Request For Candidates: Application Review Board

[Iain Lane]

Hello all,

On Mon, Aug 09, 2010 at 02:25:44PM -0400, Scott Kitterman wrote:
>On Thursday, August 05, 2010 08:42:58 pm Jono Bacon wrote:
>> As such, if you are an
>> application developer and want to get your app in the software center,
>> the process is probably too complex and involved.
>
>In what way is this new process simpler and less involved?

I'm glad that this is finally being discussed on -devel. At UDS I
expressed a significant level of concern about this specification and
the ramifications it will have on the development community, and on
Universe and the archive as a whole. It will, in my opinion, knock the
new application process out of the alignment that it's currently in,
in much a similar way to the effect that PPAs have already had.

(I've recently posted a couple of lengthy messages to this list, one
on the subject of backports being too difficult to get done, and
another on the REVU process being unsuitable. I think that both of
these trains of thought are linked to this spec, and that they were
probably sparked off by the discussions at UDS. These are where we
should be concentrating our efforts IMHO.)

At the same time, I am loathe to get into any heated or heavy
discussion which will probably not lead to anything happening right
away; this is probably a situation in which hindsight will prove
invaluable. I just want to put my concerns on record now, although
they are already in the audio records from UDS if anyone cares to look
there. There were two sessions on this.

The summary of my message is, much as my fellow MOTUs have already
outlined: please use backports. Fixing the spec that Scott linked to,
in addition to some bureaucratic changes, is the right way to achieve
what we want here.

Here's what I believe the proposers of this specification see as the
problem. Please correct me if I'm wrong:

   Getting cool new applications into Ubuntu stable releases is too
   hard. Policy is too difficult to navigate. The procedures for review
   and acceptance into the archive are impossible to
   understand. Finding sponsors is vexing. Running the development
   release when targeting the stable release is undesirable and
   difficult. There is no visibility amongst the thousands of other
   applications available in Universe.

So the proposed solution is to create a board to bypass all of
this. Essentially the board receives and evaluates applications and
then puts them into an archive where they are visible to the software
centre. These applications receive special promotion, as I understand
it.

Others have covered some issues that they see. I'll briefly outline
mine, with some overlap:

   1. This process will necessarily split the already /significantly/
      understaffed pool of package reviewers if it is to be in any way
      effective (i.e. if the ARB is to be able to review applications in
      a timely fashion).
   2. There is no path for packages into the development release.
   3. This process bypasses MOTU, Universe, the rest of the archive and
      Debian. I think this will be a really rather large hit on team
      morale.
   4. The incentive to get your application in Debian and naturally to
      Ubuntu will be removed. Why, as an application developer, would I
      want to have my application available in Debian if it's just going
      to hinder my ability to get it out to users of Ubuntu stable
      releases? (IOW: The process denigrates backports, which is the
      current preferred way of doing this)
   5. The message is that the archive is not the place for your cool app
      to live. I think this process is what Matt Z was getting at in a
      recent post on his blog[0].

I want to expand on 5 a little bit. I found a comment from Matthew T
in the blog post:

,----
| Almost every day, OMG Ubuntu features interesting free software for
| Ubuntu. Today, for example, it was Burg. Yesterday, it was
| GmailWatcher. The day before that, it was Pinta. None of those three
| are available in the Ubuntu repositories — and that’s typical.
`----

Leaving aside the fact that when this blog was posted, Pinta was
indeed packaged and available in Maverick (by me), this raises an
interesting point. The comment goes on:

,----
| Packaging applications is what OS developers do when the OS is
| unpopular. Ubuntu is now popular enough that application developers
| are increasingly interested in packaging their software for Ubuntu
| themselves (as they already do with Windows and Mac OS X). This does
| not mean they’re the slightest bit interested in becoming a MOTU or a
| DD; they want to package their own software, not anyone else’s.
`----

This is likely to be offensive to a lot of distribution developers. It
reads like: “There is no place for you to package upstream
software. Upstream developers are better placed than you to do
this. Please go away”. Just because something may not happen as
rapidly as you may like doesn't mean that the whole system is rotten.

If developers are interested in doing this properly then there is
absolutely no requirement to become a MOTU or DD. Both distributions
have methods to have finer-grained upload control: PPU in Ubuntu and
DM in Debian. This would be the proper way to have your package out
there, with all of the hard-won QA processes that we have.

But why would you do that when you can just point your users to a PPA
and not have to answer to anyone else?

I've said enough. The ideal process, if indeed there needs to be a
special class of applications, is expedited backports backed by
upstream developers becoming PPUs. Note that I don't have an answer
for the transient “Super Bowl 2010” style of applications that Rick
raised. We can fix this without rearchitecting the whole world and
risking the bond that our distribution developers have with the
distribution that they use and voluntarily do work for.

Cheers, Iain

[0]
http://mdzlog.alcor.net/2010/07/06/weve-packaged-all-of-the-free-software-what-now/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
Url : https://lists.ubuntu.com/archives/ubuntu-devel/attachments/20100814/437d698c/attachment.pgp