We got FUDCon started today here in Chile after a lot of months of hard work. The cold weather (at least for Brazilians) didn’t prevent us from have a nice time. Today we already had a lot of talks about all kind of subjects like GNOME 3, LATAM infrastructure, Inkscape and many more.
My first talk was about Fedora’s spins and remixes. I have never talked Spanish for that long and it has been a good opportunity to practice. I guess the audience got it right and I was able to reply the questions properly. You might want to take a look at the slides if you are able to read in Spanish. Also, it was great to meet the new Fedora Project Leader, Jared Smith.
Tomorrow Dennis Gilmore will give a talk about how to join Fedora Infrastructure and Daniel and Guillermo will talk about RPM packaging. Those are just the first talks, we will have a lot more! Give a look at the event wiki page for further information.
The World Cup isn’t over yet, but since Brazil is already out I expect Spain to be the champion. Random guesses apart, while the most important soccer tournament is still happening in South Africa I’m preparing myself for another event that will take place in Santiago, Chile. Next week the Fedora Project will be hosting the second Fedora Users and Developers Conference in Latin America. The first one took place in Porto Alegre, Brazil last year and I was there too.
This year we will have the presence of the new Fedora Project Leader, Jared Smith, and also interesting sessions from Toshio Kuratomi, Dennis Gilmore, Daniel Bruno. Jayme Ayres, Rodrigo Padula, Domingo Becker and many more folks from Fedora.
I will be holding a technical session about spins and remixes in Fedora, and also a hackfest regarding internationalization and localization quality assurance. It will be a good opportunity to meet some friends I haven’t seen for a while and also a great opportunity to create something good for Fedora. This is our own World Cup and Open Source is the big winner.
I have been learning how to use gettext and intltool in order to get some code internationalized. At first I thought it would very difficult to learn and get practical results but in fact the i18n libraries and tools are easy to work with.
The first choice was how to test them in a useful way and so I chose to do it using the Preupgrade code. Although I’m not very familiar with Python code I did this because this software is not internationalized yet and because it is “real code”, not just a Hello World example. I guess this is the greatest thing about open source software, you can just dig in and do whatever you want to and give it back, so IMHO this is a good place to start and a patch can come later. Another interesting thing about gettext is that it works pretty much in the same way for other languages. This is good because I intent to collaborate providing i18n support for other programs as well.
So far I extracted 80 strings into the .POT file but I was not able to get the dialog titles internationalized yet, although they are marked as translatable in the Glade XML file. Once I fix this it will be good to go and I’ll find out how to get the i18n support automatically built into the package.
Since I’m leaving the coordination of the Brazilian Portuguese translation team after four years of a rewarding job, I expect to do something different in the next years.
My main concern about the whole software translation system right now is that for many times the teams make a good work but for many reasons the translations don’t end in the final versions of Fedora. These reasons are for instance: strings freeze breaks, final freeze breaks, downstream patching and specific i18n bugs.
IMHO, at first we need to improve the testing on those situations, and this is where I intend to start helping. After that, I intent to propose some patches rather than just file tons of bugs. It is important to make the i18n string testing consistent during the release cycle, and the Fedora 13 release cycle seems a great time to start.