From:  "Resat SABIQ" <>
Date:  31 Jul 2017 06:49:45 Hong Kong Time

Re: Mozilla Add-ons: QIRIMTATARCA Til Destesi (Crimean Tatar Lang Pack) 54.0 didn't pass review


=== <Operational> part of the message & the issue: ===
This is what oviously needs to happen:
1. The addon's history that got recklessly wiped out (apparently using a thoughtless SQL statement) in April needs to be restored, & AMO needs to reflect that this extension has been around since 2009.
2. The extension is fully functional in accordance with its documented promise (to provide a Crimean Tatar localization of UI, with easy installation, & looking like a Turkish langpack from outside), & should be allowed to be used as long as all this is feasible. There's no justification to stop it all right now.
3. You insist on users using a langpack that advertises around that they use crh(-UA) locale in their browser[1][2]? Let me check in my 8 years of work, & then you can release it or not anyway you want.
Spend 1/3 of the time you intend on spending on replying, & the rest of the time on useful stuff, like addressing the points above (& later on addressing other Firefox issues like [3]). Anything else, is just a waste of everybody's time.

It's disapponting to see that even top-tier people are sending emails w/o sufficient analysis, apparently w/o even having looked at the extension (or its source). As for
a. after 8 years of working on this, i'm obviously aware of it
b. it obviously doesn't do anything about increased fingerprintability i mentioned in points 2., 3., & 4. (those points reveal navigator.language, which doesn't affect).
Several people here seem intent on picking up a fight with me, & i don't have time for any of this. These several people, intentionally or not, have also placed themselves in a horrible company along with guys like Ilminsky, Stalin, Putin, etc. I urge you to change course ASAP, & respect 8 years of work in Mozilla environment, & act in accordance with open source philosophy. 83+% (the rest in a kindred lang) localization would be gladly accepted by any open-source project i know of, especially after 8+ years of work. Instead, you have vandalized my l10n work history, stopped the extension from being used, & so far have not shown any sign of willingness to accept it into VCS! Think about all of this before you write another word, & make me waste more time arguing instead of spending that time that i can barely find on useful things (like continuing that same l10n effort).
[1] all of the feedback so far amounts to this.
[2] i think it's dumb (& perhaps even outrageous) to insist & require that people have to be highly fingerprintable, but again i don't have much time to argue.
Envoyé: lundi 24 juillet 2017 à 13:14
De: "Axel Hecht" <>
À: "Resat SABIQ" <>,,,,
Objet: Re: Mozilla Add-ons: QIRIMTATARCA Til Destesi (Crimean Tatar Lang Pack) 54.0 didn't pass review
Am 22.07.17 um 14:53 schrieb Resat SABIQ:
> Including
> *Envoyé:* samedi 22 juillet 2017 à 14:47
> *De:* "Resat SABIQ" <>
> *À:*,
> *Cc:*
> *Objet:* Re: Mozilla Add-ons: QIRIMTATARCA Til Destesi (Crimean Tatar
> Lang Pack) 54.0 didn't pass review
> Hello,
> First of all, the extension vs. langpack matter for this addon has been
> discussed multiple times, & Jorge kindly allowed it to be an extension
> several times.
> Primary reasons for it to be an extension:
> 1. It's easier to install and use: users don't need another addon or
> knowledge of how to set about:config preferences (& extra time) to do
> so.  With a lang pack, it won't change, and user would be required to
> manually change a pref in about:config: that's way too much to ask, even
> for me. User base is small as it is. And making it harder to use might
> easily zero it out (well, you have actually zeroed it out currently as
> it is).

I disagree that this code should be in add-ons, and the ability to do
this will go away with 57, as we drop support for xul-addons, and with
that, preference changes.

> 2. Some Firefox requests also reveal the UI locale[1], and this addon
> sets it to tr, making it better for users' privacy.

Again something that we'll not expose to web extensions, AFAIK. Note,
there are existing bugs on reviewing the privacy impact of
localization-related APIs. Note, I'd r- for an extensions that tries to
modify Firefox in a way that it lies about its data to Mozilla.

> 3. Required for Help->Firefox Help menu element to lead to a page in a
> kindred language.  Otherwise, this results in a page being
> displayed in the original locale (e.g., English), because the
> infrastructure for this locale isn't there yet (& might not be there for
> a long time)).

I disagree that this should be part of a language pack. AMO should
figure out that you can read content in Turkish, and offer that.

> 4. Required for Report Help->Web Forgery... menu element to lead to a
> page in a kindred language. Otherwise, this results in a page
> being displayed in the original locale (e.g., English), because the
> infrastructure for this locale isn't there yet (& might not be there for
> a long time)).

Another thing that shouldn't be done by a language pack, IMO.

Given that most of the things won't work in Firefox 57 anyway, I
recommend against continuing the path taken in the past.

As for finger-printing concerns, I recommend copying the language
settings from the turkish localization,,
and see how far that goes.


> Another important factor:
> a. I'm hoping to get the locale to be released as part of official
> Mozilla builds in the near future. When the locale is available from
> it might be less important
> for this addon to be easy to use, cause the addon page could just refer
> people to the official releases (or the addon could be de-listed
> altogether). However, there is 1 caveat even in that case: the UI locale
> would again be revealed in this case as in [1], & i would still like it
> to be masked as tr even in official releases for privacy reasons. I hope
> that is possible, but i'm not 100% sure.
> For reasons 1., 2., & a. above, i'm kindly requesting the addon to be
> allowed to be an extension, at least until: the locale is available as
> an official build, & we know whether the official build can use a
> masking locale (like tr), instead of a very finger-printable locale like
> crh(-UA).
> [1] E.g.:
>      [exec] RequestAssistant (2017-04-03 04:36:20):
> [2] Copied (w/ XP changed to 10) from prior discussions: *But i still
> believe addons
> like these shouldn't be held equivalent to official langpacks released
> by mozilla. For instance, i also have a Blend In addon, that makes app's
> platform look like Windows 10 even if in fact it's not: of course an
> official mozilla release will (most likely) never do that, but there
> should be some room for these kind of documented special features in
> developer-supplied addons.*
> [3] The langpack has been around since 2009 (v3.5). In v3.5 57.26% was
> translated, the /rest /was "fuzzy"; by v53 83.28% is translated, the
> /rest /is "fuzzy". AMO is now giving me run-arounds, & some time in
> April Andreas Wagner marked all versions disabled. Then i made some
> (completely unnecessary) changes he wanted, & now only v53 is shown. Now
> it's like it hasn't existed for 8 years, & was released only for v53. Is
> that right?
> [4] I have also offered to check the l10n for crh(-UA) into the VCS, but
> was given a run-around on that route as well. If it's checked in, you
> will be able to release or not release it any way you want (lang pack,
> would be the first option that comes to mind). But this extension has a
> right to not be a standard lang pack: we in an open-source environment
> here, not a nazi-source environment. And it works just fine now, in v54
> that got blocked for instance: there are no issues, & everything is
> documented (it has every right to exist and be listed as an addon).
> [5] So i now i feel like i've been given completely made-up run-arounds
> on both the AMO end, and the VCS end. I have barely been able to keep up
> with updates as it is. I might not have much more time for all these
> made-up roadblocks. Get your act together please. If you need to, have
> an official Mozilla Foundation meeting, & make a decision or 2. Give
> this locale a chance to continue living while it's still alive. Don't be
> like (some disgraced) people in history and now who've been doing their
> best to kill languages.
> Regards,
> Reşat.
> *Envoyé:* vendredi 21 juillet 2017 à 13:49
> *De:*
> *À:*
> *Objet:* Mozilla Add-ons: QIRIMTATARCA Til Destesi (Crimean Tatar Lang
> Pack) 54.0 didn't pass review
> Hello,
> Your add-on, QIRIMTATARCA Til Destesi (Crimean Tatar Lang Pack) 54.0,
> has been reviewed and did not meet the criteria for being hosted in our
> gallery.
> Reviewer:
> Philipp Kewisch
> Comments:
> This version didn&#39;t pass review because of the following problems:
> 1) Please change this to a language pack as requested in the previous
> review.
> This version of your add-on has been disabled. You may re-request review
> by addressing the reviewer's comments and uploading a new version at
> If you want to respond please reply to this email or visit
> .
> You can also join us in #addon-reviewers on
> To learn more about the review process, please visit
> To ensure your add-ons are compatible past Firefox 57, please migrate
> them to WebExtensions as soon as possible:
> --
> Mozilla Add-ons Team
> ===================================================================
> You received this email because you're subscribed to:
> * email me when my add-on is reviewed by a reviewer
> You can't unsubscribe from this type of email.
> Manage email notifications:

dev-l10n-new-locales mailing list