Subject: Re: [linux-audio-dev] testing the waters
From: Erik Steffl (esteffl_AT_pbi.net)
Date: Tue Jun 27 2000 - 21:04:46 EEST
???
why not contribute to existing distro?
maintaining the distribution is tremendous amount of work. the audio
part of it is pretty small. why not simply concentrate on audio part and
use one of the existing distros as base?
about the only thing that's needed (other then applications) is low
latency...
as for setup: turning off daemons that are not needed can be fairly
easily done in any distro, you can supply script to do that (even for
more than one distro). and btw most daemons just sleep when they are not
used, so who cares?
I would say that the most effort should go into kernel level, mainly
achieving low latency and good drivers, because without stable drivers
(both in stability of code and stability of API) that are good enough
(alsa seems to be good enough while oss seems to be stable enough:-) the
apps are developed much slower. once there are drivers, we can continue
with the rest of framework + apps. that's just my opinion, of course, I
am not telling anybody what they SHOULD do...
erik
Joe Miklojcik wrote:
>
> I couldn't agree more with this, and had been thinking it myself. "Linux
> Audio Distro" has been on my list of pipe dreams for more than a year now.
> I know at least five little studios and one big one that would probably love
> the current state of the art in Linux Audio software, unfinished as it may
> be, but can't get there because one of the barriers to entry is getting
> everything compiled and installed.
>
> Nonetheless, I'd rather we all concentrate on finishing up the software
> projects we've started before moving on to integrating it all into a single
> clean distro. This appears to me as one of those cases in which Linux is an
> exercise in patience.
>
> US$0.02
> ----
> Joe Miklojcik
>
> John Lazzaro wrote:
>
> > [...]
> > Here's a more radical suggestion: market a Linux distribution that's
> > customized for audio applications, and market it through your existing
> > channels, i.e. Pro Audio and music-store chains. You'll notice as you
> > read this list that a constant theme is the wish to integrate low-latency
> > kernel patches into the main Linux distribution. By putting together a
> > distribution that:
> >
> > -- included low-latency kernel patches, as well as the real-time
> > POSIX timer patches that haven't made it into the kernel yet.
> >
> > -- consciously made configuration choices from the bottom up to
> > fit in a low-latency audio world (default daemons, ect)
> >
> > -- had the full range of existing GPL'd audio applications as part of
> > the package, well integrated and documented, with tutorials ready
> > to run on the desktop.
> >
> > -- and perhaps, a few of your company's core applications, either as
> > cut-down closed-source binary versions at one extreme, or more
> > open if possible.
> >
> > you'll have an experimental vehicle for understanding how to market
> > open-source software in the pro-audio and music world, that doesn't
> > put your company's core IP at risk. A distro like this really brings
> > to mind Red Hat's marketing theme, that millions of people buy Heinz's
> > catsup even though they could by fresh tomatoes and make their own at
> > home cheaper. Many people will pay $49 or $69 USD for a well-done
> > package as I describe above, rather than do it themselves -- and while
> > its true they will be free to pass the disk along to N of thei friends,
> > it's also true that they'd have an incentive to buy a new one two or
> > three times a year to upgrade to the latest and greatest. [...]
This archive was generated by hypermail 2b28 : Wed Jun 28 2000 - 00:12:40 EEST