Trying to get to the bottom of my mailbox :-)
In short... I would not be so fast to dismiss DSSI... but there's still time
and lots of other things to implement so we can worry about this later ;-).
some comments below.
On Tuesday 20 July 2004 11.47, Werner Schweer wrote:
> On Monday 19 July 2004 23:47, Mathias Lundgren wrote:
> > Ayh!
> > Robert and I briefly discussed DSSI (http://dssi.sourceforge.net/) and
> > the possibility of combining DSSI and MESS one way or another. Is this
> > anything you've looked at, Werner? Would it be possible to, in a not
> > that horrible way, use DSSI synths through MESS, or should one go for
> > MESS via DSSI - feels like this might induce a bit of overhead? Or
> > perhaps we should wait and only go MESS until DSSI is not "disposable"
> > anymore? Does anyone have an opinion of this thingie?
> i dont understand the intention of this proposal. If they think that
> GMPI will be the solution they should work on GMPI.
I understand their motivation. GMPI was started as a collaborative effort with
_lots_ of people involved. Working on it is extremely slow and it can
probably not be speeded up... lots of bureaucracy...
Hence...to scratch an itch some people came up with DSSI as an interim
solution... (with a possible long life though the name, DISPOSABLE, claims
> Another different
> "standard" is contra productive.
True but possibly inevitable.
> If people use it and build
> synthesizer with this interface,
> it will be much harder for GMPI.
Mmm, though I think the future of GMPI has bigger obstacles than that to
overcome... several companies (Cakewalk for instance) are behind and deeply
entwined in the future of this effort. It is hard to hypothesise about where
it will go.
Nevertheless it'll be interesting to follow.
> They write:
> "The MusE sequencer has an API for hosting soft synths, known as
> M.E.S.S. It is probably the closest alternative to the present
> proposal, it addresses many of the same problems, it is working now,
> and it is an attractively small API. Unfortunately it is unlikely to
> be popular as a standard for Linux audio plugins, simply because it is
> in C++ and requires Qt for plugin GUIs."
> Yes, its unlikely that MESS be popular as a "Linux standard".
> Its intended to be "experimental". So you are invited to experiment
> with it. Some goals:
> - be small and fast
> - rich functionality
> - move complexity from plugin to host
> BTW MESS does not _require_ Qt for plugin GUIs. Every plugin
> can start a separate GUI process (like DSSI does).
Right, though I think there are Qt datatypes in MESS?
> I personally believe that its very hard to design an interface
> only on paper. DSSI does not much provide of a solution.
It's not entirely on paper, there are actually some synths adapted (fluidsynth
for instance). I'd say the status of it is about the same as MESS... working
but still a little experimental and due to change...
> In the
> MusE context its just another problem.
> My opinion is that you should ignore DSSI until there are some
> attractive synthesizer available which are using it. Instead be
> creative and try to make MESS better. Its easy to change and you
> can test your ideas because there is a complete working host-client
I agree whole heartedly with this. There are lots and lots and lots of things
to fix and add to MusE, we can add to the TODO list when there are good
> This SF.Net email is sponsored by BEA Weblogic Workshop
> FREE Java Enterprise J2EE developer tools!
> Get your free copy of BEA WebLogic Workshop 8.1 today.
> Lmuse-developer mailing list