Skip to main content

Unifying Bengali Locales for Mozilla Project

UPDATE: as intended, this has started some good amount of stir - which was half the point of this post. Many questions are being asked in several channels which has the risk of derailing the actual discussion (unifying bn) if answered. So, if there's anything particularly about this blog that bothers you, it'll be best to express it as a comment to the blog-post itself.

I'm updating the post with some side-notes (floating right textboxes, in light-yellowish background) & tooltip titles (things deserving less explanations; underlined texts with yellowish background). Apart from internal markup changes, the existing content of the post will not be changed (unless otherwise mentioned); feel free to use DIFF.

Post was removed by the author himself (under good faith, to help the situation).
Can't be cited anymore.
This post is in response to the recent adversities within Mozilla Kolkata Community, regarding the Bengali Localization for Mozilla Projects in general (and Firefox for Android in particular), quality of bn-IN localization & prospects of unified bn locale.

As this thread [snapshot] goes, most of the participating members have paramount knowledge of, experience with, contacts in, contexts with & empathy for L10N efforts, yet don't identify themselves with the bn-IN community (I fail to find many of them on mozillians.org as well). In general they do think there's a lack of mentorship; and I think they are totally in a position to fill the gap, if they wanted.

It's inappropriate to consider me even remotely capable & responsible for L10N bn-IN activities. It's sad that even when provoked blatantly (say, through this post), nobody has stepped in to tell me, "Enough of your jackassery. You're not the senior-most/capable/apt person to address this issue. Let me show you how it's done!"

It was posted publicly on FB, so posted the screenshot without taking explicit permission from Indranil da.
Why do I have to respond? Because unfortunately, I'm probably the senior-most active member, of the Bengali speaking Indian Mozillians, who proudly associates himself with the community. I'll glad to be wrong here on the senior-most part.

Let me start off by offending some people.

I'm not an active member of localization efforts (of any FOSS project). There are three primary reasons for that:
  1. Apart from the professional localizers, the major contributor base of l10n are the volunteers who are not there because they're great at that; they're there because, they're not great with anything else, or, even localization. So to say.
  2. The l10n community is very slow moving on the technology front. So the tools with valid, versatile & kick ass USPs also take hell lot of time to be accepted/used by the communities. So, tooling-contributions is also not a very interesting ground.
  3. Most importantly, I don't support localization; because, Genesis 11. I don't approve of the need of multiple languages, at a high level. But, that's an argument of another day.
CONTEXT [cite deleted]: The Mozilla Kolkata community members were asked (on an FB post) about justifying the bn-IN Fennec launch event, where the L10N contents were allegedly copied from bn-BD, and were suggested to look into the prospect of merging these two locale. The thread went haywire.
Hence, even though my language curiosity ranges from Minion-tongue to Blackspeech - I'll keep my post restricted to the disaster-recovery of the situation in hand - and not get into the technical tidbits of the l10n system itself.

Now that I'm done offending & driving off the unreasonable l10n idealists, and down with a leaner audience, let me split the discussion in four sections:

Communication Gap

In FOSS we have some well accustomed tools for different purposes. Full-duplex discussion goes in IRC, half-duplex discussions, arguments & announcements go into mailing-lists, PR & Promotional contents go to social channels, poop of mind & passive-aggressive attacks go to the personal blog, and semi-spam, semi-infotainment kittens & fail videos go to FB groups. Although, this is "not a directive, more of a guide;" but hey!

Choosing the wrong medium/channel of communication is dragging us down & we need to be careful right now (like, yesterday would've been better kind of right now). Starting the same discussion as a mailing list thread would've had much stronger base, better output & much less juvenile name-calling, feather-tearing pile of mess by the prepubescent imbeciles on a dope.

That's not the only form of miss-communication though; the age (generation gap) is adding more to it. If anything that can help this factor, is by careful moderation & mentoring by the senior members.

Also, please care for the sanctity of information (and hereby, that phrase is a real-thing). Like each and every other action, once a post/comment/status goes live online, it creates a butterfly effect - sometimes not in a very desirable way. But, deleting that doesn't solve the problem; if anything, it adds to the confusion, broken citation & possible misinterpretations.

Meritocracy & Credibility

Even in the previous FB thread, there are several topics discussed & I find a huge part of it highly agreeable (though on an unusual medium for that discussion). The community consisting of kids & needing to grow up, the need of proper mentorship, not disdaining contributors easily to let community grow, futility of schadenfreude, generations repeating mistakes & reinventing same solutions etc.

No one can be blamed for not taking up the responsibilities of mentoring the new generation; as this is completely a volunteer effort & time is precious. They can only be persuaded. :)
"I'm not on that boat, and deny all responsibilities, but am up for a tea-time debate" isn't something that can help here. If you care for your language, have good enough exposure on FOSS/l10n, have good insights about the system in place & know well about community building - DO. STEP. IN.

We can't keep ourselves out of touch, not do our part of mentoring, let the kids run around to get shit done, and then complain how it wasn't done properly. Humans are diverse, unorganized species, prone to make mistakes at best - turning mailing lists into event-feed reader & facebook into battleground, among other things. On top of that, Bengalis are one helluva notorious & difficult race, with handle-with-care & this-side-up labels on every butt-cheeks - but deal with it. Tell us the story of how you tried & failed, instead of saying how everything is a goner & can't be helped.

Apart from that, let's all agree here that Mozilla is a Meritocratic organization, and we practice the same (as much as possible) in Mozilla India as well. Hence, not everyone's every comment carries the same value/worth, neither is reputation/credibility an one way street. It's analogous to share-market index at best.

Hence, not every piece of information to be treated with similar importance/passion. Feel free to ignore when an argument or a person doesn't make much sense anymore. Disengage. And if it's not an easy topic to just be dropped, then find a better medium, channel or concerned people to re-initiate the discussion.

bn-BD vs. bn-IN

Le wild shrewd observation: bn-IN is good-to-have, but bn-BD is a necessity.

Not only that Bangladesh officially runs on that language, an incredibly huge part of the country don't know any other language but Bengali. If there has to be any software, services, products, communications, promotions - everything _has to_ be available in Bengali, and qualitatively, it _has to_ be good.

Trying to indicate/prove "we" care/know/use Bengali better than "them" is not only ridiculous, or juvenile, it's blasphemous, baseless & insulting to some fair extent. "They" are the folks who gave their lives for the love of Language/Mother-tongue in past, and "they" are the folks who are giving their time, commitment & academic GPA to integrate Indic language support in Firefox OS in present.

Anybody who'd like to object on this note, back yourself up with data, not prejudice or herd-conformity.

Prospect of Unification

In a FOSS community, if a project looks up on another similar one & takes a learning or two, that should be appreciated, and not frowned upon. I am unaware if any or how much of bn-BD localization effort got copied over to bn-IN - automated scripts give numbers of matching strings in both locales, but one can never be sure how much was parallel effort, and how much was directly copied.

But, in either of the cases, when we're talking about Software interface translation - usually each string being an one/two word phrase - how much difference does bn-IN make compared to bn-BD and vice versa? Cause, either of the redundant parallel effort or (alleged) plagiarism is a huge waste of time, process, commitment & can't/shouldn't be the way to go.

Guys who are working on bn-IN aren't the one who are much aware of the backlog, or have been the key people to split Bengali in the first place. They're just doing it because this is how it has always been done, based on the history as much as they can remember & some misplaced prejudice of language (locale) supremacy. But, does it have to be?

That was (somewhat) the question raised, in a Facebook thread (turned into a puddle) where bn-IN was celebrating the Firefox Mobile in bn-IN release.

A very important & critical question indeed: do we have to go on posing some unnecessary LOC over the language just for the heck of it, or, we can do better and unify our efforts?

As much as I understand, translation/localization of Wikipedia's articles/contents in generic Bengali is wildly more difficult than a quick phrase on a browser-interface button. From that perspective, the challenge with the unification of bn-IN & bn-BD into bn is only as much as the combined unwillingness of bn-IN & bn-BD localizers to work together.

Find a better medium/channel of communication. Take initiatives, and participate (thanks, if you already have). Be proactive to identify & fix issues.

Mozilla BD Mailing List: https://lists.mozilla.org/listinfo/community-bangladesh
Mozilla India Mailing List: https://lists.mozilla.org/listinfo/community-india
Localization Mailing List: https://lists.mozilla.org/listinfo/dev-l10n

P.S: Here's to the 100mins of my life, that I never aspired to spend writing this blog. Cheers!

Comments

  1. This was a fantastic article. Really loved reading your we blog post. The information was very informative and helpful...

    ReplyDelete

Post a Comment