Re: How we make decisions... [Fwd: Re: Proposal: replacing esound with polypaudio in 2.10]

["Gustavo J. A. M. Carneiro" <gjc inescporto pt>]

  I think I speak for a lot of frustrated users and developers out
there, and I'd like to thank you for your effort.  Sound has been
negleted for too long in GNOME.  We have needed a new sound system for
years!  Keep up the good work!

  Best regards.

On Fri, 2004-11-12 at 20:16 +0100, Lennart Poettering wrote:
> On Thu, 11.11.04 19:48, Sri Ramkrishna (sri aracnet com) wrote:
> 
> Hi!
> 
> > By the end of the thread, the author of this particular piece of
> > technology has:
> > 
> > * No idea what he really needs to get this accepted by the GNOME 
> >   community. Sure  he's been asked to fix a couple of things.  Some
> >   minor complaints about not working on non-FPU systems but really no 
> >   appreciable list.  If we're serious, lets create a bugzilla component
> >   for him and get at least some branching and testing going.  Getting
> >   rid of Esound is great goal for 2.10 or the next release after and at
> >   least one distro is using it.
> 
> BTW: I (the polypaudio guy) created a small todo list from all the
> issues (or at least those I consider sensible and useful) that were
> discussed on this mailing list recently. And I am currently working on
> closing as many issues of the list as I can. The issues I already
> closed are:
> 
> - X11 root window property based authentication
> - IPv6 support
> - fall back to getpwnam when $HOME is not set
> - use setreuid() instead of setuid()
> - don't use access() any more
> - I investigated Alan Cox's point on TCP_NODELAY. However I cannot
>   follow him on this. If I don't enable this option on the TCP sockets 
>   the transfer latency becomes horribly undeterministic. And for
>   applications that repeatedly query latency information (like every
>   movie player around) this is more than counterproductive. Therefore
>   I reenabled TCP_NODELAY after a quick and unsuccessful test phase.
> - chdir("/")
> 
> The still not closed issues are:
> - non-fp mixing/resampling
> - an esound backend (i am currenlty working on this)
> - check getadrrinfo() results.
> - support for x11 synchronization extension
> - document all functions in the source code and limit resources of all kinds
> 
> I consider this a complete list of all issues I've been able to
> identify during the discussions on this mailing list. If something's
> missing, please tell me what.
> 
> I plan to release a new version of polypaudio when i closed all issues
> from the list above (except x11 synchronization), so that nobody is
> able to find another objection for adapting gnome for polypaudio.
> 
> > I could feel that he was getting frustrated and irritated dealing with
> > this community.
> 
> Not really. This is free software after all, I didn't expect it any
> better... ;-)
> 
> Lennart
> 
-- 
Gustavo J. A. M. Carneiro
<gjc inescporto pt> <gustavo users sourceforge net>
The universe is always one step beyond logic