Re: Adding new character entities (was Re: [XHTML2] Unicode line and paragraph separators)

On 7 Apr 2003 at 18:39, Masayasu wrote:

> "Ernest Cline" <ernestcline@mindspring.com> wrote:
> 
> > Since XHTML1 has the Latin-1, Special and Symbol characters as sets of 
> > defined entities that are part of the normative definition, I see no 
> > problem in adding &ps; and &ls; to the set of special character 
> > entities for XHTML2 if a decision were made to add them to XHTML2.
> 
> Back to 1999, the W3C Internationalization Working Group advised other
> Working Groups not to add new character entities in W3C specifications.
> Following that advice, the HTML WG has "frozen" the (X)HTML entity sets,
> and various proposals for new character entities were rejected on that
> basis.  There is a suggestion to develop a universal character entity set
> for XML (not just for XHTML), but it is extremely unlikely that XHTML2
> would add its own new character entities.

Any idea on when a Universal enity set for XML might surface and who 
might need to be prodded/offered help to get work on it further along? 
If such a set were to be offered either before or at the same time as 
XHTML2, that would nullify that problem with &ps; and &ls;. Without it, 
I'll grant that depite any merits my idea has, this is a barrier that 
would need to be dealt with befre it could be implemented.  I'll still 
work on my survey of <p> in current web pages, so that the data is 
there to see whether it is worth trying to implement.

> > Can 
> > you name a single browser that supports XHTML1 as application/xhtml+xml 
> > that does not support the three entity sets?
> 
> Opera 6/7, Netscape 6.x, Mozilla 0.9.8 and earlier, to name a few.
> 
>   cf. http://www.w3.org/People/mimasa/test/xhtml/entities/results

I never have bothered with Opera which explains why I was unaware of 
its faults. As for pre-Mozilla 1.0 Gecko-based browsers, I wouldn't 
consider their failure to support the entities a serious impediment.
(By the way, what do you mean by NCR? I'm assuming numerical entities 
such as &#8233; and &#x2029; but it isn't clear from the page.) Quite 
frankly tho, based on the reported results, since the entities are part 
of the normative specification of XHTML1, I would consider Opera to be 
non-conforming for application/xhtml+xml.

Based on my current level of understanding, I can accept a lack of 
entity support when the document is transferred as text/xml or as 
application/xml as in such a case, no claims have been made at the 
media type level that the server is providing XHTML and that the client 
understands it.

Received on Monday, 7 April 2003 11:12:00 UTC