...

View Full Version : HTML5, XHTML2, and the future of the web



liorean
04-11-2007, 07:12 PM
Back in January I was asked if I was interested in writing an opinion-based article about HTML5 and XHTML2 for Digital Web Magazine. Of course I wrote it, so here goes:

HTML5, XHTML2, and the future of the web (http://www.digital-web.com/articles/html5_xhtml2_and_the_future_of_the_web/)

gsnedders
04-11-2007, 09:56 PM
To paraphrase what I said in #html-wg earlier, shouldn't it really be "Web Applications 1.0 (often referred to as HTML5)"? It also makes it seem as if the XHTML WG hasn't changed, whereas in fact it's been re-chartered and is also open to anyone (though it's questionable why anyone would want to).

What people have put in the comments (sadly) shows that all too many people still favour XHTML2, a standard without any real backing at all from implementers.

Oh, and your opinions surprise me having seen you on both whatwg and public-html mailing lists :p

liorean
04-12-2007, 12:42 AM
To paraphrase what I said in #html-wg earlier, shouldn't it really be "Web Applications 1.0 (often referred to as HTML5)"?Well, it should. Editor decision however, they felt my use of HTML5 and WA1 would be more confusing than the distinction would be worth. They wanted to use HTML5 throughout instead of WA1, except for one place where I was specifically talking about the non-HTML parts.
It also makes it seem as if the XHTML WG hasn't changed, whereas in fact it's been re-chartered and is also open to anyone (though it's questionable why anyone would want to).Well, it's basically the same WG rechartered and not a new one, though.
What people have put in the comments (sadly) shows that all too many people still favour XHTML2, a standard without any real backing at all from implementers.I think it shows they really didn't read what I wrote, or at least didn't think about what I could mean and read whatever they wanted to percieve it as.
Oh, and your opinions surprise me having seen you on both whatwg and public-html mailing lists :pExactly what were you surprised about? I thought my opinions here were pretty clear - compatibility with ALL parts of the web technology chain that have significant use is a MUST for any successor of HTML. XML only parsing is not an option, SGML never was (not even with HTML4.01), so HTML5 whether under the HTML WG or under WhatWG must standardise tag soup.
And I believe HTML WG and WhatWG HTML5 will be the same - anything else would be ridiculous! - no matter what alternatives Laurens Holst is lobbying for.

VIPStephan
04-12-2007, 12:50 PM
Interesting article.


[Ö] HTML5 whether under the HTML WG or under WhatWG must standardise tag soup. [Ö]

What Iím concerned of is that this could result in amateurs doing the same mistakes again and again, not even knowing they are doing mistakes just because all browsers will make malformed code display the same acceptable result - yet, an error is and stays an error, no matter how standadized the error handling is. This doesnít really help in making the web a better/cleaner placeÖ

The only solution would then be that browser vendors would agree that erroneous code will also result in erroneous display (while - in contrast to XHTML 2 - still displaying something) so that people donít have a choice other than producing valid code (after whatever specification). That might be the advantage of this - if applied like that.

Or am I wrong here?

I guess XHTML 2 could get us rid of a number of spammers as the effort would be too high to create a spam site that is conforming to XML standards. :D

liorean
04-12-2007, 01:04 PM
What I’m concerned of is that this could result in amateurs doing the same mistakes again and again, not even knowing they are doing mistakes just because all browsers will make malformed code display the same acceptable result - yet, an error is and stays an error, no matter how standadized the error handling is. This doesn’t really help in making the web a better/cleaner place…One of the objectives of the WhatWG is that a user agent supporting HTML5 (without legacy HTML support, that is) automatically should do the right thing with the vast amounts of invalid markup that already exists on the web. Document conformance is much more limited in scope, and conformance checkers similar to the current validators are already built.
The only solution would then be that browser vendors would agree that erroneous code will also result in erroneous display (while - in contrast to XHTML 2 - still displaying something) so that people don’t have a choice other than producing valid code (after whatever specification). That might be the advantage of this - if applied like that.Browsers can't be incompatible with the web. HTML5 browser rules aim to let browsers handle the web without having to reverse engineer the competitors by already having done that reverse engineering and writing it into a specification.
Or am I wrong here?Just a difference in perspective I guess. The WhatWG means to make it easier to do right than wrong while still making sense out of wrongs. And the HTML WG has very similar goals.
I guess XHTML 2 could get us rid of a number of spammers as the effort would be too high to create a spam site that is conforming to XML standards. :DWell, spammers take hte path of least effort. For that to work, all browsers would have to drop HTML support entirely, and that is not going to happen.

gsnedders
04-12-2007, 01:16 PM
Exactly what were you surprised about? I thought my opinions here were pretty clear - compatibility with ALL parts of the web technology chain that have significant use is a MUST for any successor of HTML. XML only parsing is not an option, SGML never was (not even with HTML4.01), so HTML5 whether under the HTML WG or under WhatWG must standardise tag soup.
Sarcasm David, sarcasm. :)


And I believe HTML WG and WhatWG HTML5 will be the same - anything else would be ridiculous! - no matter what alternatives Laurens Holst is lobbying for.
But most of Laurens alternatives are just cutting up the spec, which strikes me as even more pointless. If you look at HTML5 there are a large amount of comments within it noting things to change, as well as the visible notes within red boxes. If Hixie is the editor of both, why not start from what we already have, then change what we don't like?


The only solution would then be that browser vendors would agree that erroneous code will also result in erroneous display (while - in contrast to XHTML 2 - still displaying something) so that people donít have a choice other than producing valid code (after whatever specification). That might be the advantage of this - if applied like that.

If a specification breaks backwards compatibility on a large scale, it becomes irrelevant. If browsers break backwards compatibility on a large scale, they become irrelevant. People will simply use other browsers.

twodayslate
04-14-2007, 10:56 PM
Whoa, that was you!
I saw that on digg.

Nice, good work.



EZ Archive Ads Plugin for vBulletin Copyright 2006 Computer Help Forum