Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
[FxOS] 2.5 Schedule Info
20 views
Skip to first unread message
delphine
Aug 12, 2015, 5:58:57 PM8/12/15
to
Hi localizers,
Some important updates concerning 2.5 from an l10n point of view:
* as you have noticed (and as has been discussed in previous messages), 2.5 isn't open to localization work yet. We're expecting this to be very soon though. We were waiting to get more clarity on the upcoming schedule, and things are only slowly starting to clear up. Will keep you all posted here once 2.5 is open for l10n work
* At the time, 2.5 string freeze is expected to be on Nov 2nd[1]. They are trying a different model for this release, in order to launch multiple Mozilla products at the same time (Fall release).
However for Firefox OS this makes things complicated l10n-wise... In fact, string freeze is on the day they want to "release", since they are letting features land until that date. We therefore can not have a string freeze before then.
Potential partners will be aware of this, and will know that for a localized and *tested* version, they will have to wait for the usual 8 week l10n cycle following string freeze[2]
What does this mean for you? If you want to have a localized build available at the Release date (Nov.2), that's possible and it's up to you. You can regularly test your strings on the device while you localize during the development cycle of 2.5, and fix the issues you see as you go along.
Or if you prefer, you can also wait until the release/string freeze date and start localizing then. You will then have 2 weeks until testing cycle begins, as usual.
More updates to come soon. Please let me know if you have any questions.
thanks
[1] 2.5 Schedule wiki: https://wiki.mozilla.org/Firefox_OS/Releases/2.5
[2] https://wiki.mozilla.org/L10n:B2G/Planning#Average_timeline_for_Localization_cycle_in_FxOS
Some important updates concerning 2.5 from an l10n point of view:
* as you have noticed (and as has been discussed in previous messages), 2.5 isn't open to localization work yet. We're expecting this to be very soon though. We were waiting to get more clarity on the upcoming schedule, and things are only slowly starting to clear up. Will keep you all posted here once 2.5 is open for l10n work
* At the time, 2.5 string freeze is expected to be on Nov 2nd[1]. They are trying a different model for this release, in order to launch multiple Mozilla products at the same time (Fall release).
However for Firefox OS this makes things complicated l10n-wise... In fact, string freeze is on the day they want to "release", since they are letting features land until that date. We therefore can not have a string freeze before then.
Potential partners will be aware of this, and will know that for a localized and *tested* version, they will have to wait for the usual 8 week l10n cycle following string freeze[2]
What does this mean for you? If you want to have a localized build available at the Release date (Nov.2), that's possible and it's up to you. You can regularly test your strings on the device while you localize during the development cycle of 2.5, and fix the issues you see as you go along.
Or if you prefer, you can also wait until the release/string freeze date and start localizing then. You will then have 2 weeks until testing cycle begins, as usual.
More updates to come soon. Please let me know if you have any questions.
thanks
[1] 2.5 Schedule wiki: https://wiki.mozilla.org/Firefox_OS/Releases/2.5
[2] https://wiki.mozilla.org/L10n:B2G/Planning#Average_timeline_for_Localization_cycle_in_FxOS
Michal Stanke
Aug 13, 2015, 3:44:28 AM8/13/15
to dev-...@lists.mozilla.org, Matěj Hudec
Hi Delphine,
thank you for the information. When the freeze date is the same as
release, how it is about feature and UI testing? Because IIRC, at least
one of the l10n issues (truncation) found during last cycle was a
regression caused by button margin/padding change. It would be very
unpleasant to force localizers to choose between truncation and some
least-wrong translation.
I have nothing against continuous localization even at the cost of some
of the work gets obsolete again during the cycle. But if there is no
pre-release "freezed" time, all the testing before freeze can be wasted
by 1px change somewhere in the UI (and we wouldn't even know it).
Cheers,
Michal Stanke
Dne 12.8.2015 v 23:58 delphine napsal(a):
> _______________________________________________
> dev-l10n mailing list
> dev-...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-l10n
thank you for the information. When the freeze date is the same as
release, how it is about feature and UI testing? Because IIRC, at least
one of the l10n issues (truncation) found during last cycle was a
regression caused by button margin/padding change. It would be very
unpleasant to force localizers to choose between truncation and some
least-wrong translation.
I have nothing against continuous localization even at the cost of some
of the work gets obsolete again during the cycle. But if there is no
pre-release "freezed" time, all the testing before freeze can be wasted
by 1px change somewhere in the UI (and we wouldn't even know it).
Cheers,
Michal Stanke
Dne 12.8.2015 v 23:58 delphine napsal(a):
> dev-l10n mailing list
> dev-...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-l10n
delphine
Aug 13, 2015, 2:06:58 PM8/13/15
to
Le jeudi 13 août 2015 00:44:28 UTC-7, Michal Stanke a écrit :
> Hi Delphine,
>
> thank you for the information. When the freeze date is the same as
> release, how it is about feature and UI testing? Because IIRC, at least
> one of the l10n issues (truncation) found during last cycle was a
> regression caused by button margin/padding change. It would be very
> unpleasant to force localizers to choose between truncation and some
> least-wrong translation.
Hi Michal!
> Hi Delphine,
>
> thank you for the information. When the freeze date is the same as
> release, how it is about feature and UI testing? Because IIRC, at least
> one of the l10n issues (truncation) found during last cycle was a
> regression caused by button margin/padding change. It would be very
> unpleasant to force localizers to choose between truncation and some
> least-wrong translation.
Thanks for these very important questions.
Concerning the above: we are planning on doing an l10n testing round mid-dev cycle, on locales that are up to date, in order to identify these regressions earlier.
We should also have some ongoing automation testing on the Settings app, that will help us identify these issues more quickly (with time we hope this will cover other apps as well)
QAnalyst testers will also be doing daily exploratory testing on locales that are up to date, as well as pseudolocales.
Finally, I highely encourage that if any localizer finds such regression when testing his strings while she/he is localizing, that she/he files a bug and CC me so we can make sure to get on top of things as early as possible.
Hopefully between these 4 things, this will help catch regressions/bad issues much earlier than in previous versions :)
>
> I have nothing against continuous localization even at the cost of some
> of the work gets obsolete again during the cycle. But if there is no
> pre-release "freezed" time, all the testing before freeze can be wasted
> by 1px change somewhere in the UI (and we wouldn't even know it).
>
So yes, there is a moment in the cycle called "Release Checkpoint" on Oct 2nd (so 1 month before "release/string freeze") where features should be stabilizing. This means that 2 weeks before "release/string freeze" you should expect minimal changes to features and strings.
I'll update here should anything change. Also, I will send out an easy to read l10n calendar for 2.5 in order to help make things clearer.
thanks!
cheers
delphine
Michal Stanke
Aug 13, 2015, 4:14:45 PM8/13/15
to dev-...@lists.mozilla.org, Matěj Hudec
Hopefully there won't be many "checkpoint" breaking. :) Anyway we will
try to do our best and keep up our localization with devs. Please let
me, Matěj or both know when the testing will be on the agenda, so we
finish the current strings ASAP and test them.
As for the automated testing, I would suggest to test also the
First-Time-Experience guide. It's usually done once and not visited
again by the testers (personally when I flash and clean my Flame, I skip
it entirely), but might have bad impact on the real users (Czech Android
OS l10n has on me).
Cheers,
Michal Stanke
Dne 13.8.2015 v 20:06 delphine napsal(a):
try to do our best and keep up our localization with devs. Please let
me, Matěj or both know when the testing will be on the agenda, so we
finish the current strings ASAP and test them.
As for the automated testing, I would suggest to test also the
First-Time-Experience guide. It's usually done once and not visited
again by the testers (personally when I flash and clean my Flame, I skip
it entirely), but might have bad impact on the real users (Czech Android
OS l10n has on me).
Cheers,
Michal Stanke
Dne 13.8.2015 v 20:06 delphine napsal(a):
delphine
Aug 13, 2015, 6:33:34 PM8/13/15
to
Le jeudi 13 août 2015 13:14:45 UTC-7, Michal Stanke a écrit :
> Hopefully there won't be many "checkpoint" breaking. :) Anyway we will
> try to do our best and keep up our localization with devs.
THanks! :D
> Hopefully there won't be many "checkpoint" breaking. :) Anyway we will
> try to do our best and keep up our localization with devs.
> Please let me, Matěj or both know when the testing will be on the agenda, so we
> finish the current strings ASAP and test them.
>
Yes, I will definitely regularly give updates here.
> finish the current strings ASAP and test them.
>
> As for the automated testing, I would suggest to test also the
> First-Time-Experience guide. It's usually done once and not visited
> again by the testers (personally when I flash and clean my Flame, I skip
> it entirely), but might have bad impact on the real users (Czech Android
> OS l10n has on me).
We'll also be going over the FTE as part of our localization runs, post string freeze
Thanks for all this valuable feedback Michal
0 new messages