Change "Bengali" language name to "Bangla"

[Doug Ewell]

I suppose we could just go ahead and add "Bangla" to the Registry as a
non-primary Description field, making sure to keep the ISO 639-3
reference name "Bengali" as the first Description field. This might cool
the nerves of some folks who have been much more vituperative than Mats
in their insistence that Bangla is the One True Name of this language.

I agree with Michael and John that this sort of thing, as an
ietf-languages initiative, needs to stay focused on identifying the
language (script, region, variant) and not on providing translations.
The name "Bengali" does not fail in its purpose of identifying the
language.

It's not as though ISO 639 itself has shied away from non-preferred
names in the past; see "Limburgan; Limburger; Limburgish" as an example.
Maybe the best option is to take this whole matter to the RA.

FWIW, I suggested adding "better-known" aliases for region descriptions
back in 2011, such as adding "North Korea" to supplement the existing
"Democratic People's Republic of Korea", but the list discussion
spiraled out of control quickly and nothing productive came of it.

Folks who are interested in Description fields really should read
Section 3.1.5 of RFC 5646 first.

--
Doug Ewell | http://ewellic.org | Thornton, CO 🇺🇸