Re: Time to move GGV to the Attic?

["Christian Rose" <menthos gnome org>]

On 7/18/06, Bill Haneman <Bill Haneman sun com> wrote:
> > > On Mon, 2006-07-17 at 11:34 -0600, Andreas J. Guelzow wrote:
> > > > I agree. But I don't think that anyone is suggesting that evince is
> > > > better than GGV because it is being actively developed. GGV is not being
> > > > _maintained_, therefore no one is doing any source code modification,
> > > > therefore it doesn't need an active source code repository.
>
> Given the fact that the Gnome development environment (autotools, etc.)
> is far from static, I think there is reason to keep even "stable"
> components in CVS, where build fixes and bugfixes can easily be
> applied.  This applies even if there is no "active" maintainer, provided
> the module provides sufficient utility to Gnome to justify a place in
> the repository.
>
> While there is clearly a desire/intention to migrate fully to Evince
> someday, it seems that there are still significant 'gaps' that ggv can
> fill.  Archiving it means that support for building it is likely to fade
> away, possibly quickly.

Seriously, if anyone has even the slightest interest in keeping a
particular non-maintained module in a buildable state, they should
step up and request to maintain it.

In my experience, it's perfectly fine to state something to the effect
of "this module is in low maintenance mode; my main interest is just
to keep it buildable, don't expect any features to be added, and if I
fix bugs it will only be if I have the time for it. Releases will be
made only when needed. Patches and/or help with co-maintaining it
welcome."
We've had several cases like this, where a module lost its maintainers
and was going to bitrot away, but where someone who cared even the
slightest stepped up as an emergency maintainer and kept it at least
buildable during a period of time, until either this person found some
more time for development, or someone else came by and offered to
maintain it. Many of these modules still have active development
today, so emergency maintaining seems to be an effective way of both
preventing a piece of software from being abondened completely, and in
making it interesting for new aspiring maintainers.

If someone steps up and wants to be such an "emergency maintainer" for
the module, then you're right, the archive isn't the right place for
the module. Then it should be moved back into the main repo, since it
has someone that sort of maintains it.
The archive is only for modules where noone has so far offered to
maintain it in any way.


Christian