XHTML5

For technical support for all editions of CSE HTML Validator. Includes bug reports.

Re: XHTML5

Postby Albert Wiersch » Tue Nov 15, 2011 11:31 am

alnico wrote:My pages were valid XHTML1.1 until a year back. When HTML Validator added support for HTML5, I simply added the new tags, and removed the deprecated ones to match HTML5. When I used HTML Validator to check my syntax, I got a royal mess as I didn't know whether to move to HTML5 or stick with XHTML rules.


You could stick with HTML5 but write your documents with XML in mind.

alnico wrote:My pages are served up as application/xhtml+xml through the use of a PHP script - it's a neat script to serve up this MIME type to Standards compatible browsers, and serve a text/xml for IE6,7 etc. The script also renders and OBJECT tag instead of IMG for images, and renders an
Code: Select all
application/javascript
instead of
Code: Select all
text/javascript
.


That sounds like unnecessary server load & unnecessary complication to me. I think it would be better to just stick with HTML5 and not worry about serving XML/XHTML5, unless there is a good reason to do otherwise.

alnico wrote:I like XHTML because I notice the XML parser rendering pages on Firefox, Chrome & Opera load a page much faster than if served as text.


That's interesting. I haven't looked into that. I wonder if any "good" experiments have been done with regards to the speed of rendering XHTML vs HTML, though I suspect if the documents are written well that it doesn't or shouldn't make a significant difference. Parsing a page should not be a large % of the time needed to display a page, unless perhaps the page is "messed up" and the browser tries to fix it.

alnico wrote:These are the errors if I use HTML Validator to validate an XHTML syntax page as HTML5:
Code: Select all
<meta property="og:type" name="og.type" content="sport" />

The "meta" tag requires that the following attributes be mutually exclusive: "name" "http-equiv" "charset" "itemprop" "property". No more than one of these attributes may be used simultaneously.


What version are you using? v10.02 and v11.00 should be fine with that meta tag.

alnico wrote:
Code: Select all
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en-gb" xmlns:og="http://opengraphprotocol.org/schema/" xmlns:fb="http://www.facebook.com/2008/fbml" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:dcterm="http://purl.org/dc/terms/"  xmlns:ctag="http://commontag.org/ns#" xmlns:c="http://s.opencalais.com/1/pred/" xmlns:v="http://rdf.data-vocabulary.org/#" xmlns:foaf="http://xmlns.com/foaf/0.1/" xmlns:sioc="http://rdfs.org/sioc/ns#" xmlns:ui="http://java.sun.com/jsf/facelets" xmlns:sioctypes="http://rdfs.org/sioc/types#">

The "xmlns:v" attribute was found but is not allowed by the current configuration (because its category is not active). This could be because "xmlns:v" is a proprietary (non-HTML5) attribute.


Thanks. In the next v11 update, this message should no longer be generated.

alnico wrote:HTML5 requires that the "lang" attribute also be specified when "xml:lang" is used in HTML documents (this is not a requirement in XML documents). NOTE: The value of the "lang" and "xml:lang" attributes must be a case-insensitive match.


Thanks again. In the next v11 update, this error should not be generated for XML/XHTML documents.
Image
Albert Wiersch
User avatar
Albert Wiersch
Site Admin
Site Admin
 
Posts: 2361
Joined: Sat Dec 11, 2004 10:23 am
Location: Near Dallas, TX

Re: XHTML5

Postby MikeGale » Tue Nov 15, 2011 3:39 pm

I'd also be interested in performance tests of XHTML vs HTML. Anyone got something published?

For a long time I've assumed that a lot of code (bloat) and processor cycles are taken up with checking (and fixing) messy markup. The browser makers are locked into this support by a large mass of incompetent content. My understanding is that the techniques are fairly informal and would not be replicated if you started from scratch.

With xhtml it must obey simpler rules and if it fails, bad luck, you're dead. Much quicker. Above that (and for basically the same reasons) it is machine readable which is needed for various types of progress to happen.

It might be a great benefit for browser performance if the browser makers, in concert, started to remove this ridiculous processing (gradually). It would kill some web content but the pain would start fixing this horror!
User avatar
MikeGale
Rank VI - Professional
Rank VI - Professional
 
Posts: 604
Joined: Mon Dec 13, 2004 2:50 pm
Location: Tannhauser Gate

Re: XHTML5

Postby Albert Wiersch » Fri Nov 18, 2011 10:09 am

MikeGale wrote:It might be a great benefit for browser performance if the browser makers, in concert, started to remove this ridiculous processing (gradually). It would kill some web content but the pain would start fixing this horror!


If they tried to do that, then I think any "pain" experienced by this "improvement" would just cause people to start using a different browser that causes them less pain. :D
Image
Albert Wiersch
User avatar
Albert Wiersch
Site Admin
Site Admin
 
Posts: 2361
Joined: Sat Dec 11, 2004 10:23 am
Location: Near Dallas, TX

Re: XHTML5

Postby MikeGale » Fri Nov 18, 2011 10:20 pm

Yep. That's why I said in concert.

I saw the danger of going it alone, a day or two ago. Opera still failing on some sites.

Realistically I think you're right, it's wishful thinking. Maybe too late to do anything about it.

The unfortunate thing is that this kills programmatic access to most pages.
User avatar
MikeGale
Rank VI - Professional
Rank VI - Professional
 
Posts: 604
Joined: Mon Dec 13, 2004 2:50 pm
Location: Tannhauser Gate

Re: XHTML5

Postby Albert Wiersch » Fri Nov 18, 2011 11:18 pm

MikeGale wrote:Yep. That's why I said in concert.


Ahh yes, you did. :)

But even if they did do it in concert (which I'd give only a .01% chance of happening), then I think people would stop upgrading their browsers, thinking that every time they upgrade, it causes more problems and breaks more pages... and that would be a whole other problem in itself. :!:

I guess we're stuck with the practical! But like everything, the practical also has its positives & negatives.
Image
Albert Wiersch
User avatar
Albert Wiersch
Site Admin
Site Admin
 
Posts: 2361
Joined: Sat Dec 11, 2004 10:23 am
Location: Near Dallas, TX

Previous

Return to CSE Tech Support

Who is online

Users browsing this forum: MSN [Bot] and 2 guests