[iOS] Firefox for iOS 3.0 localization

[Jeff Beatty]

Hi everyone,

Sorry this email is a bit delayed. I'm hoping to consolidate everything
that's been said so far (plus other details) into this one email for you to
reference back to during this cycle. Let's begin, shall we?

*String freeze*: 29 February
*Localization freeze*: 20 March
*Strings profile*: 31 new strings.

- ~32% are for managing authentications,
- ~19% contain additional brand/feature names and don't require
translation (e.g., Twitter, Touch ID, etc.),
- ~16% are about bookmark management,
- ~16% are about private data management,
- and the remaining 20% are for general use of the browser.

*L10n process changes*:

- *Testing*:
- As Stefan mentioned, we have automated l10n screenshots for testing!!!
You no longer 'need' to have an iOS device to help with testing your
locale. To find your locale's screenshots, visit
http://people.mozilla.org/~sarentz/fxios/screenshots/ . We're hoping
we can make this iterative too so that you can see your changes
soon after
they hit the repos. Testing will be very important this cycle
and in future
cycles as described below.
- *Sign-offs for shipping locales*:
- We're going to experiment with eliminating sign-offs for locales we're
already shipping. Instead, on 20 March we'll ship the last commit of each
locale that meets the following criteria:
- 100% complete translation with no remaining errors on the web
dashboard,
<https://l10n.mozilla-community.org/~flod/webstatus/?product=firefox-ios>
- Have no visible l10n errors upon review of the locale's
screenshots (e.g., significant truncation, text overflow,
encoding, etc.).
- *Sign-offs for new locales:*
- We will continue to work through bugzilla for processing new locale
sign-offs. The criteria for shipping these locales will be the
same as the
criteria mentioned above.

I'm sure many of you will breathe a sigh of relief after reading that about
sign-offs :-)

As it goes with experimentation, the first time is never perfect. We're
looking forward to working with you to refine this process. You've already
given some great suggestions for closing the feedback loop and adding
context to strings within our translation environments! Please continue to
let us know where parts of the process don't make sense or need changes.

As always, if you have any questions or comments, please let us know.
Looking forward to shipping a lot of localizations on 20 March!
--
Thanks,
Jeff

[Stefan Plewako]

> Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com> w dniu 03.03.2016, o godz. 17:41:

>
> - We're going to experiment with eliminating sign-offs for locales we're
> already shipping. Instead, on 20 March we'll ship the last commit of each
> locale that meets the following criteria:
> - 100% complete translation with no remaining errors on the web
> dashboard,
> <https://l10n.mozilla-community.org/~flod/webstatus/?product=firefox-ios>
> - Have no visible l10n errors upon review of the locale's
> screenshots (e.g., significant truncation, text overflow,
> encoding, etc.).

Does it mean that Polish and other already released locales will be excluded from 3.0+?
Anticipating softening "solutions": how significancy will be judged?

While I understand that quality is important and the line must be drawn somewhere, I remember to well how many Firefox OS bugs (that weren't neglected) popped in each version. Honestly, I have hard time understanding how anyone knowing anything about localization issues on small screens could make such suggestions.

[Michal Stanke]

Hi Jeff.

Would it be possible to mark the commit used for release by a git tag
(e.g. in format "{locale-code}-{version}" or vice versa), so it can be
tracked back in case of bug reports etc.?

Cheers,
Michal Stanke

Dne 3.3.2016 v 17:41 Jeff Beatty napsal(a):

> - We're going to experiment with eliminating sign-offs for locales we're
> already shipping. Instead, on 20 March we'll ship the last commit of each
> locale that meets the following criteria:
> - 100% complete translation with no remaining errors on the web
> dashboard,
> <https://l10n.mozilla-community.org/~flod/webstatus/?product=firefox-ios>
> - Have no visible l10n errors upon review of the locale's
> screenshots (e.g., significant truncation, text overflow,
> encoding, etc.).

[Jeff Beatty]

Great question. That's something we should discuss with Stefan and the iOS
team. I imagine that shouldn't be problematic.

> _______________________________________________
> dev-l10n mailing list
> dev-...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-l10n
>



--
Thanks,
Jeff

[Jeff Beatty]

Thanks for asking. Perhaps I misphrased. To clarify, all locales that have
been shipping iOS will continue to do so. We simply won't require an
explicit sign-off like we've done in past versions through bugzilla.
Instead, if your locale is 100% complete with no dashboard errors, we'll do
a manual technical review your screenshots (or on the latest build) and
accept the last repo commit.

"Significancy" will be discussed with the l10n team in question (i.e.,
we'll file a bug about it). We certainly aren't experts in each language
and will consult with you to learn if the translation meaning has been lost
due to the truncation (making it significant) or not (making it minor).

By doing this, we're stressing the importance of completion and testing
while eliminating the (possibly) unnecesary act of signing off.

On Thu, Mar 3, 2016 at 11:57 AM, Stefan Plewako <sple...@aviary.pl> wrote:

> > Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com> w dniu
> 03.03.2016, o godz. 17:41:
> >

> > - We're going to experiment with eliminating sign-offs for locales
> we're
> > already shipping. Instead, on 20 March we'll ship the last commit
> of each
> > locale that meets the following criteria:
> > - 100% complete translation with no remaining errors on the web
> > dashboard,
> > <
> https://l10n.mozilla-community.org/~flod/webstatus/?product=firefox-ios>
> > - Have no visible l10n errors upon review of the locale's
> > screenshots (e.g., significant truncation, text overflow,
> > encoding, etc.).
>

> Does it mean that Polish and other already released locales will be
> excluded from 3.0+?
> Anticipating softening "solutions": how significancy will be judged?
>
> While I understand that quality is important and the line must be drawn
> somewhere, I remember to well how many Firefox OS bugs (that weren't
> neglected) popped in each version. Honestly, I have hard time understanding
> how anyone knowing anything about localization issues on small screens
> could make such suggestions.
>

[Stefan Plewako]

>> Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com> w dniu 03.03.2016, o godz. 17:41:
>>
>> - We're going to experiment with eliminating sign-offs for locales we're
>> already shipping. Instead, on 20 March we'll ship the last commit of each
>> locale that meets the following criteria:
>> - 100% complete translation with no remaining errors on the web
>> dashboard,
>> <https://l10n.mozilla-community.org/~flod/webstatus/?product=firefox-ios>
>> - Have no visible l10n errors upon review of the locale's
>> screenshots (e.g., significant truncation, text overflow,
>> encoding, etc.).

> Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com> w dniu 03.03.2016, o godz. 20:07:

>
> Thanks for asking. Perhaps I misphrased. To clarify, all locales that have been shipping iOS will continue to do so. We simply won't require an explicit sign-off like we've done in past versions through bugzilla. Instead, if your locale is 100% complete with no dashboard errors, we'll do a manual technical review your screenshots (or on the latest build) and accept the last repo commit.
>
> "Significancy" will be discussed with the l10n team in question (i.e., we'll file a bug about it). We certainly aren't experts in each language and will consult with you to learn if the translation meaning has been lost due to the truncation (making it significant) or not (making it minor).
>
> By doing this, we're stressing the importance of completion and testing while eliminating the (possibly) unnecesary act of signing off.

What will happen when there is in fact significantly truncated translation since version 2.0 and the problem was signaled and bug was reported long before release (since it cannot be reasonably fixed on l10n side)?

Does it mean that locale with such problem will be excluded from the release under new rules? (or what they actually mean really)

[Jeff Beatty]

Thank you for rephrasing. I understand the concern much better now.

If the truncation issue cannot be solved through retranslation but requires
involvement from the iOS team, that won't be counted against the locales
that are affected by the technical problem.

[Axel Hecht]

I think we should tag the git repo for releases, yes. It will end up
being a global tag (or branch, not sure I care either way), for all
locales at once.

That's just because managing several dozens of versions can't be done
without significant investments in tooling.

Axel

[Stefan Plewako]

Maybe holding localization teams responsible for being responsive on bmo would make better rule then (if it is not already)?

Different approach would be to actually requiring sign-offs from locales with issues but as they seem to be common it wouldn't change that much comparing to the old rules.

Additionally, I like the direction but I also have the feeling that Mozilla is starting to have disproportional demands - it would be against the rule to have fixable in localization truncation but it is OK to neglect it when it is not fixable on localization side?

> Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com> w dniu 03.03.2016, o godz. 20:43:

> Thank you for rephrasing. I understand the concern much better now.
>
> If the truncation issue cannot be solved through retranslation but requires involvement from the iOS team, that won't be counted against the locales that are affected by the technical problem.
>
> El mar. 3, 2016 12:29 PM, "Stefan Plewako" <sple...@aviary.pl> escribió:
> >> Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com> w dniu 03.03.2016, o godz. 17:41:
> >>

> >> - We're going to experiment with eliminating sign-offs for locales we're
> >> already shipping. Instead, on 20 March we'll ship the last commit of each
> >> locale that meets the following criteria:
> >> - 100% complete translation with no remaining errors on the web
> >> dashboard,
> >> <https://l10n.mozilla-community.org/~flod/webstatus/?product=firefox-ios>
> >> - Have no visible l10n errors upon review of the locale's
> >> screenshots (e.g., significant truncation, text overflow,
> >> encoding, etc.).
>

[Stefan Plewako]

> Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com> w dniu 03.03.2016, o godz. 22:55:

> El mar. 3, 2016 2:44 PM, "Stefan Plewako" <sple...@aviary.pl <mailto:sple...@aviary.pl>> escribió:
> > Different approach would be to actually requiring sign-offs from locales with issues but as they seem to be common it wouldn't change that much comparing to the old rules.
> >
> > Additionally, I like the direction but I also have the feeling that Mozilla is starting to have disproportional demands - it would be against the rule to have fixable in localization truncation but it is OK to neglect it when it is not fixable on localization side?
> >

> It's not a question of neglect but what the l10n community has influence over within their sphere. If it's a common code problem affected many locales, it needs to be reported beyond the l10n community and involve more people from development. I don't consider that neglect. The issue at that point is how to make sure it lands on the iOS team's radar. At the end of last cycle, we were able to make good progress there. We can improve more this time around too.
>


How would you describe something other than neglecting when the problem hits most locales, was reported months ago, doesn't look complicated at all - not beyond one developer attention and for sure not to the degree that the report itself doesn't get even some sort of ack, not to mention decision?

Do you seriously think we should accept that filling bug in right place is not enough to get it on iOS team's radar? I have also no idea what kind of progress you have in mind.

One could also play significance card and sure, finding other bugs two work on wouldn't be hard but it would also lead to demands proportionality topic again.

[Jeff Beatty]

El mar. 3, 2016 2:44 PM, "Stefan Plewako" <sple...@aviary.pl> escribió:
>
> Maybe holding localization teams responsible for being responsive on bmo
would make better rule then (if it is not already)?
>

That's implied, but not specifically called out.

> Different approach would be to actually requiring sign-offs from locales
with issues but as they seem to be common it wouldn't change that much
comparing to the old rules.
>
> Additionally, I like the direction but I also have the feeling that
Mozilla is starting to have disproportional demands - it would be against
the rule to have fixable in localization truncation but it is OK to neglect
it when it is not fixable on localization side?
>
It's not a question of neglect but what the l10n community has influence
over within their sphere. If it's a common code problem affected many
locales, it needs to be reported beyond the l10n community and involve more
people from development. I don't consider that neglect. The issue at that
point is how to make sure it lands on the iOS team's radar. At the end of
last cycle, we were able to make good progress there. We can improve more
this time around too.

> > Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com> w dniu

03.03.2016, o godz. 20:43:
> > Thank you for rephrasing. I understand the concern much better now.
> >
> > If the truncation issue cannot be solved through retranslation but
requires involvement from the iOS team, that won't be counted against the
locales that are affected by the technical problem.
> >
> > El mar. 3, 2016 12:29 PM, "Stefan Plewako" <sple...@aviary.pl>
escribió:

> > >> Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com> w dniu

03.03.2016, o godz. 17:41:
> > >>
> > >> - We're going to experiment with eliminating sign-offs for locales
we're
> > >> already shipping. Instead, on 20 March we'll ship the last
commit of each
> > >> locale that meets the following criteria:
> > >> - 100% complete translation with no remaining errors on the
web
> > >> dashboard,
> > >> <
https://l10n.mozilla-community.org/~flod/webstatus/?product=firefox-ios>
> > >> - Have no visible l10n errors upon review of the locale's
> > >> screenshots (e.g., significant truncation, text overflow,
> > >> encoding, etc.).
> >

> > > Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com> w dniu

03.03.2016, o godz. 20:07:
> > >
> > > Thanks for asking. Perhaps I misphrased. To clarify, all locales that
have been shipping iOS will continue to do so. We simply won't require an
explicit sign-off like we've done in past versions through bugzilla.
Instead, if your locale is 100% complete with no dashboard errors, we'll do
a manual technical review your screenshots (or on the latest build) and
accept the last repo commit.
> > >
> > > "Significancy" will be discussed with the l10n team in question
(i.e., we'll file a bug about it). We certainly aren't experts in each
language and will consult with you to learn if the translation meaning has
been lost due to the truncation (making it significant) or not (making it
minor).
> > >
> > > By doing this, we're stressing the importance of completion and
testing while eliminating the (possibly) unnecesary act of signing off.
> >
> > What will happen when there is in fact significantly truncated
translation since version 2.0 and the problem was signaled and bug was
reported long before release (since it cannot be reasonably fixed on l10n
side)?
> >
> > Does it mean that locale with such problem will be excluded from the
release under new rules? (or what they actually mean really)

[Michael Wolf]

Jeff Beatty schrieb:

> - *Testing*:
> - As Stefan mentioned, we have automated l10n screenshots for testing!!!
> You no longer 'need' to have an iOS device to help with testing your
> locale. To find your locale's screenshots, visit
> http://people.mozilla.org/~sarentz/fxios/screenshots/ . We're hoping
> we can make this iterative too so that you can see your changes
> soon after
> they hit the repos. Testing will be very important this cycle
> and in future
> cycles as described below.

Hi,

what's with the both Sorbian locales (hsb and dsb)? They are not listed.
It's really important for me because I do not have an iOS device.


Thanks,

Michael W.

[Jeff Beatty]

No idea. Thanks for pointing that out. Stefan mentioned that screenshots
failed for some locales. Stefan, is that the case for hsb & dsb?

[Jeff Beatty]

Send me some examples privately and let's discuss it calmly with the iOS
team. I was under the impression we'd resolved most in 2.0.

El mar. 3, 2016 4:28 PM, "Stefan Plewako" <sple...@aviary.pl> escribió:

> > Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com> w dniu

> 03.03.2016, o godz. 22:55:
> > El mar. 3, 2016 2:44 PM, "Stefan Plewako" <sple...@aviary.pl <mailto:
> sple...@aviary.pl>> escribió:

> > > Different approach would be to actually requiring sign-offs from
> locales with issues but as they seem to be common it wouldn't change that
> much comparing to the old rules.
> > >
> > > Additionally, I like the direction but I also have the feeling that
> Mozilla is starting to have disproportional demands - it would be against
> the rule to have fixable in localization truncation but it is OK to neglect
> it when it is not fixable on localization side?
> > >
> > It's not a question of neglect but what the l10n community has influence
> over within their sphere. If it's a common code problem affected many
> locales, it needs to be reported beyond the l10n community and involve more
> people from development. I don't consider that neglect. The issue at that
> point is how to make sure it lands on the iOS team's radar. At the end of
> last cycle, we were able to make good progress there. We can improve more
> this time around too.
> >
>
>

> How would you describe something other than neglecting when the problem
> hits most locales, was reported months ago, doesn't look complicated at all
> - not beyond one developer attention and for sure not to the degree that
> the report itself doesn't get even some sort of ack, not to mention
> decision?
>
> Do you seriously think we should accept that filling bug in right place is
> not enough to get it on iOS team's radar? I have also no idea what kind of
> progress you have in mind.
>
> One could also play significance card and sure, finding other bugs two
> work on wouldn't be hard but it would also lead to demands proportionality
> topic again.
>

[Stefan Arentz]

We (iOS team) have not been good with dealing with l10n bugs. The screenshots make that very clear as there are a bunch of cases with truncated text and other very obvious errors.

I'll find out how we can get these bugs more visibly on our radar and how to make them part of our weekly triage. It is possible that we do not see them because we do not check the right component or because they do not have the right flags. That is all easy to fix and we will.

We want to ship a quality product and that means we will deal with those bugs.

We have a couple of weeks of stabilization time before we submit 3.0 to the App Store. Part of that will be dealing with regressions and bugs and the l10n ones are a big part of that.

More on Monday!

Sent from my iPhone

On Mar 3, 2016, at 2:29 PM, Stefan Plewako <sple...@aviary.pl> wrote:

>>> Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com> w dniu 03.03.2016, o godz. 17:41:
>>>
>>> - We're going to experiment with eliminating sign-offs for locales we're
>>> already shipping. Instead, on 20 March we'll ship the last commit of each
>>> locale that meets the following criteria:
>>> - 100% complete translation with no remaining errors on the web
>>> dashboard,
>>> <https://l10n.mozilla-community.org/~flod/webstatus/?product=firefox-ios>
>>> - Have no visible l10n errors upon review of the locale's
>>> screenshots (e.g., significant truncation, text overflow,
>>> encoding, etc.).
>
>
>
>> Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com> w dniu 03.03.2016, o godz. 20:07:
>>
>> Thanks for asking. Perhaps I misphrased. To clarify, all locales that have been shipping iOS will continue to do so. We simply won't require an explicit sign-off like we've done in past versions through bugzilla. Instead, if your locale is 100% complete with no dashboard errors, we'll do a manual technical review your screenshots (or on the latest build) and accept the last repo commit.
>>
>> "Significancy" will be discussed with the l10n team in question (i.e., we'll file a bug about it). We certainly aren't experts in each language and will consult with you to learn if the translation meaning has been lost due to the truncation (making it significant) or not (making it minor).
>>
>> By doing this, we're stressing the importance of completion and testing while eliminating the (possibly) unnecesary act of signing off.
>
> What will happen when there is in fact significantly truncated translation since version 2.0 and the problem was signaled and bug was reported long before release (since it cannot be reasonably fixed on l10n side)?
>
> Does it mean that locale with such problem will be excluded from the release under new rules? (or what they actually mean really)
>

[Biraj Karmakar]

Hi Stefan,

As all locale iOS Firefox for iOS 'screenshots are now ready. But same
things I have noticed, there are problemss in screenshot. So we could set
one day per week as Firefox for iOS l10n test day. and that day we could
deal with bunch of cases with truncated text and other very obvious errors.

May be we could start this as a pilot.

If this is possible, I am happy to help with my expertise on this field.

Thank You

Biraj Karmakar
Twitter :*@birajkarmakar*

[Stefan Arentz]

Hi Biraj,

Your help is very much appreciated.

I don’t think we need a special day for L10N testing or bug fixing. I prefer a more continuous process where we will try to publish screenshots on a (semi-) daily basis. If issues are found then the best course of action is to file bugs. My team will triage those bugs and deal with them. We do that every Tuesday, but it is easy to get important bugs on our radar sooner with an email here or a ping in one of the irc channels.

S.

> On Mar 5, 2016, at 1:55 AM, Biraj Karmakar <brn...@gmail.com> wrote:
>
> Hi Stefan,
>
> As all locale iOS Firefox for iOS 'screenshots are now ready. But same things I have noticed, there are problemss in screenshot. So we could set one day per week as Firefox for iOS l10n test day. and that day we could deal with bunch of cases with truncated text and other very obvious errors.
>
> May be we could start this as a pilot.
>
> If this is possible, I am happy to help with my expertise on this field.
>
> Thank You
>
> Biraj Karmakar

> Twitter :@birajkarmakar

>
>
>
>
>
> On Fri, Mar 4, 2016 at 11:20 PM, Stefan Arentz <sar...@mozilla.com <mailto:sar...@mozilla.com>> wrote:
> We (iOS team) have not been good with dealing with l10n bugs. The screenshots make that very clear as there are a bunch of cases with truncated text and other very obvious errors.
>
> I'll find out how we can get these bugs more visibly on our radar and how to make them part of our weekly triage. It is possible that we do not see them because we do not check the right component or because they do not have the right flags. That is all easy to fix and we will.
>
> We want to ship a quality product and that means we will deal with those bugs.
>
> We have a couple of weeks of stabilization time before we submit 3.0 to the App Store. Part of that will be dealing with regressions and bugs and the l10n ones are a big part of that.
>
> More on Monday!
>
> Sent from my iPhone
>

> On Mar 3, 2016, at 2:29 PM, Stefan Plewako <sple...@aviary.pl <mailto:sple...@aviary.pl>> wrote:
>
> >>> Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com <mailto:jbe...@mozilla.com>> w dniu 03.03.2016, o godz. 17:41:

> >>>
> >>> - We're going to experiment with eliminating sign-offs for locales we're
> >>> already shipping. Instead, on 20 March we'll ship the last commit of each
> >>> locale that meets the following criteria:
> >>> - 100% complete translation with no remaining errors on the web
> >>> dashboard,

> >>> <https://l10n.mozilla-community.org/~flod/webstatus/?product=firefox-ios <https://l10n.mozilla-community.org/~flod/webstatus/?product=firefox-ios>>

> >>> - Have no visible l10n errors upon review of the locale's
> >>> screenshots (e.g., significant truncation, text overflow,
> >>> encoding, etc.).
> >
> >
> >

> >> Wiadomość napisana przez Jeff Beatty <jbe...@mozilla.com <mailto:jbe...@mozilla.com>> w dniu 03.03.2016, o godz. 20:07:

> >>
> >> Thanks for asking. Perhaps I misphrased. To clarify, all locales that have been shipping iOS will continue to do so. We simply won't require an explicit sign-off like we've done in past versions through bugzilla. Instead, if your locale is 100% complete with no dashboard errors, we'll do a manual technical review your screenshots (or on the latest build) and accept the last repo commit.
> >>
> >> "Significancy" will be discussed with the l10n team in question (i.e., we'll file a bug about it). We certainly aren't experts in each language and will consult with you to learn if the translation meaning has been lost due to the truncation (making it significant) or not (making it minor).
> >>
> >> By doing this, we're stressing the importance of completion and testing while eliminating the (possibly) unnecesary act of signing off.
> >
> > What will happen when there is in fact significantly truncated translation since version 2.0 and the problem was signaled and bug was reported long before release (since it cannot be reasonably fixed on l10n side)?
> >
> > Does it mean that locale with such problem will be excluded from the release under new rules? (or what they actually mean really)
> >
> > _______________________________________________
> > dev-l10n mailing list

> > dev-...@lists.mozilla.org <mailto:dev-...@lists.mozilla.org>
> > https://lists.mozilla.org/listinfo/dev-l10n <https://lists.mozilla.org/listinfo/dev-l10n>
> _______________________________________________
> dev-l10n mailing list
> dev-...@lists.mozilla.org <mailto:dev-...@lists.mozilla.org>
> https://lists.mozilla.org/listinfo/dev-l10n <https://lists.mozilla.org/listinfo/dev-l10n>
>

[Jeff Beatty]

Hi everyone,

Just wanted to send an update to this email.

Only *5 days* remain to localize Firefox for iOS 3.0! Please complete
localization of the latest version for your locale by *20 March*.

*L10n process changes*:
- *Testing*:

- You no longer 'need' to have an iOS device to help with testing your

locale. To find your locale's screenshots, visit

http://people.mozilla.org/~sarentz/fxios/screenshots/latest
<http://people.mozilla.org/~sarentz/fxios/screenshots/latest> . We've
already been able to address some l10n issues by testing with these
screenshots. If you'd like to follow along, the tracking bug for these
issues is here: https://bugzilla.mozilla.org/show_bug.cgi?id=1256395

- *Sign-offs for shipping locales*:

- We're eliminating sign-offs for locales we're already shipping.
Instead, on *20 March* we'll ship the last commit of each locale that

meets the following criteria:
- 100% complete translation with no remaining errors on the web
dashboard,

<https://l10n.mozilla-community.org/%7Eflod/webstatus/?product=firefox-ios>

- Have no visible l10n errors upon review of the locale's screenshots

that can be resolved through retranslation (e.g., significant truncation,
text overflow, mistranslation etc.).
- Any i18n issues affecting multiple locales should be reported in
bugzilla and listed as a depedency of bug 1256395 for tracking
and triage.
Since these issues require help from the iOS team, they will not be
blockers to shipping your locale.

If you have any questions, please be sure to ask in the next 5 days. Thank
you to all 43 out of 61 locales who have completed localization for 3.0
early!

Jeff

On Thu, Mar 3, 2016 at 9:41 AM, Jeff Beatty <jbe...@mozilla.com> wrote:

> Hi everyone,
>
> Sorry this email is a bit delayed. I'm hoping to consolidate everything
> that's been said so far (plus other details) into this one email for you to
> reference back to during this cycle. Let's begin, shall we?
>
> *String freeze*: 29 February
> *Localization freeze*: 20 March
> *Strings profile*: 31 new strings.
>
> - ~32% are for managing authentications,
> - ~19% contain additional brand/feature names and don't require
> translation (e.g., Twitter, Touch ID, etc.),
> - ~16% are about bookmark management,
> - ~16% are about private data management,
> - and the remaining 20% are for general use of the browser.
>
> *L10n process changes*:
>

> - *Testing*:
> - As Stefan mentioned, we have automated l10n screenshots for
> testing!!! You no longer 'need' to have an iOS device to help with testing
> your locale. To find your locale's screenshots, visit
> http://people.mozilla.org/~sarentz/fxios/screenshots/ . We're
> hoping we can make this iterative too so that you can see your changes soon
> after they hit the repos. Testing will be very important this cycle and in
> future cycles as described below.

> - *Sign-offs for shipping locales*:

> - We're going to experiment with eliminating sign-offs for locales
> we're already shipping. Instead, on 20 March we'll ship the last commit of
> each locale that meets the following criteria:
> - 100% complete translation with no remaining errors on the web
> dashboard,
> <https://l10n.mozilla-community.org/~flod/webstatus/?product=firefox-ios>

> - Have no visible l10n errors upon review of the locale's
> screenshots (e.g., significant truncation, text overflow, encoding, etc.).

> - *Sign-offs for new locales:*
> - We will continue to work through bugzilla for processing new locale
> sign-offs. The criteria for shipping these locales will be the same as the
> criteria mentioned above.
>
> I'm sure many of you will breathe a sigh of relief after reading that
> about sign-offs :-)
>
> As it goes with experimentation, the first time is never perfect. We're
> looking forward to working with you to refine this process. You've already
> given some great suggestions for closing the feedback loop and adding
> context to strings within our translation environments! Please continue to
> let us know where parts of the process don't make sense or need changes.
>
> As always, if you have any questions or comments, please let us know.
> Looking forward to shipping a lot of localizations on 20 March!

> --
> Thanks,
> Jeff
>



--
Thanks,
Jeff

[Artem Polivanchuk]

Hi,

Will the next screenshots update happen before Localization freeze?
Last update was on March 7th.

Thanks,
Artem

[Selim Şumlu]

15 Mart 2016 Salı 16:27:09 UTC+2 tarihinde Jeff Beatty yazdı:

Hi Jeff,

Looks like I (or we?) have been removed from the TestFlight program, and http://people.mozilla.org/~sarentz/fxios/screenshots/latest/tr/ looks empty.

Any suggestions on testing the latest build?

[Jeff Beatty]

Hi Selim,

It looks like there may have been some script failures for tr.

Stefan, could you take a look at that please?

Jeff

> _______________________________________________
> dev-l10n mailing list
> dev-...@lists.mozilla.org

[Stefan Arentz]

I’m working on more up to date screenshots today. I will also investigate with tr is missing.

S.

[Jeff Beatty]

The latest Firefox for iOS Aurora l10n build is available for testing here:
https://people.mozilla.org/iosbuilds/l10n.html

--
Thanks,
Jeff