Posted on March 24, 2015
Mozilla l10n Sprint at FOSDEM 2015
Disclaimer: This is quite overdue, sorry for that, last weeks have been crazy. Also thanks to Quentin Fremeaux aka Popzelife, Rep from France for writing his own blogpost regarding the l10n Sprint, which you can find here. Also, for those who do not know; l10n stands for “localization”.
l10n is a sensitive contribution area within communities, where Mozilla is no exception. There are so many ways one can contribute, it’s hard finding a path on your own if you don’t know where to start. To my perception, l10n is like a Monopoly game; very fun to get involved in, however it has a huge potential to spark discussions and debates (to say it softly) alone for the fact that languages and locales are not so black & white like they used to be, after technology blurred borders between languages and their digital involvement.
I was proud to be sponsored at FOSDEM by the l10n team of Mozilla. If you want to read about my beer and Mozilla adventures in Brussels last January, head over to my blogpost. Otherwise, read further for a short report of the l10n meeting we Mozillians held at FOSDEM.
In the last years, the focus on l10n at FOSDEM kept declining. It was also noticeable by the Mozillians sponsored to FOSDEM by the l10n team; although not necessarily a bad thing (we try to be focused on our impact and budget), it was unusual to lack l10n presence at FOSDEM. This year I was the only one sponsored by Mozilla l10n so I wanted to give my contribution in this aspect, making an effort to gather a few localizers for a meeting in Brussels.
We soon realized it was not suitable to hold an “classic”localization sprint where actual l10n work is done, so we tried to stay realistic and ping-pong a few thoughts to the future of l10n in our local communities. At the end, Quentin Fremeaux, Edoardo Putti, Tim Maks van den Broek, Daniele Scasciafratte, Gabriele Falasca, Lyubomir Popov and I were able to meet up in person to discuss how we can improve contribution strategies Mozilla can provide to l10n communities. We had a solid representation of different countries and communities, specifically:
- Albania (sq)
- Italy (it)
- France (fr)
- Netherlands (nl)
- Bulgaria (bg)
We noticed that a lot of communities have different workflows for localization, some among them not being best practices. In particular, the italian community widely used their forums for localization work, which is not optimized for such use cases as localization (this is pretty self explanatory).
As an example from the French community, Transvision has been successfully adapted and is used for a good amount of l10n work for the fr locale (among others). We recognized that it’s a solid tool to use within our l10n teams and highly suggest it to other communities to create their own forks of it.
Check it out on GitHub.
Depending on their contribution areas, Mozillians might be more into going neck deep into actual l10n work. This is great in the short run, but chances are that in the long run this might not be ideal. Getting involved with fellow contributors and hosting l10n sprints or other community events keeps the momentum going and should not be underrated. It might seem as fruitless work at first, but at the end of the day Mozilla is an open community where collaboration is critical for our mission. Further, we would hate to see contributors working alone on contribution areas, which sadly still happens in the l10n communities. Community Building is an investment which will ensure the health of a community for much more time to come. The urge to jump right into actual contribution work might be hard to resist, but if there is no community backing the work being done, chances are it won’t last long.
If you want to do an effort regarding this, and don’t know where to start, consider contacting your nearest Mozilla Rep, so you can get the ball rolling (If in doubt, feel free to ping me too).
At Mozilla Reps (aka ReMo) we are lucky to have established organizational structures which facilitate to a great degree the needs of all Mozilla Reps ( “certain degree” because we are not perfect and constantly improve our self”). You can even see it via our mentorship structure on the Reps Portal.
Unfortunately, not all of Mozilla’s project are well structured and defined; l10n being one of them. We have recognized a sense of old school “I know it better” attitude from some localizers who have contributed for years at Mozilla; but are eventually not familiar with the open work environment Mozilla is based on. This should not be a rant, but I think we (semi)secretly know that this is more or less a problem at Mozilla. Although a very blurry topic (we cannot and should not generalize localization matters) there is space for improvement here.
We were brainstorming about the idea to have l10n Mentors who can guide new contributors with best practices. A mix between a Webmaker Mentor and a Mozilla Reps Mentor, but specifically for l10n. Apart from being great localizers; the priority for these mentors should be community builiding, in order to facilitate contribution paths for new contributors in localization.
These are the main points we have gathered from our discussions. Please note that these are thoughts on best practices from a brainstorming session. We highly recommend them, but we might be also missing a point here or there. At the end of the day this solely serves to initiate a discussion and for future reference.
I personally am highly fond of the l10n Mentors idea and would love to see it coming to life. If you are into l10n, feel free to reach out to me so we can discuss possibilities.