I was checking my email last night when I realized that there was a lot of activity on debian-i18n in the run-up to rc1 for etch d-i. Christian Perrier had sent me several email asking about the status of debian-tl and I had replied that nothing was wrong, just that I had been busy.
I should have posted about this on the mailing list in order that some of the work can be offloaded to others, but it may be a little too late for that.
Checking the status of d-i tl translations for levels 1-5, and considering what bubulle told me (he's not worried about tl since I replied in email), we're not in bad shape. However, we're not at the level of translation that I'd like TL to be in (i.e. 100% in all levels). It will need care and feeding, but maybe I can catch up before rc2. I need to put in some time on it after I take the LPI-1 cert exams on Saturday.
Somewhat related to TL, in discussions with Ariel, we are thinking of getting more involvement in localization work outside our current circle of influence. Rosetta is an excellent tool for doing collaborative translation that helps draw input from volunteers who are new to translating open source. As an introduction to l10n, it is okay. But for serious work, and for long term commitments, I feel that using l10n tools that are on the desktop are more efficient. I should look into the existing tools and see how they can be integrated with some form of network communication for coordination and revision control. Or are there methods that are already there, but not well documented? I should investigate this and write up something about it so that others who are interested in l10n can get into it.