W3C home > Mailing lists > Public > public-html@w3.org > August 2008

Re: Translation control in HTML5

From: Felix Sasaki <fsasaki@w3.org>
Date: Fri, 01 Aug 2008 14:17:27 +0900
Message-ID: <48929C67.3000808@w3.org>
To: public-html@w3.org
CC: public-i18n-its-ig@w3.org

A comment below, marked with >FS:

On Fri, 1 Aug 2008, Jirka Kosek wrote:
> Ian Hickson wrote:
> > On Thu, 31 Jul 2008, Jirka Kosek wrote:
> > > > Very good point.  lang="only de" would also work here while even being
> > > This would completely break syntax of lang attribute values which should
> > > be following BCP 47 (http://www.rfc-editor.org/rfc/rfc4646.txt)
> > 
> > We define lang, we can easily define it as being something akin to:
> > 
> >    [only] <bcp-47-code>
> 
> And what is point in doing it differently from dozens of other markup 
> languages?

The main benefit of this proposal is that it would automatically work in 
already shipped tools. It also avoids adding yet another global attribute, 
especially one for a relatively minor use case, with an unproven track 
record (i.e. we don't know if authors will use it anyway).

>FS: In a sense "minor use case" is like a chicken and egg problem ... but anyway. I have put the ITS IG into the loop 
(as Jirka did before).
Participants of the IG deal with lots of automatic and manual translation tools every day, including many
HTML documents. Maybe they can provide more data on the question "minor or major use case"?
During the development of ITS we discussed many issues for the "Translate" information problem which 
have been discussed in this thread, including overlap with styling or language, information for the whole document vs. only
parts / singular elements and attributes, defaults and exceptions etc., and came up with the current solution. We
also co-ordinated with other localization industry groups with similar concerns to make sure to not mess up things
 and to have a re-usable solution.

Btw., Google is also participating in the development of the BCP 47 for language tags. So you might have software 
which depends on lang="" already, independently of the "Translate" use case. Maybe ask Mark Davis to confirm.

Felix


> Moreover, having separate lang and translate attributes will be much 
> more friendly to CSS selectors.

That's certainly an important advantage of keeping these in separate 
attributes, yes. We have to balance all the various pros and cons in 
coming up with a solution (assuming we want to address this use case).
Received on Friday, 1 August 2008 05:18:28 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:16:21 GMT