Re: [apps-discuss] Last Call: <draft-ietf-appsawg-rfc3536bis-02.txt> (Terminology Used in Internationalization in the IETF) to BCP

[Frank Ellermann <[e-mail]>]

On 17 June 2011 05:21, Mykyta Yevstifeyev wrote:

> I proposed improving the "control character" definition

Let's keep it simple.  ECMA 48 5th edition claims that its
4th edition corresponds to ISO 6429:1992, and that there
should be a new ISO 6429 based on ECMA 48 5th ed.

If that actually happened your proposal to add a non-free
ISO 6429:1992 reference could be arguably obsolete.

The control codes only mean something in the context
of specifications using it, e.g., TUS doesn't use most of
the C0 or C1 control codes (adding its own oddities).

Wrt i18n ECMA 48 5th ed. could be even interesting, it
covers BiDi.   But I'd never trust in any BiDi spec. unless
it was written by one of the usual suspects (Martin, John,
Harald, Paul, etc.)

> we're trying to give the terms normative meaning within
> IETF, since the intended status is BCP

The business with the C0 + C1 terminology is IMO correct
in the draft, and explaining what if anything specifc control
codes do is out of scope.   The "net UTF-8" BCP obsoletes
most of these codes, and the i18n terminology draft has a
pointer to the "net UTF-8 BCP" (IIRC).

 > The document makes normative reference to an obsolete
> document - ISO/IEC 10646:2003 whereas ISO/IEC
> 10646:2011 is published.  The reference should be
> corrected.

Above all it should be the ISO 10646 version corresponding
to the referenced TUS version, if you checked that I'd agree.

-Frank