Re: [gentoo-dev] Removals reply

[Alec Warner]

On Sat, Feb 2, 2013 at 9:07 PM, Vaeth <vaeth@××××××××××××××××××××××××.de> wrote:
> Sorry, but I feel that I must explain once more:
>
>
>>>> When I came to Gentoo many years ago, this was a very rare problem,
>>>> but the removal of packages has tremendously increased, and it is
>>>> not only me who is observing this problem - there were already some
>>>> threads in the forums, and people planning to but not coming back
>>>> to Gentoo for this reason.
>>>
>>>
>>> Awesome so they could've get involved and maintain it themselves if they
>>> seen
>>> it so crushial for their lives.
>>
>>
>> Agreed. That's why there are last rights announcements with a 30 day
>> delay before the package is actually removed.
>
>
> So this 30 day delay will enable these people to get involved,
> especially for all the packages which were removed in the last years?
> Now it is apparent that an archive for dropped packages (in the
> form of keeping masked packages or some other form) is not needed:
> I want to join your club with the time machine!

Gentoo has been running for over 10 years without it, so I would argue
it is not *needed*.
Would it be nice? Sure! No one is saying 'hey your idea sucks!' We are
saying 'hey we are not really interested in doing this, but you should
feel free to do it yourself, and we would think it is cool and totally
support you.'

>
>
>> In a nutshell, the people complaining about removals should stop
>> complaining and start volunteering to maintain either the packages
>
>
> Yep!  That's the right attitude: Give the people 30 days (even those
> people who are currently not at Gentoo for whatever reason) to know
> years in advance all the software they might ever need and tell them,
> if in doubt, just to maintain hundreds of packages!

I don't believe for a second that it is the role of Gentoo to have
packaged 'any software the user might have ever needed, or will ever
need.' I'm unsure if this is the point you are making?

> It is *of course* their fault if they do not!
> Later, if they need a package, you can blame them that they have
> not voluntereered for all these packages they possibly might have needed,
> because years ago they had 30 days time to think about it (even longer
> if they took the time and resources to backup on their machines all
> tarballs of all packages).
>
> To be serious: If somebody *has* the resources to backup all dropped
> tarballs, he should please donate these resources to Gentoo, because
> Gentoo nowadays definitely cannot afford to waste a byte.

It has nothing to do with disk space. The tree has software packages
in it. I would argue that *most* of the software:

1) Has someone in Gentoo to maintain it.
2) Has a herd in Gentoo to maintain it.
3) Has a proxy-maintainer outside of Gentoo and a dev willing to
commit on behalf of that person to maintain it.

A minority of the software available is un-maintained. When I started
the treecleaner project, I wanted all the software in the tree to have
a maintainer; packages that had no maintainer would be removed. This
was basically voted down by the Gentoo Community. Instead we adopted a
less aggressive policy:

http://www.gentoo.org/proj/en/qa/treecleaners/policy.xml

Treecleaners also quasi-manage maintainer-needed packages.
http://www.gentoo.org/proj/en/qa/treecleaners/maintainer-needed.xml

Note that there are currently 734 packages 'assigned' to maintainer needed.

When I created the treecleaners project I really tried to make the
policies very straightforward so that when maskings occurred that
users would not be confused as to why the package was being removed,
and I also tried to make it clear how a user or developer could 'save'
a package.

>
> Or how else can it be explained that this idea of enabling people
> to use previous packages if they need to is fought so intensively?

Again I don't think anyone dismisses the idea. The problem is
implementing it is perhaps not as trivial as you think. If I pick on
the git migration as an example; we have essentially been 'trying to
migrate to git' for like 2 years. That is going very slowly, even
though everyone is basically on board with the idea.

This is not a small agile project. Very often you need folks in the
community willing to drive things to completion. mgorny's work on
python-r1.eclass and his multilib stuff is one recent example.

I think what you have failed to do is find someone in the developer
community who is really eager to implement your idea.

>
> All it costs is some amount of disk/mirror space.
> Or is it really that you *want* to blame the user and put
> pressure on him to maintain packages?

Why do you want to put pressure on *me* to maintain software I do not
want to maintain? Or put pressure on *Gentoo* to mirror someones
source code or binaries, for software Gentoo are no longer interested
in distributing?

>
> Strange that as soon as user's resources are required, Gentoo's
> attitude was always quite the opposite: E.g. it just *always* installs
> bash completion or systemd files (or previously also static libraries,
> although I must admit that the situation for the latter was now
> improved, fortunately) - the user must care about cleaning
> for himself if he does not want to waste resources.
>
> Sure, both is an admissible attitude - only the user is responsible
> for everything.
>
> It is just: *I* want to contribute only to a distribution which
> also cares somewhat about the users.
> Maybe some developers feel the same, but it is of course up to you
> to decide this.
>
> Regards
> Martin
>