GSoC idea - Plugin repository on pidgin.im

[Ibrahim Awwal]

Hi,

I'm a first year undergraduate EECS student at UC Berkeley and a
longtime user of Pidgin. I realize this is a little late but I just
thought of the idea and was not sure whether you guys would like it so
I decided to send this email before wasting my time applying. One of
the nice things about Pidgin is that it's so extensible and there are
a lot of plugins, but it can be hard to find them since there is no
one place to go to look for them. There is the Sourceforge tracker but
it's fairly out of date and right now it seems like the relationship
beween pidgin.im and the Sourceforge project page is a bit unclear.

I was envisioning that for a Summer of Code project, I could build a
plugin website similar to https://addons.mozilla.org , although
probably on a smaller scale to begin with. I would envision at least
having plugin categories, allowing submission of plugins which would
require some sort of administrative approval before being shown on the
site, keeping track of plugin versions, authors, and websites, and
download counts and user ratings.

I would also be interested in using this database as a backend for a
built-in version checker for plugins inside Pidgin, similar again to
what Firefox does. Currently, if I use plugins I have to remember to
check their websites every now and then for updates, and that's a lot
of websites to remember to check.

As for my experience, I have a significant amount of previous
experience with PHP and MySQL, and I'm also currently learning
Django/Python which seems like a cleaner way to build websites, but
I'm not sure about the performance hit. I have also worked with Google
App Engine (which is similar to Django from what I've seen), which
might be a good way to offload the load to another server since it
seems like the main pidgin.im server has had load issues in the past.
In fact, now that I've thought of that I would probably say that if
this project were to be accepted, I would use Google App Engine so as
to avoid putting a load on the main servers. I also consider myself to
be a pretty good programmer in general, and I do have experience with
C and Java in addition to the above languages.

Depending on what you guys think, I would follow the following timeline:
The first half of the summer or so, i.e. up to the mid-term
evaluations, I would work on the web app back end, basic front end,
and providing the data in a clean way (for the update checker or other
applications that want to use the data). For the second half, I would
work on polishing the front end and the update checker.

Again, I apologize for the late email but this was just something I
thought would be cool to have for Pidgin. Even if it is not accepted
for a GSoC project, I would like to work on it on my own time but
obviously if you guys don't feel like it's a good thing to have I
wouldn't go through with it. Let me know what you think,

--Ibrahim Awwal