I have often heard variants of "You should consider unifying the language" (the polite version) or, "What were you thinking when you split the language?" (the not so polite one).

The language in context is Bengali. And that it exists as Bengali (India) and Bengali (Bangladesh) ie. bn-IN and bn-BD

Often I do not have enough time/patience/motivation to explain in detail. This is one such attempt to write up the background and attempt to explain things a bit.

The somewhat obvious response is that we did not split the language and it is not on us to unify it. If the audience is reasonably well versed in language technologies, at this point in time one brings up the definition and scope of localization (l10n) and internationalization (i18n). Or, perhaps munge everything together into the current fashion of globalization (g11n). Either way, http://www.w3.org/International/questions/qa-i18n.en is where it all ends up.

However, if the audience is not very technically proficient and the question tends to veer towards the rhetorical, the correct place from where to begin this conversation is http://en.wikipedia.org/wiki/Ankur_Group (that page needs some edits; that is a different story)

"In a world" where language computing and language enablement on Linux was motivating large groups of participants in contributing to Free and Open Source Software, the Ankur Bangla Project started as a deeply collaborative efforts of individuals who traced their roots to either side of the geographical border and sometimes were located at places far away from those they called home.

Over a period of time the leaders of the group arrived at the conclusion that the focus areas were diverging and there was an opportunity to enable every idea towards success. A substantial volume of conversations and discussions eventually with two distinct groups - one which was comprised of contributors who had roots at Bangladesh and another who were from West Bengal and other places. Personally, the entire decision to split out into specific focus areas and do the best of what one could do is one I took many notes from. The leaders were emotionally invested, deeply technical, tremendously passionate about the end objective and yet, completely pragmatic when arriving at this step. I have, since then, yet to see an equivalent display of disciplined reason and emotion applied to an issue that is by its very nature tremendously volatile.

Those who were not yet part of the efforts or, were peripherally involved are perhaps unaware of the context. But there it is. As much as the issues of "each group would need to re-invest to build community", "additive power will be lost" were discussed, it was also reasoned that if there is a sustained effort towards enabling language computing (especially for Bengali) on the digital medium, there needs to be a diverse and deep range of approaches which need to be adopted.

In short, no one split the language. The locale definition was created. Take a look at http://www.unicode.org/cldr/charts/latest/summary/bn.html to have a good understanding of what it means (and along side, also look at http://www.unicode.org/cldr/charts/latest/summary/zh.html or, http://www.unicode.org/cldr/charts/latest/summary/en.html because these are non-obvious examples that are not often thought about) An additional/new locale does in no way result in reduced contributions to upstream projects. A trendline of reduced contributions across large scale upstream projects happen if we [1] do not sustain outreach in a structured manner [2] do not handle new contributors in a welcoming way [3] cannot emphasize on the value of contributions [4] have not defined an annual goal/objective

There was a point in time when a new contributor worked on translations for Gnome Terminal and during a release cycle, downloaded the binary, clicked on the Credits and was pleased to bits (https://git.gnome.org/browse/gnome-terminal/commit/?id=14aa74c1a998abea17e93dea082dcf9860674628) . If we are no longer able to bring that sense of joy and happiness to our contributors - then no amount of "language unsplitting" would help us get to where we want to be.
Shared publiclyView activity