Came back yesterday from Zaragoza, where we had the III hackers meeting. It wasn’t as productive as I expected though 🙁 I thought we could have had people working more intensively in fixing bugs, but only 2 bugs were solved, and very simple ones.
So, it’s clear that we’re doing something wrong. For the people already involved in GNOME (garnacho, kal, telemaco, tapia, juanan, dexem, etc), these meetings are productive, since we dedicate some time to hack on our already ongoing projects, we share ideas, and we have fun together at nights. But for new people, I think we really need some way of having them start working on bugs immediately, so that we can have this sort of weekends be real hacking days, and have them as GNOME contributors right after the weekend.
We talked about this during the weekend, and some possible solutions raised, like re-starting the IRC talks, so that we teach people how to hack before we get together on the weekend. Any other ideas for making these weekends more productive?
Apart from that, as always when meeting with the GNOME Hispano people, it was really fun, lots of laughs, good music (we went to a R&R bar in central Zaragoza on Saturday night), good food and, unfortunately, little sleep.
Also, one of the hopefully productive parts was about the book, since we talked about getting all people more committed to completing it, which is not much work, since we already have around 400 pages!. Hopefully people will respond to the challenge, so that the book is ready before GUADEC 2006 in Barcelona.
On the “Contributing to GNOME” subject, I have realised just lately what was blocking me from contributing (besides from the lack of time – but we all know that :-P). This is mainly concerning contributing code and to some extent documentation.
Glib. Glib and especially the GObject system, are in fact quite non-trivial to grasp. There are excellent tutorials around for high-level GTK and the various language bindings, but none readily googlable on the inner workings of GTK and GNOME.
Trying to work on something or just browsing through some code, it is quickly obvious that you’ll need a good understanding of glib/gobjects to do non-trivial tasks.
If all you know is Java, Python and C (like me) you’ll be thoroughly confused the first time you look at the code behind a GNOME app. Writing a widget/gobject takes a whole lot more than writing a Java/Python object.
With time I’ve come to understand the GObject system to some extent, but don’t feel very comfortable with it. On the other hand it didn’t take me more than a day to start playing around with QT4 and feeling quite comfortable – even though I’ve never really digged C++
Well… Didn’t know where to post this rant, I just came to think of it when I saw your blog.
Cheers
Mikkel
Well, the purpose of these meetings is to get people to hack on high level bugs, fixing apps, applets, etc, rather than working on library code. So, there is really no need to have them know about the inner workings of glib/gtk.
Of course, what is needed indeed is to have those people all know GTK/GNOME programming a bit before coming to the meeting.