l10n Memo for the Next Meeting

My personal short memo for the next meeting, even if I’m sure Axel is already on this:

  • Aurora is supposed to be string frozen, so that localizers have a full cycle to update their localization, test their work and sign-off the best changeset available for Beta. This worked quite well for 5 releases, why did everything go wrong this time? We’re just a couple of days away from the end of this cycle (Firefox 10 release, Jan 30th), a backout on toolkit broke everything1 and then a bug on devtools added even more confusion.
  • Being a Mozilla localizer already requires an awful amount of technical skills, please don’t even think of adding more stuff on top of that (“why can’t we or localizers just retrieve the previous string from hg blame?”).
  • Working on two different repositories is painful (see Native Fennec), I realized that I can’t transplant changesets around because often they change more strings that I need, so I have to move text around manually. I’m scared of seeing what will happen when I’ll merge my work from central to aurora.

1 Thanks to our l10n logic this is not literally true, since products fall back to the English string. From my point of view, this still means “breaking things”: exposing a partial translated UI means lowering the quality of our work, and that’s not something I like to do.

6 thoughts on “l10n Memo for the Next Meeting

  1. Hi Flod:

    “Working on two different repositories”… just TWO repositories!? Oh I see, it’s just AURORA FX you are talking about. Thunderbird, Beta, 3.1x etc are outside the context.

    Coming out: in my cygwin subdirectory there’s 11 (eleven) folders starting on HG. Even after discounting for my personal messy factor, the Mozilla l10n phase space encompasses more than just two repositories. It is a hell of a job for the localizers (median group size = 1) to stay abreast. As the Red Queen said: “It takes all the running to stay in the same place”.

    Regards

    smo

  2. just TWO repositories!? Oh I see, it’s just AURORA FX you are talking about. Thunderbird, Beta, 3.1x etc are outside the context.

    Wrong, I just meant 2 repositories (l10n-central and mozilla-aurora) ;-)

  3. You may know that I’ve been a localizer for ages (well, back to 1999, which is an eternity in Mozilla terms). Also, yes, if you look into the bug, I’m probably to blame for this backout. Also, if you look there, you’ll see that Axel was involved before the backout happened.
    Still, yes, I probably should have put more pressure on investigating this way earlier, but it looked like something had been *fixed* not like, as we found out very late now, that something actually had been *broken*. This means that Firefox 10 will ship without us being able to reasonably find out if plugin crashes or hangs are getting worse or better – in the end, *because of the L10n impact*, we agreed that we will not be able to reasonably diagnose *any* plugin/Flash crash/hang problem with Firefox 10 unless they’re really really really bad – for any not world-changing issues, we’ll just not have the data to even realize they’re there. If we wouldn’t have done this backout on Aurora – which, as you note, didn’t *break* everything, even though it might look bad, and where localizers still have 6 weeks of beta to make it completely OK – then we also wouldn’t be able to diagnose those things for Firefox 11. It’s bad enough to be in that situation for one release, it’s unacceptable for two, I hope you agree with me there.
    Yes, we messed up, and I’m somewhat to blame because I didn’t insist earlier that we investigate this. Unfortunately I believed it wasn’t a bug, but a feature, which was a mistake, in hindsight. Even as a localizer, I thought it would be better to have one sentence in a plugin crash message not translated than to miss 80-90% of all plugin crash/hang reports.
    I hope with that explanation, you can agree with this as well.

  4. Hi Robert, thanks for the explanation.
    Honestly I understand that something like that can happen and backing out that change was the only solution, still I believe it’s being managed badly: according to bsmedberg’s last comment we’ll have another break in Aurora’s string freeze when this bug is fixed, which seems quite silly.

  5. Pingback: Once upon a time there was a string freeze… | Yet Another Tech Blog