Michael Everson everson at
Fri Apr 11 22:17:03 CEST 2003

At 12:09 -0700 2003-04-11, Mark Davis wrote:

>I have no idea whether there are orthographic difference between 
>Azeri written in Iran has major/minor/subtle/nuanced differences 
>from Azeri written in Azerbaijan. And neither do you.

Of course I do. Azeri written in Iran is written in Arabic.

>1. The assumption that the above (elided list) are not encoded. That's
>wrong; they are *already* valid RFC 3066 codes. You say "To do so is
>idiotic"; well, RFC 3066 already has this entire list. Is that idiotic???

Individual registrations do not exist for them.

>2. The assumption that software requirements should have no influence on RFC
>3066 registrations. If that is the position of the IETF, Harald can let me
>know right now, and I won't bother pursuing this issue.

Martin has asked for specifics about the software "requirements".

My view is that 639 and SIL and 3066 provide codes that refer to 
languages, not to locale bundles. That is what they do. That is what 
they are there to do. If locale bundlers want to use codes, they 
should. But they should not force us to register things that don't 

>  > order to make LINGUISTIC distinctions, not to perform kludges for
>  > Bull. It takes time and effort on my part, on the part of IANA, and
>  > on the part of all the people who AREN'T Microsoft and whatever
>  > "others" you refer to to weed through the mass of duplicate
>  > registrations in order to determine what the unique entities are.
>"unique entities"???

Yes, since you are proposing to duplicate two codes for Azeri which 
refer to the same thing.

>Speaking of bull, those cows and horses are *LONG* out of the barn.

Cows? Horses? I am not sure what the referents are here.

>That is the point that the long list makes. Every item in that long 
>list is a valid RFC 3066 code,

They are structurally well-formed. That doesn't mean they refer to 
anything different one from t'other.

>but nobody knows whether there are "real" orthographic differences 
>between all of them or not. (And what are the solid criteria for 
>such a judgment anyway? Scholars might see a difference where the 
>layman doesn't.)

Differences in spelling, I should think.

>Sadly, it is not our choice. And, sadly, I have no magic wand to make
>Microsoft realize that no matter what its customers say, there is no need to
>make these distinctions.

I said, Mark, that it is not for language codes to make all of these 
distinctions. I never said that distinctions of various kinds did not 
need to be made.

You still have not told me whether you know if az-Latn and az-Latn-AZ 
differ in any way. If they do not differ, then the codes are 
duplicates for the SAME THING.

>In examples like I gave earlier on this list (the
>"Theatre Centre..." example); these are NOT locale distinctions, these are
>language distinctions, by any practical measure.

I am sorry, I did not understand this example, and I cannot find it 
in 100+ messages all with the same subject line. (Really, I am sorry.)

>(I do not confuse these: differing in timezone does not, for 
>example, qualify for a different language ID, although it may 
>qualify for a different locale ID.) If you want to ignore real 
>distinctions that people make out in the world because of some 
>ivory-tower notions of purism, then as I say, I hope that Harald 
>lets me know soon, so I can do something sensible with my time 
>instead of continuing this.

Answer Martin's question. I am not your enemy here, Mark, though I am 
being, intentionally, very very critical, so that we are all on board 
about this process and its result. And yes, I will insist that a 
language tag tags a language and not some other entity. That isn't 
ivory tower, it's just fact.
Michael Everson * * Everson Typography *  *

More information about the Ietf-languages mailing list