Luke Schierer discusses Pidgin, Open source and life
- — 10 October, 2007 07:04
A lot of open source projects fail to ever get off the ground let alone be successful. What, in your opinion are some of the keys that have made Pidgin what it is today?
The most critical thing to Pidgin's success is that those developing it use it. We have a need, we work to solve that need.
The next most critical thing is that we accept patches.
The next most critical thing is that we try hard to recognize those who have contributed in ways that are meaningful to them. We have failed at times, but usually having a series of patches accepted means more freedom to do things your way instead of ours, a shorter turn around time on the patches you submit, and, when things are working right, eventually the ability to commit code directly.
Projects fail for all sorts of reasons. They expect too much of their users. "Finish writing this while I ignore it and even look at your work only sporadically." Or there are personality conflicts. Or they become obsessed with who "owns" that bit of code. Or a better tool comes along.
Or, and I think this one is most frequent, there really is no "itch," no need for that project among people willing to work on it.
(For instance) Tons of people want support for voice and video over IM. Almost none of those people are willing to invest any time or any effort into seeing it happen. Until some are, we will likely keep pushing it to the back burner.
If that is the case for a whole project, if no one wants to work on it (and open source programming is a lot of work, and most of it done in free time), the project will absolutely fail.
What are some new features under planning/discussion/consideration for future releases?
We are working on a new version of the MSN code. Two summer of code projects have contributed to it, and many of our MSN users would benefit from it. We also need to work on the ICQ code. Someone popped up on the mailing list to talk about implementing a new version of AIM, since it appears the newest versions of WinAIM (AOL's windows client) are doing something different from the oscar we have implemented. Sean would like to get voice and video support going, but has not found the time. There are many other things we will look at. Some of them we haven't even thought of yet.
Where do you see Pidgin five years from now?
Today we have finished the core/ui split and have at least four clients built around it (Pidgin, Finch, Adium and Meebo). In five years there might be a native windows client, a QT client, or all sorts of other things. Or perhaps no one will be using IM five years from now. I could not begin to guess. If I had tried five years ago, I'd have been wrong.
Do you like the name Pidgin (which, correct me if I am wrong, is named after the universal language) more or less than GAIM?
Pidgin and Finch are much better names than gaim and gaim-text. libpurple (as a name) absolutely rocks. Both Pidgin and libpurple are the sort of punny names that, looking around, tend to amuse developers. (libpurple comes from prpl, which is short for PRotocol PLugin. We have referred to the code implementing AIM, Yahoo, MSN, and so on as "prpls" for years now). Pidgin also better expresses who we are and what we do far better than gaim ever could. By the way, it has been "Gaim" or "gaim" and not "GAIM" since around 1998. The idea that "gaim" could be an acronym really became ludicrous when we introduced support for other protocols, putting the idea that we are primarily an AIM client to death rules.