Re: Non-DDs as official Debian package maintainers
-----BEGIN PGP SIGNED MESSAGE-----
[responding to the list, assuming the private posting was an error]
On Tue, 27 Sep 2005 16:57:21 +0100
MJ Ray <email@example.com> wrote:
> > > DDs should not steal credit from maintainers. Sponsors should
> > > take responsibility for checking and vouching for what they
> > > upload, but that does not make them the package maintainer.
> > That is your opinion. I obviously disagree.
> Take it up with developers-reference if you wish, but I support
> NMs getting the credit.
I talk about responsibility. You seem to focus on credit. Possibly two
sides of same issue - let's try:
You support NMs taking the responsibility? How can they practically do
that? What happens if they fail? How do we deal with that, compared to
how we deal with DDs failing their responsibilities?
> > > Co-maintenance could not prove packaging skills, because you
> > > are fixing bugs for the NM. It is far better for someone to
> > > learn by fixing it themselves (with as much help as you are
> > > able to give, of course). Hopefully, they will upload on
> > > their own eventually.
> > Do you say that a requirement for getting into Debian is that no
> > bugs are fixed by others?
> No. I say that you cannot prove your skills if someone else is
> fixing your bugs without your involvement.
You prove some skills by fixing bugs. I agree on that.
What I fail to understand is what skills you prove by fixing all in a
package on your own.
> > > That is why I am against requiring maintainers to all end
> > > @debian.org. If any such requirement were added, I would try
> > > setting up firstname.lastname@example.org forwardings to screw that
> > > requirement.
> > I have no problem with that. It would simply mean that you -
> > visibly - act as a proxy for outsiders. My point is that we need
> > always an insider to take responsibility, and in order for our
> > social credit reward system to work, the responsibility must be
> > visible.
> Hiding non-DD maintainers will make it harder for the social
> credit reward system of NM to work. It's not perfect already.
Again: I talk about responsibility, not about credit. Let's try if we
are talking about the same:
Is our NM system based on letting those through that take a certain
amount of responsibility?
I would rather that responsibility for Debian resources was given only
If credit is possible without responsibility then I have no problem at
all giving credit to non-DDs.
> > > I am in favour of sponsors being listed in the control file,
> > > but they should not take credit for work they aren't doing.
> > > If you are requiring sponsors to do all maintainer things,
> > > then you are effectively abolishing sponsorship. What is the
> > > non-DD maint then? A DD's lackey?
> > >=20
> > > Wouldn't just requiring sponsors as uploaders satisfy your
> > > need to find the sponsor?
> > No. A sponsor is more than an uploader - a sponsor is _exactly_
> > taking all responsibilities for the package on behalf of the non-DD
> > that wants to do so herself but cannot because it requires being
> > part of the community.
> If you changed the policy on this, I'd expect mass-orphaning of
> sponsored packages.
I would too - and that is what worries me: We have packages in Debian
that noone within Debian takes responsibility for.
I'd rather that irresponsible Debian developers orphaned such packages
and they were either dropped from Debian officially (those outside of
Debian are off course still free to maintain them outside of Debian) or
adopted by Debian developers that actually care for them.
> > Let's imagine for a monent that we just add a new Sponsor: field in
> > addition to the Uploaders: field. What does it mean? What
> > responsibilities (and credit, if you prefer talking about that) are
> > tied to that task?
> Acting as first-line QA on the package.
> Hope that clarifies,
Yes, it does. And we disagree.
* Jonas Smedegaard - idealist og Internet-arkitekt
* Tlf.: +45 40843136 Website: http://dr.jones.dk/
- Enden er nær: http://www.shibumi.org/eoti.htm
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (GNU/Linux)
-----END PGP SIGNATURE-----