This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.

Bug 6298 - Provide a parser override
Summary: Provide a parser override
Status: NEW
Alias: None
Product: Validator
Classification: Unclassified
Component: Website (show other bugs)
Version: HEAD
Hardware: PC Windows XP
: P2 enhancement
Target Milestone: ---
Assignee: This bug has no owner yet - up for the taking
QA Contact: qa-dev tracking
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2008-12-10 16:30 UTC by Simon Pieters
Modified: 2018-05-09 20:10 UTC (History)
3 users (show)

See Also:
jordancarrillo530: needinfo+


Attachments

Description Simon Pieters 2008-12-10 16:30:02 UTC
From Bug 6296.

Please provide a parser override, at least between HTML5 and XML/SGML.

If you provide an override between XML and SGML, and the user chooses SGML parser with XML DTD or vice versa, you could either refuse to validate or change to an appropriate DTD that works with the parser.
Comment 1 Simon Pieters 2008-12-17 14:50:44 UTC
Or you could list both HTML5 and XHTML5 in the "Document Type" list.
Comment 2 Olivier Thereaux 2008-12-17 15:16:27 UTC
(In reply to comment #1)
> Or you could list both HTML5 and XHTML5 in the "Document Type" list.

I thought of doing that indeed. That'd work. I got mixed message as to whether it is a good idea to use the XHTML5 name however, as far as I can tell it's unclear what the best name is. Any pointer to what the currently accepted best term is?
Comment 3 Simon Pieters 2008-12-17 15:36:54 UTC
The spec uses "XHTML5" and so does Validator.nu. The term is pretty widely used on the Web. Any other name would probably have less backing and just be confusing for users, I think.
Comment 4 Karl Dubost 2008-12-18 13:20:02 UTC
(In reply to comment #3)
>  The term is pretty widely used on the Web.

hmm not widely used on the Web, only a few occurences. I would go through the Working Group and discuss that with Sam Ruby (new chair starting in January) and others. Just to make it a group decision. It Seems that Sam is interested by fixing the issues of communications with XHTML 2 WG.


Comment 5 Olivier Thereaux 2008-12-18 13:33:50 UTC
(In reply to comment #4)
> I would go through the Working Group and discuss that 

Got a few pointers from Mike Smith:
http://lists.w3.org/Archives/Public/public-xhtml2/2008Jun/thread.html#msg45
http://www.w3.org/html/wg/tracker/issues/52
 so things don't seem *that* simple. 

Mike suggests, as a relatively safe (if imperfect) alternative, to use something like HTML5 (XML Syntax).
Comment 6 Dean Edridge 2009-01-03 15:45:10 UTC
(In reply to comment #4)
> (In reply to comment #3)
> >  The term is pretty widely used on the Web.
> 
> hmm not widely used on the Web, only a few occurences.

It *is* widely used on the web, I know this because I have had Google alerts for XHTML5 for the last 3 1/2 years. XHTML5 has also been mentioned in books, magazines and on the BBC's web site. 
 
> I would go through the
> Working Group and discuss that with Sam Ruby (new chair starting in January)
> and others. Just to make it a group decision. It Seems that Sam is interested
> by fixing the issues of communications with XHTML 2 WG.

This has already been sorted out Karl [1]

The way to solve the problem for good is to throw away the rejected XHTML 2 proposal. It's not implementable, was rejected by the browser vendors 8 years ago, it offers no benefits over XHTML1 or XHTML5 and it's very existence continues to hold back the progress of the web. And despite what some people might say, it will not lead to the successful implementation or deployment of XForms. 

You can't use "HTML5+XML" even in the short term as both HTML5 and XHTML5 can use "XML syntax" (some XML syntax is valid in text/html). This will just confuse people as it's not the syntax that distinguishes HTML5 from XHTML5 [2] You'll have to use "HTML5/XHTML" while we wait for the issue to be "officially" resolved. This label would be OK for the short term as the spec makes it clear that XHTML can not be used as text/html but there is no such wording for "XML syntax" or "HTML5+XML" so that will have to be changed.

[1] http://lists.w3.org/Archives/Public/public-html/2007Oct/0386.html
[2] http://dev.w3.org/html5/spec/Overview.html#html-vs-xhtml
Comment 7 Karl Dubost 2009-01-03 16:03:40 UTC
(In reply to comment #6)
> It *is* widely used on the web, I know this because I have had Google alerts
> for XHTML5 for the last 3 1/2 years. XHTML5 has also been mentioned in books,
> magazines and on the BBC's web site. 

I should have backed up what I said. :)

http://www.ask.com/web?q=xhtml5 Showing 1-10 of 12,700
http://www.ask.com/web?q=xhtml1 Showing 1-10 of 533,000

I'm following also the discussions on different alerts ;)


The proposal of Mike is reasonable  and there are still a few issues to solve in terms of community and agreements. 
See http://intertwingly.net/blog/2008/12/15/Co-Chair-HTML-WG and the comments.

(btw I have no preferences over a term. I just pointed out that there are still two voices.)


Comment 8 Dean Edridge 2009-01-14 13:12:56 UTC
(In reply to comment #7)
> (In reply to comment #6)
> > It *is* widely used on the web, I know this because I have had Google alerts
> > for XHTML5 for the last 3 1/2 years. XHTML5 has also been mentioned in books,
> > magazines and on the BBC's web site. 
> 
> I should have backed up what I said. :)
> 
> http://www.ask.com/web?q=xhtml5 Showing 1-10 of 12,700
> http://www.ask.com/web?q=xhtml1 Showing 1-10 of 533,000

And what does this prove?

> 
> I'm following also the discussions on different alerts ;)
> 
> 
> The proposal of Mike is reasonable


Karl, I have just explained that using "XML" instead of "XHTML" is problematic. Only "XHTML" can distinguish between HTML5 and XHTML5 as it is possible to use XML syntax in HTML5 text/html web pages.

I'm sure Mike's just trying to be nice and keep the peace, but it is not a good solution.

> and there are still a few issues to solve
> in terms of community and agreements.

So what? We can't let silly politics hold back the progress of the web and the validation of XHTML web sites Karl.
 
> See http://intertwingly.net/blog/2008/12/15/Co-Chair-HTML-WG and the comments.
> 

Yeah, I've already seen that, so what? Sam's heading in the wrong direction and this has been pointed out on his blog and on www-html.

> (btw I have no preferences over a term. I just pointed out that there are still
> two voices.)
> 

<sigh>


Comment 9 Damien B 2009-01-14 14:25:54 UTC
> I should have backed up what I said. :)
> > 
> > http://www.ask.com/web?q=xhtml5 Showing 1-10 of 12,700
> > http://www.ask.com/web?q=xhtml1 Showing 1-10 of 533,000
> 
> And what does this prove?

If you look closely you'll see that the term "XHTML5" has virtually no existence outside of the WhatWG circle. Proposing "HTML 5 (HTML5 syntax)" and "HTML 5 (XML syntax)" is clearer, fully in accordance with the Draft, less misleading regarding XHTML and won't prevent the few people really accustomed to "XHTML5" from using the validator.