Re: [LAU] Audio Distribution Proposal...

From: Ray Rashif <schivmeister@email-addr-hidden>
Date: Thu Jun 04 2009 - 12:39:21 EEST

Anyone is free to create yet another distribution. Nobody should have the
authority to prevent any such initiative, even if it's an inconvenience
(adds more to choose from; adds more confusion?). What everyone should do,
however, is find a balance between convenience and accessibility, by which I
refer to the current norm of having a specialised platform (operating system
+ packages) in order to make Linux audio "accessible".

The problem with the current approach is that maintaining a specialised
distribution means basically just that - maintaining a distribution. This
involves the same manpower required to run a proper desktop platform, if
your intention is to "reach out to the masses". One other problem is that
each of these "mainstream" parent distributions have a particular release
cycle, which the respective initiative must follow. Often times, since Linux
audio is an effort primarily still in development stages, users see the need
to get their software from upstream project managers. In essence, a Linux
audio distro initiative is a bleeding-edge initiative. Even a stable
platform like 64Studio may not be considered functional by the masses.

The solution to this is to:

(1) have a rolling-release cycle
(2) provide only add-on packages; not a distribution
(3) have a framework to easily build software from upstream and add to the
package manager

For the vast majority, from what I have seen and heard (not experienced
though), Sidux meets these criteria. It is:

(1) Debian-based; convenient
(2) Rolling; accessible
(3) There are tools to create a package from a local build

For others, and from personal experience, there are Gentoo and Arch. I can
speak for both, because I've spent quite a lot of time on the former, and am
settled entirely with the latter for over 2 years. The main difference
between these two is that one is solely a source-based platform, which I've
come to find is more on the inconvenient side. With a bleeding-edge
initiative, one may have to compile software, but not the entire platform.
This is what Arch solves. To both sides of the spectrum of Linux users (new
and advanced), Arch has its demerits. However, for a specialised platform,
the merits are overwhelming:

(1) Rolling.
(2) No split packages; package foobar means software foobar; you get what
and how upstream intends you to get.
(3) Easy and code-simple package management for both binary and source,
thanks to:
    (a) simple buildscripts; it takes one mere minutes to add a local build
as a binary to the package database.
    (b) (2) directly means there are no further headaches while scripting a
new package.
(4) Due to the meta nature of the distribution, there's absolutely 0 need
for another distribution if the intention is to make it accessible for
specialised use.

However, both of these are not for the "masses" or mainstream Linux users.

Conclusion: See the project highlighted in the previous mail? That's the
saviour. An agnostic repository (within the parent distribution domain)
serving a whole load of users - precisely what CCRMA (or the other similar
initiatives) is. It's really simple - you don't need another distribution or
installer disc. You just install your desired Debian-based distribution and
add packages from that repository. Make that something like Sidux and you're
rockin'.

So I'd urge anyone thinking of creating an audio distro to put his/her
efforts into packaging for the Debian Multimedia Team (or help out in others
like CCRMA, Gentoo ProAudio, ArchAudio etc).

_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@email-addr-hidden
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
Received on Thu Jun 4 16:15:01 2009

This archive was generated by hypermail 2.1.8 : Thu Jun 04 2009 - 16:15:01 EEST