Re: ITS elements needing span or span-attributes

Hi Yves, all,

Although I wrote the list in Yves mail, I am not sure if we should adopt it.

The reason: <span> contains ruby. If we allow ruby, rb etc. to contain
ruby, we are not conform to the ruby spec anymore. See
http://www.w3.org/TR/ruby/#definition , where it says e.g.:

rb  	Common  	(PCDATA | Inline - ruby)*

That is: nesting of ruby in ruby is not allowed.

I think this is only one example of problems which arise if we allow ITS
attributes / elements to be used as local markup / global rules about
other ITS markup. We have not thought through all consequences this
could have.

So my proposal is:

a) Take more time to think this through
b) Not to allow ITS markup on top of other ITS markup

I think if a) does not help to solve problems like mentioned above, b)
will be the choice.

Cheers,

Felix


Yves Savourel wrote:
> Hi all,
> 
> Felix did my action item of looking into the ITS elements to see which ones needed.
> (Thank you Felix :)
> 
> See below (I don't have much more to say).
> 
> ==========
> Subject: Your action item http://www.w3.org/2006/07/19-i18nits-minutes.html#action01
> 
> Hi Yves,
> 
> I had a look at your action item on ITS elements that can contain text.
> I also looked at where ITS local attributes are useful (which seems to be a super set).
> 
> this needs span in the content model:
> span itself
> [23] 	locInfo.content	 ::= 	 text
> the yet to be declared global termInfo element
> [47] 	rb.content	 ::= 	 text
> [50] 	rt.content	 ::= 	 text
> 
> this needs the span attributes:
> [24] 	locInfo.attributes (has them already)
> the yet to be declared global termInfo element
> [48] 	rb.attributes	
> [51] 	rt.attributes
> [54] 	rbc.attributes
> [57] 	rtc.attributes
> [59] 	rp.content	 ::= 	 text
> [60] 	rp.attributes
> 
> Cheers,
> 
> Felix