W3C home > Mailing lists > Public > www-international@w3.org > January to March 2013

Re: I18N-ISSUE-207: Term data category and dfn element [.prep-ITS-20]

From: Felix Sasaki <fsasaki@w3.org>
Date: Tue, 15 Jan 2013 15:55:07 +0100
Message-ID: <50F56DCB.8080203@w3.org>
To: Yves Savourel <yves@opentag.com>
CC: www-international@w3.org
Am 15.01.13 15:17, schrieb Yves Savourel:
>>> We should be consistent: put either all mappings in
>>> the specifications or in the BP document.
>> Understand. So then rather do this normatively in the spec,
>> and summarizing maybe in a section "existing HTML markup and ITS"?
> Whether we do it in the specification or the BP, it would likely to be a normative change. That is, if we do in in a BP, text like "Applying the Id Value data category to xml:id (in XML) or id (in HTML) attributes in global rules is not necessary, since these attributes are the recommended way to specify an identifier." would need to be changed.
>
> What probably matter most in the long term is compliance: Do we want an ITS processor that supports Terminology for HTML to 'understand' <dfn> by default, or hope it does follow the BP?

For ITS 1.0, we were happy with the latter, and created rules for 
mapping many other data categories to HTML
http://www.w3.org/TR/2008/NOTE-xml-i18n-bp-20080213/#relating-its-plus-xhtml
We also cretated a mapping for "Terminology" - but interestingly not to 
"dfn", but to "dt"

<its:termRule selector="//h:dt" term="yes"/>


>
> As a user I'd rather have processors I can count on to do the proper mapping by default.
>
> but, in the other hand, defining all rules for HTML5 is likely to be a bit challenging:
>
> - What about data categories like Elements Within Text? After all <b>, <u>, etc. should be mapped to withinText='yes'.
> - Do we also define the translatable attributes?
> - How do we address the thorny case of Preserve Space for <pre>?
> - Is the content with <del> translatable or not?
> - etc.
>
> This could get ugly and time consuming, and may be easier to handle in a BP for which time constraint is more flexible.

I agree, see also as a backup the history in what is a term - for ITS 
1.0 we mapped to "dt", now we are discussing "dfn". So I'd vote for a 
best practice and flexibility.
At Richard (with regards to your other mail about the status of this 
discussion): sure, this will be discussed in the MLW-LT group too - they 
are aware of this thread and the i18n Wg will receive a "formal" answer 
too. For the time being we are trying to gather opinions on this and 
other comments, and everybody is happy to chime in.

Best,

Felix

>
> So I don't know...
>
> cheers,
> -yves
>
>
>
>
Received on Tuesday, 15 January 2013 14:55:37 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 15 January 2013 14:55:37 GMT