Reshat Sabiq's requests for two Tatar orthographic variants

"Reshat Sabiq (Reşat)" tatar.iqtelif.i18n at
Wed Jan 10 07:19:05 CET 2007

Hash: SHA1

CE Whitehead yazmış:
> I can second your request for
> "two Tatar orthographic variants"
> "iqtelif and janalif"
> if that is just what you want, and
> if I know whether these are variants of the script or the language.
> I think they are of the former category (script).
I think proper lingo would call the orthography variants, with the
script being Latn.

> So is janalif going to just refer to Qazan Tatar orthography of the late
> 1920s and 1930s?  
I think that's the right scope for janalif, if we name it that. In fact,
janalif is identical is letter set to the NTA, except that the same
vowel is mapped to different letters in Qazan Tatar vs. some other
languages. Still, NTA would identify any of those alphabets w/ enough
accuracy, since it's paired w/ a language.

> You should decide yourself what is most appropriate.
> That is:
> If the tag is for a script variant,
> * is it or is it not specific to Tatar?
> * or is it for the moment specific to Tatar,
> * or is IETF is to decide whether it is specific to Tatar?
> I think that should be your call not ietf's!
> I would hope you would decide whether the second requested tag to be
> "ussrlatn" or "janalif" (applying to tt only? is that what it will be
> the script for?  I gather yes; that's good and specific!! otherwise if
> it has a wider application we need to know the tags of all other
> languages and variants it will apply to; we can always add more of those
> too if you do not get them all now).
I tend to favor NTA for all Turkic languages, and the list can decide
whether or not non-Turkic languages should get a variant name, and what
it could be. I'm having trouble thinking of a good variant name for 70
languages. janalif is my fallback if for some reason you turn down NTA,
or whatever we call it.
> I'll be happy with either ussrlatn or janalif but want to make sure
> exactly what it will refer to.  So I can support either in this case,
> but want you to list the variants it will apply to.
NTA, for starters is applicable to what i listed in the original request:
az, ba, crh, kk, krc, ky, sah, tk, tt, uz
Others could be added later, as needed.
> "per-language (janalif for tt, canalip for kk, year for others)"
> I cannot vote for using the year for unspecified variants.
Well, they are specified as in unified alphabet latinization campaign of
1930s, so in the worst case scenario non-Turkic languages could get a
year-only variant.
> Now we have more requests.
> So you've left me confused; here I am supporting the two initial
> requests but I see that one of the variant subtags (janalif or ussrlatn
> or ussr 1928) is supposed now to be for all 70 languages.  This looks
> like a change to your original request, in that now you are asking it to
> apply to all languages;
> I guess then list the tags it applies to; I will support it then.
As Frank pointed out, it could be scoped as follows:
tt-Latn-october-NTA (or even, tt-Latn-october-NTA-janalif, which is
So i'm not asking for diff. things, but the question is what is the
right scope.
> And I hope you will decide whether you want me to support option 1, 2,
> or 3, or a combination because I cannot decide which is best; I am not a
> speaker of these languages.
> Thanks for understanding.  So don't inundate me with options, decide
> which is the best; I am sympathetic to your request but want you to tell
> me what option to support.  (Maybe other members of the group feel
> otherwise I don't know but that is how I feel at least!  And as you see,
> I've just sent the options back to you to choose from.)
I think nesting as above is excessive.
I am inclined to the following for Turkic and non-Turkic, respectively:
a. tt-Latn-NTA
b. tg-Latn-year, as in tg-Latn-1928, unless you folks think of a variant
name that could be used instead of year, maybe smt. like tg-Latn-soviet,
but anything i can think of for this case sounds or could sound
"loaded", even if it's not the intention, as some folks mentioned. And
of course, i'm not particularly interested in this case, other than in
how it might affect the variant for Turkic languages, as it was a mass
reform for everybody.

P.S. JFTR, last time i called latinization of 1930s the worst
orthography reform in the history of mankind, but i think it's really
the 2nd worst, the worst being the one in 1939.


- --
My public GPG key (ID 0x262839AF) is at:
Version: GnuPG v1.4.2.1 (Cygwin)


More information about the Ietf-languages mailing list