerns the Gabor and MnM libraries, that have certainly  
>> some
>> interesting functionalities that would be worth having in Pd, I  
>> also could
>> imagine alternative ways to integrated them into Pd (as well as into
>> Max/MSP/Jitter) relying on the data structures that are already  
>> existing in
>> these environments.
>>
>
> Certainly, but do you plan to open-source them?

"Plan" is certainly not the right word, but I think, open source  
licensing has to be seriously (re-)discussed in this context.

>>> For me, only the underlying FTMlib and the ftm.object and ftm.mess  
>>> objects
>>> are really required.
>>
>> Can you develop more on this point?
>
> It means that at this point i personally don't need mnm or gbr
> externals. The basic FTM functionality is sufficient - i would
> probably make a Python binding and use numpy/scipy for higher level
> stuff.

Ok.
Currently, we don't have the resources for this, but of course we  
would continue to support whoever who would like to continue to work  
in this direction as good as I can.

N.

> gr~~~
> FTM & Co
> wiki: http://ftm.ircam.fr/
> mailing-list: http://listes.ircam.fr/wws/info/ftm
> bug reports: mailto:ftm-bugs at ircam.fr