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

Group maintainership

Rightio, given recent discussions I figure I might as well explain my
position on this group maintainership proposal to the list.

Before I continue, I should make it clear that I'm talking about what
I'd like for the packages I'm currently looking after.  I have tried to
mention this in my previous postings.  I'm not making any claims about
what's good for qt or for libs or base, just the less critical stuff
that I'm currently maintaining.

It basically boils down to this:

- I enjoy the work I do for debian;
- I am aware of the limits upon my time.

As a result, there is a small collection of packages that I have
volunteered to maintain.

- I get to enjoy maintaining these packages;
- I feel responsible for these packages and in particular this means
  that the dirty work as well as the fun work for these packages gets
- I don't feel personally responsible for other material (such as qt,
  libs, base, etc.) which I know I generally do not have time to help
  out with.

I should point out here that most of debian works like this.  People
mark off a little section of debian to look after; they get the
enjoyment and satisfaction of looking after this section and in return
these sections get an individual who is personally responsible for their

Consider now a move to group maintainership.  This has the advantage of
bringing a variety of expertise to help resolve difficult issues and
assist with large volumes of work.  However, it also has the potential to
lessen both the sense of personal satisfaction and the sense of personal

For this reason group maintainership seems to generally happen for large
and complicated packages or for packages that require a diversity of
expertise (e.g., gcc, openoffice.org).

The KDE modules I'm maintaining do not fall into these categories.  I
can see an argument for group maintainership of qt, kdelibs and/or
kdebase.  I don't see such a strong argument for group maintainership of
modules such as kdesdk or kdetoys.

I've already been accused of being overly possessive of "my packages".
I honestly don't think this is reasonable, given that:

- individual maintainership is the norm, not the exception;
- the packages I maintain are not in a particularly bad state of disrepair;
- most of the outstanding bugs are upstream bugs which are just as
  accessible to everyone (I have to submit patches upstream like everyone
  else), and so individual maintainership is not holding anyone back from

I can see arguments for group maintainership of core material such as
qt, kdelibs, etc, and I have no problem whatsoever with that.  However,
because this is a good solution for large core packages does not mean
it needs to be imposed upon smaller or non-core packages.

And of course it's neither pleasant nor necessary to have one's
character called into question for resisting this imposition.

Ben. :)

Attachment: signature.asc
Description: Digital signature

Reply to: