Re: ECMA-cyrillic alias iso-ir-111 sore

Keld =?iso-8859-1?Q?J=F8rn?= Simonsen <keld@dkuug.dk> wrote:

> My understanding was that you wanted KOI8-E (E for error?)
> to be the specification in RFC 1345?

NO, NO, NO! I was saying it's a better name for the charset designed by ECMA in
the early 1980s and registered with ISO-IR under No. 111 because the charset in
question is a variant of the beloved old Soviet KOI-8 (whereas what RFC 1345
lists is not KOI-8 but a cross between ISO_8859-5:1988 and windows-1251).

E stands for ECMA, European, or extended, see this superb page on Cyrillic
charsets:

http://czyborra.com/charsets/cyrillic.html

> I think it would be helpful to
> list which names would be refering to which specifications after the
> renaming you propose would have been done.

I'm not proposing a renaming, at least as I see it. My take is that since the
ISO-IR registry has always been the normative reference, the correct
interpretation of charset names ECMA-cyrillic and iso-ir-111 has always been
the KOI-8 variant one even with the misleading RFC 1345, it's just that
implementors had practically no way of knowing this. So in my proposal 1 I'm
asking IANA not to change what these charset names mean, but to clarify in the
character-sets document what the correct charset is and has always been, i.e.,
the KOI-8 variant defined in the ISO-IR registration document. And since it IS
a KOI-8 variant, my proposal 2 is to give it a descriptive name as such as an
alias.

BTW, I take the fact that the charset tables on your FTP site render ECMA-
cyrillic alias iso-ir-111 as the KOI-8 variant rather than the RFC 1345 charset
(and I suppose it was you who fixed it in GNU recode 3.5) as a further
indication that despite the RFC 1345 error the names ECMA-cyrillic and
iso-ir-111 always properly meant the correct KOI-8 variant charset really. So I
just want the record straight everywhere, consistent between ISO-IR, your FTP
site, and IANA character-sets.

MS

Received on Tuesday, 8 April 2003 09:04:18 UTC