By Craig Buckler


By Craig Buckler

RIP XHTML 2The W3C has announced that XHTML 2 will be dropped when the Working Group’s charter expires on 31 December 2009.

The first W3C XHTML 2 draft specification appeared in 2002 and it was last updated in 2006. The project was an ambitious re-working of the language of the web. It sought to address the inconsistencies, banish obvious presentational tags, and implement clear and concise mark-up. Some concepts were excellent, e.g.

  • Any tag could be made a link by assigning an href attribute (<a> was retained, but would have become redundant).
  • Nested <section> tags controlled the document structure and hierarchy. A single <h> tag could denote headings at any level — the author would not need to manually manage <h1><h6> tags.
  • A new <nl> navigational list tag was introduced.
  • RDFa allowed semantics to be annotated in the mark-up.
  • The language was modularized so extensions could be added.

Several XHTML 2 recommendations have have been ported to the HTML 5 specification.

Unfortunately, XHTML 2 was possibly too revolutionary. It was too different, too strict, offered little backward-compatibility, and was criticized for not supporting real-world coding practices. The specification was never completed (few are) and no major browser vendor made a serious attempt to implement the standard (IE still does not support any flavor of XHTML served as XML).

How will this announcement affect XHTML developers? It won’t — not unless you have been developing XHTML 2 solutions in the faint hope of eventual browser support. XHTML 1.0 and 1.1 are still an implemented standard in most browsers and XHTML 5 will support the strict syntax, lowercase tags, and trailing slashes you want to use.

The death of XHTML 2 has one major benefit: additional resources can be allocated to HTML 5. Browser vendors and developers can now work toward a single unified specification. We hope.

See also:

  • jonrawlins

    To be quite honest, I don’t think there are a lot of people even using XHTML 2, purely for the reasons you stated, on the most part it’s generally good practice to stick with XHTML 1.0 and 1.1, until HTML 5 is all the rage.

    Thanks for keeping everyone updated as usual :D


  • graphicmist

    Can’t wait for html5 to be supported by all browsers…

  • worthless

    XHTML 5 will support the…

    I think you mean HTML 5.

  • V1


    No, XHTML 5 as that will become the strict version of HTML 5.

  • @worthless
    No – I mean XHTML 5.

    There will be an XML serialization of HTML 5 which must be served with an XML MIME type. That means no IE support (yet).

  • Chris McKee

    Shame I was looking forwards to the xhtml 2 standard a damn sight more than XHTML 5; thank god I wont need to touch it for about 5 years thanks to browser legacy.

  • This is good news as far as I’m concerned – 2 standards defeat the point of standards (“Standards are great, lets have more of them!”), and HTML 5 is by far the more practical of the 2 standards. Now more efforts can be put into improving that rather than developing 2 separate standards apart from each other.

  • Pingback: W3C to Increase Resources on HTML 5 «()

  • Sski

    As a hobbiest in HTML markup, the constant changes are becoming difficult to keep up with. IE8 already broke one of my sites even after it validates at W3C. Maybe I’m just getting old, I guess I’ll just have to revert to “page makers”.

  • @Sski
    Validating at the W3C validator only guarantees that YOU are following the rules as defined by the language. That doesn’t mean that what you write will produce the expected results, since you can never be sure that BROWSERS are going to follow these rules. IE had (and still has…) a lot of bugs in it’s implementation of various standards. IE8 fixed a lot (or should I say “most”) of the CSS 2.1 related bugs, misinterpretations and lackings from it… and some ECMAScript and DOM related ones too. You wrote your code with the assumption that IE had the correct behaviour and/or that it will always have a certain bug or behave in a certain way that appears right. If you’re a good web developer, the first is probably not true for you, but the latter is true for most of us… myself included.

    As for XHTML 2.0… I was really looking forward to it, for the reasons mentioned in the article, plus the ability to make any element an image/object with the “src” attribute, and use the contents as alternate content. Damn you backdraw compatability!

    I really wish people would stop calling the XML serialization of HTML 5 “XHTML 5”. It’s confusing, to say the least… I mean, one day (hopefully), XHTML 2.0 will be back, it will reach versions 3, 4… and then what?
    I’d call it “HTML 5-XML” or “X-HTML 5”, or simply “HTML 5 as XML”.

  • @Sski
    HTML standards move very slowly. There’s a lot of talk about what should/not be added and browser vendors do their own thing, but completed W3C specifications are rare. The HTML 4.01 standard was published in 1999 and XHTML 1.1 was published in 2001 (and IE still doesn’t support it).

    HTML 5 is a draft, there are several raging disagreements, and full browser support is possibly many years away.

    Remember that valid code does not necessarily mean that it will work! Validators can only go so far and, even if your HTML is fine, your CSS or JavaScript could be broken…

    Many people speak perfect English but still make no sense!

  • @Craig Buckler

    Many people speak perfect English but still make no sense!

    I think the word you’re looking for is “politician” :-D . OK, it’s not just them, but still…

  • Alok

    really it is gr8 post and informative as well.

  • It has ceased to be. It is bereft of life. It is an ex-specification.

    I just want to say how much I love the Monty Python reference in the excerpt on the front page. Not enough of that these days, that’s for sure.

  • @boen_robot
    XHTML 5 is a confusing term if you understand the history of the specifications, but it’ll soon become commonplace. XHTML 2 is unlikely to be resurrected so HTML 5 and XHTML 5 will simply evolve along the same path. Ultimately, that’s a good thing even though some great XHTML 2 features have been dropped.

    We can worry about new names and numbering conventions if a competing XHTML-like standard ever appears!

  • @Matthew Magain
    You’ll also be pleased to know that the article image has been grabbed from “Meaning of Life” … “It’s a Mr Death — he’s come about the reaping”.

    Perhaps XHTML 2 had the salmon mousse?

  • Just to round it off: in regards to XHTML 2: “Well, you’re dead now, so shut up!”

  • Pingback: Ken Snyder()

  • Ken: You could just use XHTML 5….

  • momos

    Still, it seems really weird to drop XHTML 2 as a future standard. I alone know at least 8 people that were eagerly waiting candidate recommendation status.

  • OK I am confused, is HTML 5 actually XHTML now or is XHTML 5 != HTML 5?

  • Pingback:

  • Pingback: Libology Blog » Trends in the WebSphere()

  • @halfasleeps and Ken Snyder

    OK I am confused, is HTML 5 actually XHTML now or is XHTML 5 != HTML 5?

    HTML 5 is the new proposed standard. It’s an evolution of HTML so, yes, you don’t need lower case tags, closing slashes, etc. (however, HTML5 does support XML conventions if you choose to use them).

    XHTML 5 is an XML serialization of HTML 5. In other words, it’s the same HTML 5 syntax, but you must follow strict XML rules such as lower case tags, closing slashes, etc.

    You can use which ever you prefer. However, XHTML 5 must be served with an XML MIME type which means it’ll fail in IE. It won’t be possible to serve it as text/html (as you can with XHTML 1.0) because you’ll be effectively be reverting to HTML 5.

  • Pingback: links for 2009-07-07 | burningCat()

  • momos

    This actually makes me think about VHS vs Betamax. VHS won because the porn industry invested in VHS, need I say more…

  • This actually makes me think about VHS vs Betamax. VHS won because the porn industry invested in VHS, need I say more…

    Yes, you do need to say more, how does this relate at all? Although the only reason the porn industry invested in VHS was because Sony restricted what content was allowed on Betamax, so no adult material, and they lost out because of it.

  • Shame I was looking forwards to the xhtml 2 standard a damn sight more than XHTML 5; thank god I wont need to touch it for about 5 years thanks to browser legacy.

    I agree. I wouldn’t mind seeing RIP HTML 5 appear tomorrow. :) I have no beef with HTML 5 other than it isn’t really HTML but some bastardization of a relatively decent standard. Rename it AppML or something, and I’m good (but still won’t use it).

    At this point, I’m not really sure why so much effort is going into HTML. It’s a decent, workable standard that plays fairly well with browsers. HTML 5 is a spec, not a standard, since no one actually implements it all. Besides, we can pretty well display whatever we want using CSS or XSLT or a server-side DSL that outputs HTML + whatever. I guess I don’t really see the point. It’s like trying to make the best buggy-whip when people are now driving automobiles.

  • OK so it is more like (x)(html5) rather than (xhtml)(5)

  • momos

    Yes, you do need to say more

    Betamax was technically superior to VHS. But hat some strict rules like: “no adult material”

  • Pingback: SitePoint Podcast #19: Compuwhatnow?()

  • Pingback: XHTML 2 Hits the Garbage Can :

  • porno

    hello.. thank you for this article.

Get the latest in Front-end, once a week, for free.