jkmeter libzita-convolver

[Free Ekanayaka]

Hi Reinhard,

|--==> On Thu, 26 Mar 2009 10:22:20 +0100, Reinhard Tartler <siretart at tauware.de> said:

  RT> Felipe Sateler <fsateler at gmail.com> writes:
  >>>IMO, at least one (preferably active) team member should be
  >>>co-maintainer of every package adopted by pkg-multimedia. I'd even
  >>>propose to make it mandatory for new packages.
  >>>
  >>>What do you think?
  >>
  >>I'm not really convinced... after all, if the package is well
  >>maintained it should be OK. If the original maintainer goes MIA, the
  >>package can just be removed...

  RT> While I also agree that my suggestion is quite strong, I don't buy your
  RT> argument. Let's have a look at our packages:

  RT> 46 packages:
  RT>  - http://qa.debian.org/developer.php?login=debian-multimedia%40lists.debian.org

  RT> 26 packages:
  RT>  - http://qa.debian.org/developer.php?login=pkg-multimedia-maintainers%40lists.alioth.debian.org

  RT> makes a total of 72 packages. How do you determine if the original
  RT> package maintainer went MIA and should be removed from the team?

If the package is sponsored by a team member, it would make sense to
set the Maintainer to:

pkg-multimedia-maintainers at lists.alioth.debian.org

and include the contributor and the uploader himself in Uploaders
(sorry if this was implicit for you).

Bug reports are usually a good way to get an idea of the health of a
packages. If there are too many, the package is probably not very
actively maintained. On the other hand if there is none, that smells
like no user is really caring about the package.

The QA pages Reinhard points are a very good way to get a sense of the
situation, I think all team members should go through them once every
while.

  >>While it is always good to have co-maintainers, I don't really think
  >>we are in a position to require that, specially from an active team
  >>member, since manpower seems to be running a bit short.

  RT> But manpower is exactly my point! IME additional package require more
  RT> manpower that we don't have. My suggestions says basically: Extra
  RT> packages are only accepted if the addition doesn't reduce the manpower
  RT> per package ratio.

I agree, so it makes sense to explicitly to a contributor whether he
is ready to maintain the package even after the first upload.

I believe that some decrease of the manpower per package ratio is
unavoidable, because team members will have to review the
contributor's work. However you can consider this a trade-off with the
hope that a given contributor will eventually improve to the point
where less or no review work is needed. That will pump up the ratio :)

Ciao!

Free