Re: Official ISO 639 changes

From: Martin J. Duerst (
Date: Wed Apr 26 2000 - 04:11:14 EDT

At 00/04/25 04:13 +0200, Keld J$BS(Bn Simonsen wrote:
On Tue, Apr 25, 2000 at 10:11:23AM +0900, Martin J. Duerst wrote:

> > Some suggestions/questions:
> >
> > - Please make sure that this page is served indicating
> > iso-8895-1 as its encoding, according to
> >
> Or just RFC 2070.

Keld, while I'm a proud coauthor of said RFC, I don't
know why you are proposing it. As of just very recently,
the IESG has reclassified it as historic. See the
excerpt below. Also, nothing much really has changed from
RFC 2070 to HTML 4.0(1), except that now everything
is together, i18n is no longer something special.

> > - Please use <th>...</th> instead of <td><b></b></td>
> > for the header cells in the table.
> Why? I think it is better to use the HTML already done as more
> browsers would understand it and it has the desired effect.

I haven't checked all the browsers, but <th> is *really* old.
I would say "or just RFC 1942", if that weren't already
historical, too.
If you can point me to a browser that doesn't use <th>,
that would be helpful.

> > - Please use <span class='changed'> or some such instead
> > of just <I> for the changed pieces. Use a stylesheet
> > to make it display the way you want.
> This is also prune to not being honoured by a number of browsers.
> Better be conservative here, so as many people as possible can
> read our specifications.

Using stylesheets among else makes it much easier for
blind people to read the pages. And using more semantic
markup makes it easier to change the presentation.

> > - New definitions of 2-letter codes make it more difficult
> > to define what code should be used in ietf-based protocols.
> > Originally, the idea for the update to RFC 1766 was to use
> > the two-letter code where this already existed, but the
> > three-letter code for the other languages. This will need
> > some more thought, and probably a cutoff date or a precedence
> > rule. The table would provide enough information for that.
> I think we should be careful with such rules, viz. the recent
> discussion on Lithuanian characters in 10646.

Of course we should be careful. But the Internet will work
much better if it's always clear which language tag to chose.

Regards, Martin.

 From the IESG:

The IESG has approved the Internet-Draft 'The 'text/html' Media Type'
<draft-connolly-text-html-02.txt> as an Informational RFC. This
document supersedes the current MIME registration of the text/html
content-type, which references a now-obsolete version of the HTML

This has been reviewed in the IETF but is not the product of an IETF
Working Group. The IESG contact persons are Keith Moore and Patrik

In the same action, the IESG reclassified the following RFCs
as Historic:

RFC1866 Hypertext Markup Language - 2.0
RFC1867 Form-based File Upload in HTML
RFC1942 HTML Tables
RFC1980 A Proposed Extension to HTML: Client-Side Image Maps
RFC2070 Internationalization of the Hypertext Markup Language

This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:21:02 EDT