RE: [WIDGET PC] i18n comment 6: Use of its:dir

Hi Marcos,

Without understanding why folks think these things, it's a little difficult to provide specific discussion points, but the key point behind ITS is that it describes types of information you need for a well-internationalized and easily localizable specification.  The implementation of those ideas is secondary.  

One such type of information relates to bidi support.  If your markup is to support use for the languages of the millions of people who use a script such as Arabic, Hebrew, Thaana, N'ko, etc, you need to provide a simple way to set and change the base direction for parts of the document. It is important for that information to be inherited through contained elements - you can't do that using Unicode controls.  Markup is also preferable to control codes because it makes editing and maintenance of code much easier.

ITS happens to suggest a name for the its:dir attribute and describe what values you'd need, and what that means.  But it's not about supporting ITS here.  It's about providing a mechanism for people using right-to-left and particularly bidirectional text to achieve what they need to. I don't think that is optional. Call the attribute whatever you want, I believe you need it and its behaviour to be available to users.

I didn't review the Widgets spec myself (we have so much to cover that we have to share things around), but I'm starting to think maybe I should try to find the time, if I can. For example, I noticed that the its:dir attribute can be used on the name, author, and description elements, but I would have expected that for a widget in Persian, say, you'd just set the attribute on the widget element and it should take care of all those without the author having to separately and laboriously markup them up.  (Think about HTML - you put dir on the html element, not on every p, div, list, etc.)  You'd only need to use dir on name, author, span, etc if you need to *change* the base direction.  (We had a similar gap in SVG Tiny markup before Xmas, which they fixed as soon as we pointed it out to them.)

Well, I hope that's of some use.  I'll try to take another look next week, if I can.

Cheers,
RI


============
Richard Ishida
Internationalization Lead
W3C (World Wide Web Consortium)

http://www.w3.org/International/
http://rishida.net/



> -----Original Message-----
> From: Marcos Caceres [mailto:marcosc@opera.com]
> Sent: 10 July 2009 16:49
> To: Phillips, Addison
> Cc: ishida@w3.org; public-Webapps@w3.org; public-i18n-core@w3.org
> Subject: Re: [WIDGET PC] i18n comment 6: Use of its:dir
> 
> 
> 
> On 7/10/09 5:40 PM, Phillips, Addison wrote:
> > (personal response)
> >> Fantastic. Unfortunately, implementer feedback has raised concerns
> >> about ITS and so the WG has put ITS features "at risk" (and marked
> >> as
> >> such in the soon to be released CR spec). We will see what happens
> >> in
> >> CR; hopefully implementers will understand the value of
> >> implementing
> >> it.
> >>
> >
> > I noticed that in the transreq. Obviously the I18N WG is concerned about
> any such feedback: can webapps please share what the concerns are? I know
> that many developers consider bidi support "kind of scary"; I hope that the
> implementation issues are not simply fear driven. Is there some way that the
> Internationalization community can support implementers?
> >
> 
> The concerns are that people won't implement ITS (or that authors can
> use the appropriate Unicode markers to achieve the same thing as ITS).
> 
> Kind regards,
> Marcos

Received on Saturday, 11 July 2009 06:19:30 UTC