What does it mean to be part of gnome games?

["Thomas H.P. Andersen" <phomes gmail com>]

Hi all,

There is a lot of discussion about being part of gnome games lately.

Let's take a look back. In the past we had all the games in a single
module "gnome-games". This was done for convenience. They were
released together, had a single bugzilla name, one mailinglist, etc.
They also shared some code in an internal library. We also created one
tarball and shipped that. It was up to distributions to use the
configure flag to select which games they wanted to build and ship.

While having it in one module made some things easy there were also
drawbacks. We had a module with code in C, C++, python, vala, js, and
scheme. The configure file was 1000+ lines,  the compile time was
long, and the build/distcheck broke all the time due to the many
dependencies. (It was while everything was in one module that we
decided to move to a common language.)

We always had more games than we had time to maintain and develop.
Dropping games was hard because it meant to delete them completely
from the module. Including new games required that they would fit into
the module and that we were comfortable with taking over the
maintenance.

In the end we decided to just split up gnome-games into separate
modules. One git repo, bugzilla, etc, per game. All that is left of
"gnome games" is the shared mailing list, #gnome-games, and a page on
the wiki (which should just be removed imo). We have even started to
split the games on different maintainers. We still collaborate and
work together as a group on the games but that is just a detail.

So I think think that people should not try to become part of "gnome
games" because it simply no longer exists. A game using gnome
technologies does not need our blessing to become part of gnome. Just
apply for a git/bugzilla account like any project. We may or may not
choose to help out with development/maintainership. Just like all any
other project that will happen based on whether we like the game and
whether the code is in a hackable state. Since most of the current
developers have a preference for vala then that might be a good
choice. Python or other languages might attract others though.

- Thomas