Re: Language environments

[Paul Eggert]

> Date: Wed, 21 Nov 2001 18:26:24 -0700 (MST)
> From: Richard Stallman <address@hidden>
> 
> The name was artificial, but the language is real.

The language is real, but unfortunately all its names are artificial....


> The language is now often called Bosnian/Croatian/Serbian.

Those three names are mostly political ones.  It's no coincidence that
those names match those of the three nations in the area.

If you really want to delve into these matters and add support for it
in Emacs, then here's a bit of background about this messy situation.
The spoken language (as opposed to all the political guff) is divided
into three major dialects as follows (pardon my lack of accents):

  Cakavian (northern Croatia)
  Kajkavian (near Zagreb)
  Stokavian (most other locations)

Stokavian is further divided into:

  Ekavian (most of Serbia; sometimes called "Serbian")
  Ijekavian (a.k.a Jekavian) (west Serbia, Montenegro,
    Bosnia & Hercegovina, Croatia; sometimes called "Croatian")
  Ikavian (Dalmatian coast)

Note that the linguistic reality is quite different from the political
one.  For example, many Croatians consider Cakavian and Kajkavian to
be dialects of "Croatian", even though this is linguistically dubious.

However, from Emacs's point of view, written communication is more
important.  Here, perhaps the most important linguistic variants are:

  Cyrillic Ekavian -- specifically, the Ekavian of the Belgrade dialect.
  Latinic Ekavian.  (These two variants are both official in Serbia.)

  Cyrillic Ijekavian as used by many Bosnian Serbs and Montenegrins.
  Latinic Ijekavian, used for publications in Croatia,
    by Bosnian Croats, and by Montenegrins.  Bosnian Muslims also
    use it but I think they have their own spelling preferences.

Almost certainly it would be controversial to limit Emacs to these
four choices.

I should mention that, ISO/FDIS 639-1, which has been submitted for
final editing as a FDIS, has four choices: Serbo-Croatian (sh),
Serbian (sr), Croatian (hr), and Bosnian (bs).  Please see:
<http://www.rtt.org/ISO/TC37/SC2/WG1/639/>.  Unfortunately, as you can
see, these four choices don't map well into written reality.

Is this complicated enough for you?  It is for me.  I don't envy the
poor guy who would have to decide how to shoehorn all this stuff into
Emacs and keep all the Balkan users happy.

(I can't resist mentioning that in Bosnia, at least one major
newspaper prints alternate pages in Cyrillic and Latinic; perhaps
Emacs should be programmed to switch orthography after each formfeed?
:-)