XHTML WTF

2 Jul 2009 3 pm eastern

The web’s future isn’t what the web’s past cracked it up to be. 1999: XML is the light and XHTML is the way. 2009: XHTML is dead—kind of.

From the W3C news archive for 2 July 2009:

XHTML 2 Working Group Expected to Stop Work End of 2009, W3C to Increase Resources on HTML 5

2009-07-02: Today the Director announces that when the XHTML 2 Working Group charter expires as scheduled at the end of 2009, the charter will not be renewed. By doing so, and by increasing resources in the Working Group, W3C hopes to accelerate the progress of HTML 5 and clarify W3C’s position regarding the future of HTML. A FAQ answers questions about the future of deliverables of the XHTML 2 Working Group, and the status of various discussions related to HTML. Learn more about the HTML Activity. (Permalink)

Tags: W3C, XML, XHTML, HTML, HTML5, WTF

Filed under: HTML, HTML5, Markup, Web Design History, Web Standards, XHTML

116 Responses to “XHTML WTF”

yatil! said on 1 January 1970 at 12:00 am:
gp said on 3 July 2009 at 2:06 pm:

“The web’s future isn’t what the web’s past cracked it up to be. 1999: XML is the light and XHTML is the way. 2009: XHTML is dead—kind of.” -XHTML WTF – Jeffrey Zeldman Presents The Daily Report
Nathan Arnold said on 2 July 2009 at 3:11 pm:

WAT!?!?!?!
Andrew Mager said on 2 July 2009 at 3:11 pm:

I <3 HTML 5.
Niki Brown said on 2 July 2009 at 3:12 pm:

I’m actually kind of excited for HTML5… its time for something new!
Jeff Siarto said on 2 July 2009 at 3:12 pm:

This isn’t all bad news. Most browser companies are getting behind HTML 5 (which is a good thing) and XHTML 2 has been dead in the water for some time. Aren’t web standards more powerful when everyone is behind a single standard?
Steve said on 2 July 2009 at 3:16 pm:

I don’t understand why this is happening. Didn’t XHTML fix all the inconsistencies in HTML? HTML5 seems to bring back a lot of those inconsistencies from what I’ve seen so far. WTF indeed.
Adam said on 2 July 2009 at 3:16 pm:

The sad fact is that XHTML2 hasn’t been getting any traction – so if this means that the folks responsible for all the good ideas in XHTML2 start contributing to the the de facto future spec (HTML5) then good on them.

I think a lot of us find the idea of revolution attractive, but in actual practice reformation is far more pragmatic, realistic.
Anthony Barns said on 2 July 2009 at 3:17 pm:

Surprising but looking forward to HTML 5.
Michel said on 2 July 2009 at 3:20 pm:

Alright, I agree — this was kind of expected. I don’t think XHTML 2 had much future, now that HTML 5 emerged and the focus shifted to it.

What is strange to me (and I am not yet sure I’ve found the answer to this question) is that HTML 5 is fine, but where is XHTML (5?). I have nothing against HTML (without the ‘X’ in front), but in HTML there are some omissions — I can, for example, write badly formatted code, and it will be perfectly valid in HTML 4.01 and HTML 5. This is not nice… I like my code to be well written, my tags properly opened and closed (and correctly nested), etc. But HTML is too loose and will allow me to make mistakes — which I don’t (obviously) want.

So… I wonder: Where is the ’stricter brother’ of HTML 5?… Or we are going back now?
Ramiro Ruiz said on 2 July 2009 at 3:21 pm:

Well I’m going to miss XHTML but I agree with Jeff. This can be good news
Alex said on 2 July 2009 at 3:21 pm:

So what happens now? Leave tags open? Validate code to what standard? Sigh, Googling HTML 5 resources..
Wiley said on 2 July 2009 at 3:21 pm:

Video tags. Enough said.
mike said on 2 July 2009 at 3:21 pm:

reading the FAQ, it sounds like the HTML5 working group will now be responsible for the same deliverables that the XHTML2 working group was going to produce.

sounds more like the groups are being merged (i would imagine lots of overlap here)
Josh Lasdin said on 2 July 2009 at 3:23 pm:

This is depressing news. However, I guess it will just be up to the designer/developer to use HTML5 semantically, continuing “sensible code” and not letting it fall to the sloppy standards of pre-XHTML. :/
Erik said on 2 July 2009 at 3:23 pm:

It looks like the W3C intends XHTML to be alive and well as part of the HTML5 specification. (See the FAQ linked in Zeldman’s post for more about this.) If the good stuff from XHMTL2 can make it into (X)HTML5, it’s all for the better.
Andrew Vit said on 2 July 2009 at 3:24 pm:

XHTML is not dead… They only stopped working on what was going to be the alternative to HTML5. You can still use HTML5 with XHTML syntax served as application/xhtml+xml…

See here
Sergio Corrales said on 2 July 2009 at 3:29 pm:

wow! seriously?! FML! Come on, we need to get one set of standards, not two!
Fred said on 2 July 2009 at 3:30 pm:

Wait, this isn’t news about Michael Jackson?
Tantek said on 2 July 2009 at 3:32 pm:

It makes sense to stop work on XHTML2. It stopped being relevant many years ago.

Disclosure: I was a representative in the HTML working group that worked on XHTML2, and disagreed strongly with the priorities of the rest of the members of the working group. From a blog post over six years ago:

The biggest problem that I have with XHTML 2 is not just its misnaming/misplacement as a “future version of HTML”, but also the amount of the HTML Working Group’s time that it consumes at the expense of what I think are imminently more important things for the HTML Working Group to work on.

Jeffrey, we were all shouting about this six years ago:

Mr. Zeldman: http://www.zeldman.com/daily/0103b.shtml#skyfall

Mark Pilgrim: http://diveintomark.org/archives/2003/01/13/semantic_obsolescence

Eric Meyer: http://meyerweb.com/eric/thoughts/2003a.html#t20030114

But the W3C staff/chairs didn’t listen.

It took the Great Web Standards Schism of 2004 and the subsequent formation and momentum of the WHATWG to actually iterate on HTML4.01 (as my past blog post had suggested should have been the top priority of the W3C HTML Working Group *six years ago*).
Matthew Kempster said on 2 July 2009 at 3:38 pm:

WTF?!

jkz
Alex Lemanski said on 2 July 2009 at 3:39 pm:

I think this could be a good thing if some of the Resources can be redirected to working with Microsoft to support application/xhtml+xml in IE. That way we can truly have full support for XHTML cross browser. Lets make this the goal now instead!
Dougal Campbell said on 2 July 2009 at 3:40 pm:

I, for one, welcome our new HTML5 overlords!

XHTML 2 was sort of a mutant offspring, anyways.

I plan to use XHTML 5 served as text/html, just to tork off the purists, just like I do now with XHTML 1.0.
Jonas Downey said on 2 July 2009 at 3:46 pm:

Man, I am super glad I spent time switching from HTML4 to XHTML1. Maybe XHTML1 was like Vista, you should have just skipped it and waited for a better version to come out. :)
beth said on 2 July 2009 at 3:56 pm:

Guess the web won’t be moving to XML after all!
Alan Bristow said on 2 July 2009 at 3:57 pm:

I like the strictness of X(html), guess I’ll just have to move HTML5, eventually, and kid myself that it too has a nice strict policy and adhere to it.
Bryan Hoffman said on 2 July 2009 at 3:58 pm:

This is going to be a real issue for those of us who are vested in systems that use XHTML, such as Wordpress.

This site is powered by Wordpress. What are your plans Jeffery?
Robert Spangler said on 2 July 2009 at 3:59 pm:

So, since xHTML seems to be leading into a dead-end, have you started to make the switch back to HTML 4.1 vs. using xHTML?

I’m thinking about it after reading Cameron Moll’s post on it
Grant Elliott said on 2 July 2009 at 4:01 pm:

Someone needs to be shot for stopping xhtml 2’s release. HTML is bad enough as it is. I was looking forward to actually looking forward to CSS and the new ways to section and do navigation and such. I just scratched the surface with my xhtml reading and im in love with it. HTML 5… not so much. Haven’t we been working to move away from pages filled with crap defining colors and towards pages containing useful info and tags to define who is what? This is moving backwards. Its a bad move.
Masterleep said on 2 July 2009 at 4:02 pm:

Do not panic, XHTML fans. XHTML lives on as the XML serialization of HTML5, called XHTML 5. What is dead is the less backwards compatible language XHTML 2 that was going nowhere fast anyway.

See, it’s better – it jumps 4 version numbers in one fell swoop!
Jordan Clark said on 2 July 2009 at 4:04 pm:

I guess this day has been coming for along time now…

The ideal of XHTML was always a bit much to ask, anyway; the language was always going to be shunned by the vast majority of mainstream websites. The only people who would ever have used XHTML correctly are standardistas.
Scott said on 2 July 2009 at 4:09 pm:

All that I ask is to make unclosed tags invalid. Something about the trailing slash in an image tag just makes everything feel right. Grant me that and the future of xHTML makes no difference to me.. (Oh yeah, I also don’t want to see any capital letters anywhere. Thanks.)
Neal G said on 2 July 2009 at 4:10 pm:

Like a few people have already mentioned XHTML5 remains. In all honesty this was a good move for the W3C, espcecially for people who are fans of XHTML because it was causing them to choose between XHTML5 and XHTML2.

Me personally, I could care less since I’ve given up on XHTML years ago, and strictly use HTML 4.01.

I’ll never understand why people would serve their XHTML pages as text as almost everyone does. And even if you do serve your XHTML as XML it’s nearly impossible to reach 100% validation at all times (see also: CMS)
Rimantas said on 2 July 2009 at 4:11 pm:

@Grant Elliott: you do realize that XHTML1.0 semantically is the same as HTML4.01?
And HTML5 gives you what you ask: it has more elements for document’s structure.
I am pretty sure, that anyone mourning for XHTML does not know much about it.
Dan Wilkinson said on 2 July 2009 at 4:11 pm:

Oh man…what am I going to do with my XHTML Fist t-shirt? Does this mean it qualifies as “vintage” now?
Andrei Eftimie said on 2 July 2009 at 4:14 pm:

I don’t understand most of you people complaining about Tag Soup and bad code. You can have perfectly fine code in HTML 4.01 Strict, as well as in HTML 5.

Just write good code. Stop complaining. And try to understand that HTML offers much more flexibility and much more safer[1] then XHTML.

[1] And by safe i mean if XHTML would actually be treated as XHTML and not HTML (as in XML parser and not a SGML parser). Then if one of the editors of your site would introduce a wrong formatted content, the whole page would fail.
Seth said on 2 July 2009 at 4:14 pm:

I’ve been doing this for so long that I honestly can’t tell the difference. There are very few things that I note as different between HTML and XHTML…and it’s the X. The real differences that I use on a daily basis are minimal and more along the lines of “Best Practices” than “Working or Broken”. Like the closing /> or . Unless I’m missing something, why does XHTML matter? What are the major issues/features we’re not getting in this transition? What are we losing/gaining? If XHTML was doing amazing things then why is it dead?

I say this with a lot of real world experience and honestly not being able to tell the difference. I’ve been doing this for 13+ years and day to day it’s mostly just preference…from what I can tell. I do say I code CSS and HTML mostly because the X is too long :) and as most designers I’m somewhat verbally lazy.
Byekick said on 2 July 2009 at 4:14 pm:

Some earlier comments mentioned looking for HTML5 resources – in which case can I recommend two for your consideration:

HTML5 Doctor has lots of great, really really practical advice aimed at stuff you can use, not impenetrable specs.

One of the people involved in HTML5 Doctor is Bruce Lawson at Opera

I was lucky to hear Bruce at @media last week talking with Molly about HTML5 and what it really means – you couldn’t find a calmer, more helpful voice. His site is using HTML5, so have a peak at view source too.

My own take on this is that W3C have done the web a bit of a favour here, if leaving XHTML’s future looking more than a little uncertain. We’ve done the days of trying in vain to support two different (browser) standards – we really don’t need it again in markup.

HTML5 can only be as good as we make it – so get involved if you can.
Andrew said on 2 July 2009 at 4:21 pm:

My concern is the extensibility. We’ve been screwing around with cramming semantic data into html using microformats and now RDFa, when all we really need is an extensible version of html that will let us define the meaning of our content. Now what?

Between this and Outlook 2010, I feel like reinstalling GoLive 5 and cranking some Fastball.
Aaron said on 2 July 2009 at 4:29 pm:

I figured it was coming, and glad to see it so we can get the show on the road with HTML 5.
Bye-bye XHTML | Interactive Design Lab said on 2 July 2009 at 4:31 pm:

[...] http://www.zeldman.com/2009/07/02/xhtml-wtf/ (No Ratings Yet) Loading … Thursday, July 2nd, 2009 News [...]
Twitted by markmaloney said on 2 July 2009 at 4:34 pm:

[...] This post was Twitted by markmaloney [...]
XHTML 2 is Dead at Space Ninja said on 2 July 2009 at 4:51 pm:

[...] I didn’t see this coming. Zeldman reports that the W3C is not going to renew the XHTML 2 working group’s charter this year. That [...]
Aaron Burrows said on 2 July 2009 at 4:51 pm:

I was really liking XHTML2 over HTML5, but XHTML2 lost so much momentum. I’m not in favor of the sloppiness of HTML5, but we can still code with XHTML strictness. I always served XHTML1 as text/html so I guess it’s of little consequence. Maybe this will be a good thing, and HTML5 will be finished and adopted much sooner than the two slightly (yet significantly enough) different languages would have been. I’m looking forward to seeing what becomes of it.
XHTML 2 is Dead - Pop Art Blog said on 2 July 2009 at 4:52 pm:

[...] I didn’t see this coming. Zeldman reports that the W3C is not going to renew the XHTML 2 working group’s charter this year. That [...]
Derek Beumer said on 2 July 2009 at 5:49 pm:

I think the biggest potential hit here is to education and advocacy. Having the XHTML standard to reference when advising new and old developers alike gives weight to the argument for well-formed code, helping to promote the adoption of good coding habits without the individual needing to have already grasped for themselves the full reasoning behind their use.

Vagueness in standards is a primary detractor to their adoption, and is something that HTML 5 has sought to eliminate in many cases, which makes the flexibility on syntax all the more puzzling.
Ian Hickson said on 2 July 2009 at 5:55 pm:

I plan to use XHTML 5 served as text/html, just to tork off the purists, just like I do now with XHTML 1.0.

You actually can’t, because “HTML5″ vs “XHTML5″ is defined not by the syntax but by the MIME type. So just serving anything as text/html means that it is HTML5. :-)
Josh Stodola said on 2 July 2009 at 6:09 pm:

I always knew the stupid rules of XHTML had no benefit what-so-ever, just a bunch of false positives for XML fanboys to pitch in salesman fashion. It was a big waste of my time learning to conform to that nonsense, all because I wanted to see a green “Congratulations” when I click the validate button. Sigh. I wonder what the next five years has in store.
Jeff Croft said on 2 July 2009 at 6:10 pm:

Good riddance.
Jesse Gardner said on 2 July 2009 at 6:21 pm:

I could be wrong, but the inconsistencies drive me mad and make me care less about conforming to standards.

For example, as I understand it, a closing slash in an img tag is required in XHTML, but not allowed in HTML4; whereas a closing tag IS allowed (required?) on a break tag.

Can anyone point me to good resources regarding the differences/similarities of both?
Jeffrey Zeldman said on 2 July 2009 at 7:12 pm:

Can anyone point me to good resources regarding the differences/similarities of both?

Dave Shea discussed why he switched from XHTML 1.0 back to HTML 4 in Switched.

And I can’t vouch for every entry’s quality, but …

* XHTML 1, HTML 4, or HTML 5?

* XHTML 1.0 Transitional vs. HTML 5

* X/HTML 5 vs. HTML 2

* HTML 5 vs. XHTML 2

We’re planning a discussion on this topic on A List Apart soon.
nick said on 2 July 2009 at 8:02 pm:

Time for you to switch as well, Mr Zeldman :)
Billee D. said on 2 July 2009 at 8:04 pm:

Crazy. I’m kind of torn, actually. On one hand I really like the possibilities that HTML 5 will offer — and the AEA website is a nice example of starting down the HTML 5 path — but on the other hand I have heard about (and experienced) some issues and inconsistencies working with it.

Sites like http://html5doctor.com/ will help to increase interest in HTML 5, but I think I’m stuck using XHTML until HTML 5 has more browser support. I still wish that I could actually serve-up XHTML as XML…you know, like it’s supposed to be. :-/
Joost said on 2 July 2009 at 8:34 pm:

Serving XHTML formatted code as HTML/SGML never made much sense to me.
Serving XHTML as XML hasn’t been really practical.

Not really surprised by this decision.
Ralph said on 2 July 2009 at 8:38 pm:

@Bryan Hoffman: Robert Nyman offers a simple code to allow you to deliver html with WordPress: http://bit.ly/Lgusm
XHTML2 is dead, long live HTML5 | Web Directions said on 2 July 2009 at 8:45 pm:

[...] a lively conversation over at zeldman.com with excellent input, particularly from Tantek Çelik, who has been involved with the development [...]
horuskol said on 2 July 2009 at 8:52 pm:

This is not too surprising a development as HTML5 development includes XHTML5 – so why have two working groups duplicating efforts.
Dale Cruse said on 2 July 2009 at 9:04 pm:

Looks like I have something fresh to work on this weekend.

I wonder if I should apply HTML5 the book companion website I’m creating.
Coding like it's 1999 ~ Authentic Boredom said on 2 July 2009 at 9:04 pm:

[...] like you made the right call, Camron. Zeldman is reporting that the XHTML 2 team has made a decision to not continue their work past the end of the year. It [...]
XHTML��ˣ�HTML5��ʱ�� - WEB��׼��ר�� - ��̳ ��ҳ-��վ-��-��-�� - Powered by Discuz! said on 2 July 2009 at 9:18 pm:

[...] [...]
Horus Kol » XHTML2 Being Closed in Favour of X/HTML5 said on 2 July 2009 at 9:26 pm:

[...] Jeffrey Zeldman puts it in his recent blog post – XHTML is dead, [...]
JeffCroft.com: Zeldman: XHTML WTF said on 2 July 2009 at 9:50 pm:

[...] http://www.zeldman.com/2009/07/02/xhtml-wtf/#comment-43921 [...]
Andrew M said on 2 July 2009 at 10:04 pm:

Is Google’s leverage of HTML5 driving any of this change by the W3C? It seemed as though XHTML was the future and had been so for the past few years.

see “Google Throws Its Weight Behind HTML 5″
http://www.webmonkey.com/blog/Google_Throws_Its_Weight_Behind_HTML_5
Erik Vorhes said on 3 July 2009 at 12:45 am:

But, @nick, that means he’ll need to rethink his clever (and appropriate but currently and inexplicably invalid in HTML5) practice of marking up commenter names with <cite>!
Jeff said on 3 July 2009 at 2:37 am:

Hello,

Just for clarification, with that news, does that mean I should use other doctypes when I start developping a website ?
I am kind of lost about the technical implications of this…
Jaap said on 3 July 2009 at 3:51 am:

It seems there’s a lot of misinformation going around about HTML5, I would’ve thought web developers were better informed…

Time for some PR for HTML5?
Davide said on 3 July 2009 at 3:51 am:

Can anyone point a resource outlining the difference between the future HTML and XML? In the latest year I was trained to believe that HTML il EVIL because you can leave tags open. Will one be allowed to do that in HTML %?
Davide said on 3 July 2009 at 3:52 am:

Also, for the sake of chronicle, whenever I tried to parse XHTML as XML through ruby I failed miserably, falling back to hpricot, which treats everything as HTML, and works very well, so maybe some looseness isn’t that bad.
Davide
Mike Whitehurst said on 3 July 2009 at 3:59 am:

There’s a long time to debate this, IE6 doesn’t support HTML5. IE8/FF/Safari have partial support. So we’re back to that again. We can’t actually use HTML 5 yet because we want to provide a good experience accross the range of browsers our visitors have. XHTML will remain the best way to do this for many years to come.
Jordan said on 3 July 2009 at 5:15 am:

Despite having the same shock when I saw the announcement, I was under the impression that this isn’t necessarily a bad thing because HTML 5 will essentially incorporate all of the XML focused elements of XHTML (namespaces, etc).

I could be missing some big omission, but it has seemed to me for a while that HTML 5 looks much more exciting and likely to succeed than XHTML 2.
Danny said on 3 July 2009 at 5:18 am:

XHTML 2.0 + XForms + Xlink + any other XML based standard could be a future of the web. Just read point 6 of the FAQ – there is no clear idea how to implement extensions. And there is always problem with semantics and missing tags (requests for audio/video/menu/date/section/ and finally my own) which can’t be easilly implemented in HTML 5 and could be in XHTML by namespaces. Practically XHTML 5 is dead because IE does not support application/xhtml+xml. One more thing. W3C is became more and more company related/addicted. If You haven’t noticed they just removed ogg codecs (http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2009-June/020620.html) because it’s not in the interest of Apple. So, what standards are we talking about? Standards made by corporations to fill their pocket?
Rijk said on 3 July 2009 at 6:40 am:

It isn’t very amazing that many people still get confused, but once you get down to basics, things are really simple…

XHTML isn’t dead. Both ‘XHTML2′ and ‘XHTML5′ could/can be used, in theory, for enforcing validity and distributed extendability. The HTML5 effort doesn’t take away any of those (mostly theoretical) possibilities.

You can write with your precious extra backslashes and add things like RDFa or even self-cooked extensions all you want if you can deal with namespace wrangling, and if you send it out as application/xhtml+xml with the HTML5 Doctype: <!DOCTYPE html> it will be valid. And if you send it out as text/html, it will also work and not even get marked as invalid, because HTML5 is nice for you, except that you can’t get a ‘valid’ stamp when using extensions not yet part of the HTML5 spec, like RDFa. Which doesn’t mean you can’t use them.

What is different between XHTML2 and (X)HTML5, for those interested in the X in XHTML, is mainly that the latter doesn’t introduce incompatible new elements to the HTML vocabulary, which would break rendering of your documents in all existing browsers when you send it out as a text/html document.
Rijk said on 3 July 2009 at 6:44 am:

@Mike Whitehurst: in what way is using XHTML better in providing a ‘good experience’ for IE users?
XHTML WTF – Jeffrey Zeldman Presents The Daily Report » CSSclip*News | CSSclip | web design inspiration and gallery said on 3 July 2009 at 7:10 am:

[...] Link to "XHTML WTF – Jeffrey Zeldman Presents The Daily Report" » [...]
Jonny Axelsson said on 3 July 2009 at 7:12 am:

This was pretty much inevitable for the last half decade. Either XHTML 2.0 would become irrelevant as a REC, or it wouldn’t become a REC at all.

XHTML 2.0 was destined to be the next generation HTML3.0, an idealistic W3C spec not fulfilling market needs. Apart from the excruciating process which is the W3C track, most of the good ideas in HTML 3 reincarnated in HTML 4 in a much better version (some even in HTML 5). Likewise the good ideas in XHTML 2, and there are a few, are bound to be taken up by HTML 5, or possibly HTML 6.

Furthermore the syntax-agnostic approach of (X)HTML5 makes a lot more sense. The strength of HTML, among the most successful languages around, never lied in the syntax, but in what it could economically express.
Xhtml ist tot said on 3 July 2009 at 7:36 am:

[...] – Link eintragen Xhtml ist tot, lang lebe Html. Nachrufe von: Erik Eggert, Peter Kröner, Jeffrey Zeldman 03.07.09 – 13:26 [...]
Hemebond said on 3 July 2009 at 7:45 am:

Great. Ditching what could have been a great, clean, extensible markup language for some browser-maker-driven mess with tags for every little thing you might want to put in a page; but not in a standardised way. Are you as excited as I am?
Mike Whitehurst said on 3 July 2009 at 8:09 am:

@Rijk: I’m referring to cross-browser support. We’re at a point now where we have good consistent support accross every browser. We’ve fought hard for it. Introducing a new specification means we have to revisit the process of browser-specific markup/code – to find suitable workarounds where the feature support “isn’t quite there yet”.
stefan said on 3 July 2009 at 8:27 am:

Mike Whitehurst said on 3 July 2009 at 3:59 am:

There’s a long time to debate this, IE6 doesn’t support HTML5. IE8/F/Safari have partial support. So we’re back to that again. We can’t actually use HTML 5 yet because we want to provide a good experience accross the range of browsers our visitors have. XHTML will remain the best way to do this for many years to come.

Amen.
Anonymous said on 3 July 2009 at 9:14 am:

[...] [...]
Jeffrey Zeldman said on 3 July 2009 at 10:00 am:

XHTML 2.0 was destined to be the next generation HTML3.0, an idealistic W3C spec not fulfilling market needs.

Agreed. Many of us have said this for years. While speaking to the W3C in Spain, I met the main dude behind XHTML 2, a lovely, gracious, brilliant gentleman, and was unable to convey to him the earthbound, practical needs of folks like you and me. Likewise, he was unable to persuade me that “IMG” needed to be fixed or that a shift to elements no browser supports was a wise move when IE was still having trouble supporting “OBJECT.”

(I’m sure I’m oversimplifying our conversation over a lovely dinner. I spent half the meal chasing our then two-year-old daughter around the restaurant.)

Even in the first edition of Designing With Web Standards, I expressed concern with the direction of XHTML and the notion that a more abstract and variable XML was the markup language of everyone’s future. On the other hand, XHTML 1.0 was and is a stable, clean, and logical standard markup language that every browser supports (with the exception of MIME type in IE). At the time of the first and second editions, even with HTML 5 in the wings, XHTML 1.0 still made sense to me as a solid toehold for web development for years to come.

We’re at a point now where we have good consistent support accross every browser. We’ve fought hard for it. Introducing a new specification means we have to revisit the process of browser-specific markup/code – to find suitable workarounds where the feature support “isn’t quite there yet”.

Yes and no.

There are two issues with HTML 5. One, brilliantly expressed by John Allsopp in Issue No. 275 of A List Apart, runs like this:

We don’t need to add specific terms to the vocabulary of HTML, we need to add a mechanism that allows semantic richness to be added to a document as required. In technical terms, we need to make HTML extensible. HTML 5 proposes no mechanism for extensibility.

The W3C’s decision to dump XHTML 2 (while inevitable given its purist outlook and lack of uptake in the market) may foist HTML 5 on us before the community and the builders of HTML 5 have come to anything like solid agreement about the strategic and specific tactical direction of the specification vis-a-vis semantics, elements, extensibility, and so on.

The other problem is that of browser support. And here’s where I net out at “yes and no.” Yes, we’re going to have problems, especially with IE, and didn’t we just spend more than a decade working through all that? But also, no, we shouldn’t fear or attack new specifications on the grounds that browser makers may resist. The web is changing, and specifications must advance. (Not everybody can do everything via JavaScript libraries.)
Noel Jackson said on 3 July 2009 at 10:17 am:

WordPress is most definitely not biased towards XHTML, HTML 4.01, or HTML 5 – code your templates however you feel you’d like to. :)
Faruk Ateş said on 3 July 2009 at 10:23 am:

The web is changing, and specifications must advance. (Not everybody can do everything via JavaScript libraries.)

I would dare go so far as to say that we will all be dependent on JavaScript libraries for emulated functionality for a lot longer than we all would like to be, if we don’t think pragmatically about situations like this one.

If the status quo is satisfying enough to you, then by all means, attack such efforts. But to me, the status quo is far from satisfying enough, and I’m not going to be satisfied until every single browser with 1% market share or more is capable of doing what Safari can already do today.

So for me, it’s very clear and simple: (more) change is needed.
Orionvega said on 3 July 2009 at 10:27 am:

Never a dull moment.
Renato Feijo said on 3 July 2009 at 10:33 am:

What worries me most is the obvious disconnect between HTML5 and CSS3 specs. To the , I would add the necessary interoperability with styling mechanisms that ensure acceptable accessibility across multiple devices and media, and provide more consistent ways for designers to layout content.
Renato Feijo said on 3 July 2009 at 10:36 am:

Sorry, that should read “To the ’strategic and specific tactical direction of the specification vis-a-vis semantics, elements, extensibility’ I would add…
XHTML2 Is Dead. In Similar News, Water Is Wet, Fire Is Hot. | FarukAt.eş said on 3 July 2009 at 10:40 am:

[...] guy who once wrote The Case for XHTML, but Jeffrey Zeldman captures my view on this perfectly in this one single paragraph: Designing With Web Standards, I expressed concern with the direction of XHTML and the notion that [...]
Rijk said on 3 July 2009 at 10:40 am:

The main problem with the complaint “HTML 5 proposes no mechanism for extensibility.” is that it is only partially true. If you use the XML serialization, you can have the same extensibility as XHTML1 and XHTML2 offered – it is NOT a step backwards as some try to imply. But such extensibilty isn’t all that useful, and requires messy namespaces.

People who want that kind of extensibility in HTML5 (in the text/html serialization) either want direct support for namespaces, or something much like it and just as messy, like CURIEs. Or you give up and just try to get plain direct recognition for your favorite vocabulary in the HTML spec (SVG, MathML).
Joao Silva said on 3 July 2009 at 10:44 am:

As interesting discussion as it is what makes this good news?

We as web developers moved from html 4 to xhtml1.0 and now we will move back to html5, now i know that is good for the development of the web that finaly there is a goal and a light in the end of the tunnel and that w3c are focus on only one language for the future.

However how long do you think will take for the browsers to support this language as well as browsers such as IE6 or IE7.
Yes IE 7, i dont think they will change the rendering engine to accomodate html5 if we take in consideration their attitude with outlook 2010.

I don’t know but any progress is a progress so happy days.

My cent thanks
J Cornelius said on 3 July 2009 at 10:57 am:

Regardless of our pursuits it will be years, potentially decades before W3c working groups, browser vendors, and software vendors all agree on a single spec.

In support of XHTML, extensibility is crucial if we are to remain open to what the future brings. Who knows if the semantic names given to HTML5 elements will still be relevant in the future? Perhaps an exciting new development renders them useless. The novelty of article and aside tags is endearing. But these and the other new semantically named elements provide no real value over div class=”article” or div class=”aside”.

If a page doesn’t validate, should it render? I think not. Programatically, strict adherence to the spec and forcing markup to comply is good for everyone because it removes ambiguity. So long as we need to consider improperly nested, unclosed, or other invalid markup the tools we use and, consequently, the web as a whole will suffer for it.
Pedro said on 3 July 2009 at 11:03 am:

I think it would be time to actually rethink the whole HTML concept. Think about it: it means HyperText Markup Language. Hyper TEXT, people.

Isn’t it time to come up with a broader way of describing content on the web and the relationships between these pieces of content? Bottom line, why keep expanding HTML to encompass new things and not start from scratch in a more generic way?

Just my 2 cents.
Rijk said on 3 July 2009 at 11:18 am:

@Mike: is XHTML somehow more cross-browser compatible than HTML? Never noticed that. I don’t get the point of your first message at all :)

If you mean that we better stick to the vocabulary defined in HTML 4 and also used in XHTML1, then yes, that’s the save choice, and it isn’t that bad…
Best of the Week 7/3 - Creattica Daily said on 3 July 2009 at 11:24 am:

[...] XHTML WTF [...]
Greg Hemphill said on 3 July 2009 at 11:28 am:

As to the future, it’s looking more and more like HTML5 will be the one spec to rule them all. Thus, if there are things missing from HTML5 that you feel passionate about… you no longer can ignore their omission by thinking you’ll just use XHTML2 instead. You want a feature, work on getting it in html5.

At the end of the day I hope this helps get HTML5 finalized, because it appears MS isn’t interested in adding HTML5 goodies (or CSS3) until the specs are finalized or near to it. By having only one spec to focus on hopefully thing will move faster.
Stoneship > Comments on Jeffrey Zeldman’s XHTML WTF said on 3 July 2009 at 11:39 am:

[...] are a few interesting comments on Jeffrey Zeldman’s XHTML WTF article: I don’t understand why this is happening. Didn’t XHTML fix all the inconsistencies in HTML? [...]
Jeffrey Zeldman said on 3 July 2009 at 11:51 am:

I’m not going to be satisfied until every single browser with 1% market share or more is capable of doing what Safari can already do today.

Agreed. But you and I may not be satisfied for decades to come. Let’s hope the next generation of WaSP is up to the task of working in partnership with browser and tool makers and designer / developers. (Or that Microsoft abandons Trident for Webkit.)
Mike Whitehurst said on 3 July 2009 at 12:06 pm:

@Rijk: Poor choice of words on my part. The obvious point is that we will have to wait a while until our favourite web browser has suffifienct HTML5 support.

I was really thinking about (but not explaining properly) the accessibility issues that HTML5 will inevitably create. XHTML & CSS are all about seperating content from presentation. I remember 3 petri-dishes and some wind-up dinosaurs :o)

I haven’t spent any time reading the HTML5 specification yet, but I’m certain there’ll be visually impaired, hard-of-hearing, handheld using, printing (and so on) people who (if they knew what it was) would be quite worried about the effects that HTML5 will have on them.

I wouldn’t say I’m resistant to change, but at the moment there are a lot of questions that I don’t have answers to. I’m sure ALA will provide most of them in good time, but right now I’m a bit “WTF” and that scares me.
Nathan said on 3 July 2009 at 12:28 pm:

There seems to be a lot of dis-information about HTML5 out there. It’s not a step back from XHTML 1.0, but a natural progression. XHTML 2 is the one being given up on, not XHTML in general. XHTML 5 lives on and will be very useable.
Berserk said on 3 July 2009 at 1:12 pm:

The good news is, according to this comment at html5doctor.com using the xhtml-y syntax for links without a natural end-tag (such as <br> being written as <br />) is valid in html5. With Bruce Lawsons follow-up comment I can only assume that this is correct.

That makes me happy (and I wouldn’t miss xhtml2 any way, it always seemed like utopia).

Now, if only they could agree on one codec for <video> html5… To compare that failure (not a failure of Hickson, but of Mozilla, Apple and MS) with the fact that <img> doesn’t specify which format images should be in is a moot point. A correct comparison would be if MS required .gif, FF .jpg and Safari .png *for the same image*. Unless this is fixed I’m afraid <video> is dead and we are stuck with Flash movies for a long time. (Sorry for going off target…)
Destry Wion said on 3 July 2009 at 3:47 pm:

I somehow think RDFa is going to get a lot more attention now, likely where xhtml folds into html5. Interesting to see what unfolds during the folding.
Marc said on 3 July 2009 at 4:21 pm:

As far as I’m concerned, about dang time they officially killed XHTML2 and focused on what has been for a while the obvious future direction of the One Spec To Rule Them All, so far as leading-edge web markup goes. And I say this as someone who currently uses XHTML1.0 strict for everything I do, because, ignorant though I usually am, I’ve been paying enough attention to understand where things should be going and indeed where they thankfully are going.

I don’t mean to be insulting, but the number of comments here that come from people who are obviously completely clueless about what HTML5 really is, or, for that matter, what XHTML1 really is and isn’t, is disappointing to say the least. Seriously, if the Powers That Be hadn’t jumped on the XML bandwagon, what we call XHTML, sans a tiny bit of XML-ness that never went anywhere, would have been labeled HTML4.5. Or even HTML5, if you wanted to go up a whole version number, in which case we’d just be discussing HTML6 today and there’d be no confusion at all.

It’s not like HTML5 hasn’t been in the works for months already, and XHTML2 hasn’t been effectively dead for… well, years, really.

Me, I’m looking forward to it hugely. Now, to all of those panicking above: go read up on HTML5 and stop saying things that make you sound ignorant and silly. I’m sure y’all are plenty smart when you’re paying attention.
rafael's.blog - Prace nad XHTML 2 zostaną zerwane z końcem roku said on 3 July 2009 at 4:40 pm:

[...] Jeffrey Zeldman – aka Godfather of Webstandards Dave Shea – aka Creator of CSS Zen Garden [...]
Are we going to unite or wave? « InteractionLove said on 3 July 2009 at 5:46 pm:

[...] last month we’ve been overwhelmed by news of what our online future is going to look like, XHTML is apparently dead, Firefox 3.5 is going to save web-typography and last but not least both Opera and Google unveiled [...]
Thinkyhead said on 3 July 2009 at 6:00 pm:

Now for the next editions of your books, search and replace “XHTML” with “HTML5″ and you’ll be all set. Am I right??
Jeffrey Zeldman: XHTML WTF said on 3 July 2009 at 7:21 pm:

[...] Jeffrey Zeldman: XHTML WTF. Reading the comments, it’s scary how many people are totally ill-informed about HTML5 and XHTML5. Posted 4th July 2009 at 1:22 am [...]
Jeffrey Zeldman: XHTML WTF | dv8-designs said on 3 July 2009 at 7:43 pm:

[...] Zeldman: XHTML WTF July 3rd, 2009 admin Leave a comment Go to comments Jeffrey Zeldman: XHTML WTF. Reading the comments, it’s scary how many people are totally ill-informed about HTML5 and [...]
NewsLand - [Info] XHTML 2's dead, baby said on 3 July 2009 at 7:48 pm:

[...] to be. 1999: XML is the light and XHTML is the way. 2009: XHTML is dead?kind of." Link: <http://www.zeldman.com/2009/07/02/xhtml-wtf/> — d. — Newsgroup robomoderato – gli articoli vengono approvati [...]
Kam XHTML WTF, kam HTML5 FTW « Rimantas Liubertas said on 3 July 2009 at 8:08 pm:

[...] Užtat nestebina, kad Zeldmanas į W3C sprendimą leisti XHTML2 mirti negimus savotiškai sureagavo: XHTML WTF. [...]
Faruk Ateş said on 3 July 2009 at 9:56 pm:

Agreed. But you and I may not be satisfied for decades to come.

I suspect the same, but that’s why I’m all the more eager to try and make things move forward faster. Modernizr is one such effort.

Plus, I’m willing to fight this battle for decades to come. In the end, the Web and all of its users will win.
Doug S. said on 3 July 2009 at 10:36 pm:

At least there’s (X)HTML5.

Personally, I was much more excited about XHTML2 than I ever was about HTML5. H tags instead of H1-6? More intelligent structure? All the good things and very few of the bad ones.

Now XHTML is gone. I’ll probably stick with XHTML 1.0 Strict until IE8 becomes the de facto IE browser. Then I’ll look into (X)HTML5.
�Ϥƥ֥˥塼�� said on 3 July 2009 at 11:16 pm:

[...] [...]
Ron McElfresh | McSolo › Opinions › A few more comments on HTML, XHTML, HTML 5 said on 4 July 2009 at 12:13 am:

[...] HTML WTF, Jeffrey Zeldman says “XHTML is dead—kind of.” The dozens of comments following [...]
Henry said on 4 July 2009 at 12:28 am:

I’m hoping it doesn’t take long for ‘most’ user to get modern web browsers!
bruce said on 4 July 2009 at 4:36 am:

@beserk

Yes, it’s true that trailing slash for br, img etc is optional for HTML 5. More in my valediction to XHTML 2 http://www.brucelawson.co.uk/2009/goodbye-xhtml-2/
Jeffrey Zeldman said on 4 July 2009 at 7:20 am:

From Bruce’s post:

What does this mean to you? Nothing. Your current XHTML 1.x sites still continue working (except in IE, if you serve them as XML rather than HTML).

Yup. Moreover,

HTML 5 took some good ideas from XHTML 2 – the idea of deriving the “level” of a heading from its context, although it preserves using h1…h6 for backwards compatibility rather than a generic h element. XHTML 2 allowed “href anywhere” so anything can be a link, and HTML 5 has a similar idea, although it preserves backwards compatibility by allowing the a element to surround block-level elements. The XHTML 2 element nl for navigation list is doubled in the HTML 5 nav element that wraps a ul or ol. The main side-effect of the end of XHTML 2 is that its resources will now be given to HTML 5.

In other words, those who looked forward to using XHTML2’s more abstract approach to markup will be able to do so (but in a more backward-compatible way).

HTML 5 was initiated (and thus supported) by browser makers and was intended to move fast-tracked W3C-style standards activity with the fast-moving app-based web. Once the W3C adopted it, its eventual adoption as a working spec was inevitable.

And, really, the abstract direction of XHTML 2 and its lack of connection to a changing web doomed not only XHTML 2, but XHTML. Those of us who came to admire the clarity of (most of) XHTML’s markup rules were unable to persuade the brains behind XHTML to stay on track with web development rather than reinvent markup in ways some browsers might never support. HTML 5 arose in part because XHTML 2 looked like a derailment of the future of the XHTML we had converted to and evangelized.

XHTML 2 was never going to pan out, thus there was no urgency to learn about it, and we enjoyed a period of stability as browsers came up to spec.

XHTML 1.x will be a stable rec forever, and we can keep using it forever if it meets our needs. The urgency at this point is to familiarize ourselves with HTML 5 and voice concerns, if any.

Limited HTML 5 already works in browsers; it’s used in aneventapart.com and plenty of other sites.

Now for the next editions of your books, search and replace “XHTML” with “HTML5″ and you’ll be all set. Am I right??

Heh-heh. Wrong as Prohibition. Thanks for playing.
Jeffrey Zeldman said on 4 July 2009 at 7:57 am:

From 2003:

If XHTML can be branded so as to create the association with validation, separating structure and presentation, accessibility, and other techniques that have worth in their own right, then maybe they can all get traction together. But that’s not a technical argument, it’s a social one, and anyone who claims that these loosely coupled concepts are really tightly coupled is either misinformed or lying.

—Dive Into Mark

2 Jul 2009 3 pm eastern

XHTML WTF

XHTML 2 Working Group Expected to Stop Work End of 2009, W3C to Increase Resources on HTML 5

116 Responses to “XHTML WTF”

Speak up!

The Deck

Job Board

Here

There