What was the biggest development challenge in Pidgin 2.2.0 and how did you resolve it?
With 2.2.0 we merged in several of the Summer of Code 2007 projects. Each of these had their own challenges and issues across the summer.
From my own perspective, the biggest on going challenge was the amount of processing power needed to run our new trac ticketing and wiki install. With help from the trac developers, we were able to pin this down to an expensive lookup in checking permissions for re-assigning tickets, and optimize it.
For an open source project to function successfully, people are needed in a number of different roles. Developers are of course always in the shortest supply, but my own role is more of a support role. I triage many of the incoming tickets, and I read all of them. I do much of the work keeping our server running, and I co-ordinate with the packagers for the various Linux distributions.
How many Google Summer of Code students contributed to 2.2.0?
Three of the students had their work merged in for 2.2.0. Much of Eric (Polino)'s work was already in by that point or he would be a fourth. It looks like the MSN code, a fifth project, will still be merged in, but probably not for this next release. Perhaps the one after.
What was it like working with Google summer of Code students?
We have had a very positive experience with the Google Summer of Code, both this year and in previous years. Most of the students work diligently and produce good results. Some of them have stayed with us, expanding our development team. Others have moved on to other projects. Either way, they leave with a stronger appreciation for what goes into an open source project, and the unique mentality that comes with working on something that is not run with the goal of making money.
Will this be an ongoing partnership?
This was our third Summer of Code. I see no reason why we would not continue in the future.
You mention in a September blog that you were told that 90 percent of users think that every change you have made since 2.0.0 is a step backwards.
I, and my co-developers think that the reality is actually the reverse. (There is a) disconnect between those complaining and our own picture of the user base.
How do you respond to this criticism?
If it were true, obviously we would have to rethink some things. That being said, it is impossible to please everyone. At the end of the day, if we have improved the experience of those we personally know, we've succeeded. Historically we have known (and are) a diverse enough user base to match the use patterns of a large body out there. I expect this will continue, and that if we are happier, many of our users will be also.