Re: ISO/IEC 10646 as Document Character Set

Albert Lunde (Albert-Lunde@nwu.edu)
Thu, 4 May 95 14:21:18 EDT

At 1:10 PM 5/4/95, connolly@w3.org wrote:
>
>This is not a democracy :-) We go by technical arguments, not by
>shouting. I don't see why we need to put ISO10646 as the document
>character set in HTML 2.0. Everybody can do everything they need to
>do -- and reliably -- even if the 2.0 RFC only specifies ISO-8859-1
>
>The current draft allows user agents to support others, and even
>suggest that they support ISO10646:
>
>http://www.w3.org/hypertext/WWW/MarkUp/html-spec/html-spec_2.html#SEC8

I'd suggest leaving ISO-8859-1 as the document character set in 2.0, and
saying in an appendex that we intend to address internationalization in a
forthcoming document which will specify ISO10646 as the document character
set regardless of the MIME charset parameter. This can warn developers
about the problem of introducing incompatible numeric character references.
(We don't want them to use the MIME charset "as" the document character set
in cases that are not subsets of Unicode.)

With regard to the question of determining the MIME charset from internal
information vs HTTP, I think we should rely mainly on the MIME charset
mechanism, but I wouldn't be totally adverse to adopting some designated
hack using something like <META> if it was clearly indicated as an optional
transition mechanism while full charset support is being phased in.

I think for the general question of meta-information on disk we should
adopt a seperate advisory document advocating a specific disk format for
including MIME-like meta information for text and/or binary objects. Use of
this format should not be required, but suggested as an option for
implementations. The more we can lean on MIME the happier I'll be ;)

Generally, I'd like to get 2.0 out the door with Latin-1 and warn people
against incompatible practices that would conflict with our plans for 2.x
internationalization.

---
    Albert Lunde                      Albert-Lunde@nwu.edu