[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Proposal for collaborative maintenance of packages

On Wed, 2005-12-28 at 23:50 -0200, Gustavo Franco wrote:

> > (...)
> > I wasn't addressing your point specifically. Instead, I'm raising
> > another one. The whole Debian system is a serious pain and an
> > impediment to cooperation. Ubuntu is not much better.
> > 
> > I should not have to be a DD to commit to
> > the archive -- ANYONE should be able to commit perhaps with
> > some minimal registration requirements.
> > 
> > This is how Wikipedia works and why it is successful. With minimal
> > fuss I have contributed some comments and a couple of changes.


> There is no serious pain in how Debian and Ubuntu are handling their
> packages, IMHO. 

How would YOU know??? You're a DD. I'm not. You have access
to resources .. I don't. But I am the one actually building
the software. I am not meaning to be rude: I'm pointing out that
insiders are NOT qualified to comment on how outsiders feel
about how easy it is for THEM to contribute.

I run Ubuntu .. I can't even test my own Debian package
on my own system.

> The wikipedia style of colaboration just don't work when
> you need to prepare a package from scratch (as you address below).

That may be so.

> That's good, you cited a real world problem above with the deb packages
> of your software. If the package for your software is outdated you
> ("upstream") or any user, should file a bug (severity: wishlist) against
> the package through our "bug tracking system" requesting for its update.

That isn't really the point. you're asking me to do even more
administrative work .. when I'm seeking a way to actually
do some of the real work myself .. to relieve the burden on the
current maintainer.

> The current maintainer simple don't have enough time to update it ? 

The current maintainer is part of a team dealing with a major *set*
of packages. My package is not even really in that set. His priorities
are rightly to put that set first (and due to a weakness of Debian
every release and patch of the core of the set requires all the 
other packages to be re-uploaded even though none of them have
changed, this is because every change to the core changes the ABI)

> If
> it's important to the users to upgrade to that new version, they will
> (and they usually do) add more requests on that bug report.

There aren't many users because the dang thing isn't being 
distributed by Debian because of all the delays.

>  the ultimate
> solution is create a group into alioth.debian.org to share the workload.

I tried to do that and was refused.

> I disagree, mainly because i started reading the oficial documentation
> at www.debian.org/doc/ when i was 16 years old. I think you don't want
> to learn how to prepare deb packages taking care about all the QA
> issues,

Not at all. As I said I spent a month preparing the packaging
myself. My problem isn't in meeting QA requirements .. its
in getting the dang work actually committed to the archive
in a timely manner. 

I actually LIKE very much the Debian
approach of producing Policy documents -- even if I disagree
sometimes with the policies, they're standards to work by
and documents that can be changed. Indeed, the Policy
documents are the main reason why Debian is so good and thus
highly regarded -- and so many other distros are Debian based :)

My problem isn't with the quality requirements, but with
the turnaround: upstream releases are sometimes as frequent as 
once per week (occasionally several times a day!).

> I don't believe that they're raising the bar in terms of package quality.

You're missing something: the 'package quality' is secondary.
What's important is the software IN the package. 

> > BTW: I know of dozens of pieces of fine software that SHOULD
> > be in Debian. 

> Have you heard about WNPP (www.debian.org/devel/wnpp) ? 

Yup -- I have some idea of most of the Debian processes 
and policies .. though I'm definitely no expert on them.

> Do they filed
> bugs requesting package preparation ? 

I doubt it -- they'd need to be very eager to get their
stuff into Debian to learn enough about it to discover
what a WNPP is. Most of these people are more interested
in actually developing new ideas and new software than
spending time learning about a packaging system. They usually
provide source code and scripts that build the package on 
some Unix system and leave it at that.

Most of the people who are interested in THEIR communities
are happy with source builds.

The result is that a lot of potential users OUTSIDE their
communities are not getting to even hear about their work.

I might be willing to package some of these things myself.
What puts me off is not doing the packaging work -- its
having to hassle about getting a DD to upload it. And then
hassle them every time there is an upgrade. I don't really
like hassling people.

> We can't keep high quality and let
> anyone with a good piece of software but no caring about Debian
> packages, package your stuff and upload it.

Yes you can, if you have an entry level repository.
Lintian is a pretty good first order check isn't it?

> > EG: My comp is on the net sometimes and sitting here idle.

> What about the security of this procedure ? It is in fact a good idea
> but not easy to implement as it sounds.

I agree. I don't even claim it is a good idea, just that
it is AN idea worth considering. Especially if there were
some entry level repository where novices would be prematurely
throwing stuff in to be built.

This would be REALLY USEFUL to me. I only have access to
an X86_64. I can't build for 386, PPC, Alpha, ia64 etc.
So an entry level repository with an autobuilding service
would be very useful to me as an *upstream* developer
to make sure the code actually builds and runs on all platforms.

John Skaller <skaller at users dot sf dot net>
Felix, successor to C++: http://felix.sf.net

Reply to: