Contact Us

XHTML died alone, the semantic web is next

Posted on 20/7/09 by Felix Geisendörfer

Story time:

Thursday, July 2nd 2009. Officials announce the death of XHTML2. Multiple suspects, including HTML4, HTML5 and XHTML1 have been taken into pre-trial custody.

The investigation is difficult. XHTML1 seems to have no motive as XHTML2 was his son. The father can barely speak about his loss without bursting into tears. However, he is a very strict man and everybody knew he had extremely high expectations in his son. Could his bad temper have killed XHTML2 for making a simple mistake?

HTML4 however has a motive. He was once the big star in town. But then came XHTML1 and took that fame away. Every theater, including "the Fire Fox", "the Opera" and "the Web Kit" loved XHTML1 and told HTML4 it is time to retire. Not only that, many critics also saw great talent in his son XHTML2 who was competing with HTML5. So did HTML4 kill XHTML2 out of revenge, to free the way for his own son?

The prime suspect of course is HTML5. He was a direct competitor to XHTML2. The two had very different professional opinions. XHTML2, just like his father, was a strict perfectionist. HTML5 however is often characterized as a forgiving, practical performer. But investigators have found that HTML5 had the support of many cutting-edge theaters since at least 2004 and had no reason to fear XHTML2's competition. Did HTML5 feel he had to kill XHTML2 to simply destroy his strict philosophy?

3 years later. Investigations on the XHTML2 case have long been closed. XHTML2's death was classified as suicide. The media said XHTML2 could not gain the support of any major theater and killed himself out of desperation.

Now a new case is erupting the news landscape. HTML5 had a step brother, XHTML5, and his body was just found in a motel room. He was the new leader and last hope of a movement called "the semantic web". Probable cause of death - microformat overdose.

XHTML5's death marks the end of the movement. He and his followers thought they could make performance art accessible to a new audience of robots and intelligent machines. Those machines were very hungry for information, but could not understand performers like HTML4 and HTML5. "The semantic web" thought that by enforcing the strictness invented by XHTML1 and through the heavy injection of microformats they could perform at a level of perfection that could bridge the gap between the visuals and the underlaying information.

From early on, the "semantic web movement" was very violent. Non-supporters were labeled as reactionists. The standards developed by the movement made them totally blind for reality. The biggest theater, known for its greedy operators and huge audience, "the Internet Explorer" simply ignored the movement.

But then came JSON, a theater janitor who would permanently transform the stage. JSON himself was raised by JavaScript, a class of machines invented at the old "Netscape" theater. The JavaScript machines were originally only used to do basic lighting and other low-level work at the theaters and nobody really understood their true power. But through JSON they were able to communicate with other machines. Machines who previously only talked to XHTML1's grandfather - XML.

JSON just happened to be much more eloquent than XML, and his slang became quickly popular among a new generation of machines. And he became friends with HTML4 and HTML5. They could not understand each other directly, but the JavaScript machines were able to express JSONs thoughts to HTML4 & 5. So HTML4 & 5 realized that JSON could take over XML's job as a playwright.

And so the stage was slowly transformed. Plays were written by JSON who then performed them for the machines. JavaScript and other machines enabled HTML4 & 5 to perform the plays in their unique ways that were still more attractive to humans.

The XHTML family died because it wanted to write and perform for humans and machines alike. But the machines were not ready for it yet and wanted their own performances. Accepting this truth ultimately helped the humans who could now enjoy more entertainment & knowledge than ever, thanks to their automated friends.

=== The End ===

If you did not like the performance, entrance fees are unfortunately non-refundable. But feel free to release your grief over the death of XHTML in the comments.

-- Felix Geisendörfer aka the_undefined


You can skip to the end and add a comment.

Éber Freitas Dias said on Jul 20, 2009:

Haha, this was fun!

Kalileo said on Jul 20, 2009:

XHTML strict will live forever!

Mardix said on Jul 20, 2009:

Very nice story. I cried at the end.

tomo  said on Jul 20, 2009:

I am missing the right german punctuation here, but I think you'll get it:

You are an uber-geek!!


Neil Crookes said on Jul 20, 2009:

Completely agree, in the future, webpages will not be written in html, instead we'll expose the content in JSON and use JavaScript to add it to the DOM... ;-)

Felix Geisendörfer said on Jul 20, 2009:

Neil Crookes: Right, except I don't think JavaScript will not be the only middle-layer between data and presentation. PHP, Ruby, etc. will still be generating tons of HTML. But I do think most of the web will accept JSON as a format for publishing data in the long run.

NOSLOW  said on Jul 20, 2009:

@Felix: gimme a
! :P

@Neil: ExtJS is doing that now ;) Beware though, HtmlHelper and FormHelper will become strangers.

Jake Rocheleau said on Jul 21, 2009:

Really interesting read, found this very informative as a front-end web developer. Keep up the great posts!

Igor Goldkind said on Jul 22, 2009:

I think rumours of the demise of the Semantic Web are a little premature. The movement has to do with creating a more efficient architecture, not one particular coding solution. Read more

Felix Geisendörfer said on Jul 22, 2009:

Igor Goldkind: Right. Unfortunately I have found no correlation between the words XML and efficiency so far. What am I missing?

Sergey Chernyshev said on Jul 22, 2009:

I think I'll second Igor's comment plus I have not found any correlation between XML and Semantic Web either - why are you referring to Semantic Web at all, I just didn't get it.

Nate Abele said on Jul 23, 2009:

Sergey Chernyshev: the point is that JSON is taking over the job that people thought embedded microformats (i.e. the whole point of the "semantic web") would do. This is obvious from the text.

Sergey Chernyshev said on Jul 23, 2009:

Ok, if you think that whole point of "semantic web" is embedded microformats, then no questions about it.

It's a good idea to learn what "semantic web" is before reading/writing such articles.

Sorry for not being polite.


Nate Abele said on Jul 23, 2009:

Sergey Chernyshev: Pardon the pun, but now you're merely debating semantics. ;-) Sure, in and of itself, the idea of the semantic web is about some abstract, high-level concepts on document architecture; great. But the point, the stated purpose and end-goal for which the concepts exist are thing like being able to re-purpose content embedded in a document through the use of structure and meta-data.

Microformats are a practical application of this, and to try to say that they're not, you'd only be kidding yourself.

Sergey Chernyshev said on Jul 23, 2009:

No, what you're referring to is latest development to embed structure directly into documents. You're saying Microformats, but it's actually a term for something different, but I can assume you were probably referring to RDFa. In any case, "death" of XHTML2 has nothing to do with Microformats or RDFa and of course has very little to do with Semantic Web.

It just bedazzles me how people can write something like this without giving it enough research. Article sounds very intriguing though and gets lots of attention - almost as much as popular articles about collider.

Felix Geisendörfer said on Jul 23, 2009:

Sergey Chernyshev: Lets all stop for a second and explain what you mean when you say "semantic web". I'll go first. This is what Wikipedia says, and this is what I mean when I talk about the semantic web:

The Semantic Web is an evolving development of the World Wide Web in which the semantics of information and services on the web is defined, making it possible for the web to understand and satisfy the requests of people and machines to use the web content.

My article is merely saying: Hey this is stupid. The HTML-family is about the worst format I can imagine for expressing data and relationships of data. So lets stop trying to mess it up in all kinds of ugly and xml-ugly ways, accept the fact that we are screwed and lets find an alternative data format for machines. If serving different content to machines than you'd serve to people fits into the concept of the "semantic web", hey then I am its biggest supporter. But if it does not, then please lets move beyond XML and all its ugly sisters and appreciate JSON as a true child of the web and lets serve this to our machines friends.

Sergey Chernyshev said on Jul 23, 2009:

Felix, I think you said it all - if you notice, Wikipedia excerpt that you posted doesn't say that requests people and machines are the same, meaning that embedding structured data into HTML is just one of the efforts - for ages, RDF was served separately from HTML, in different requests and in many formats (RDF/XML and N3/Turtle are the most popular).

If you don't like XML, but so passionate about JSON, you should also know that RDF is being served as JSON too - RDF is not a format, it's data model.

All that being said, Semantic Web is not only RDF, but more - if you're interested in details, you can start attending one of many Semantic Web Meetups: (just don't attend only one event and judge by your first impressions).

Felix Geisendörfer said on Jul 23, 2009:

Sergey Chernyshev: Ok, maybe I was absolutely wrong. Maybe the "semantic web" is not going to die because "semantic html" is stupid, but because the "semantic web" has become an utterly meaningless catch-all term for exchanging data over the internet.

Nate Abele said on Jul 23, 2009:

Sergey Chernyshev: Also, you keep hinting at what "semantic web" means in order to justify your position that we don't understand it, but so far you haven't been able to succinctly explain what it *actually* is. This is usually a sign that someone doesn't understand themselves.

Sergey Chernyshev said on Jul 23, 2009:

Guys, I'm sorry if I sound like I don't give away too much, it's just that I spent a lot of time getting the glimpse of this and just can't put it into comment section here. If you're really interested, come to some conferences or local events like Meetups I mentioned.

Jon Adams said on Jul 24, 2009:

@Sergey: Sounds spooky. What is this, some kind of cult?

There aren't any scheduled meetups in my area, anyhow. I live in Somalia.

I was under the impression that the whole "semantic web" movement was as Felix described/implied, myself; Trying to make data simultaneously human and machine readable. Last I read into it I walked away thinking it was a noble idea but not very practical. If semantic web isn't that, and is a whole lot more, why not at least try and explain it?

And, again, as Felix mentioned, if it's just this catch-all term for exchanging data over the internet (i.e. sending different requests for human vs machine readable data) then... ok.. that's just a label. that's just the way it is. that's what we're going to do, and we're on the same team.

Igor Goldkind said on Jul 24, 2009:

semantic web definition:
1. An evolving extension of the World Wide Web in which web content can be expressed not only in natural language, but also in a format which can be read and used by software agents; therefore, permitting them to find, share and integrate information more easily.

It derives from W3C director Tim Berners-Lee's vision of the Web as a universal medium for data, information, and knowledge exchange.

2. A set of formats and languages that find and analyze data on the World Wide Web, allowing consumers and businesses to understand all kinds of useful online information.

—Defined in "The Semantic Web in Action", by Lee Feigenbaum, et al.;

Scientific American; December, 2007; page 65.

This is useful because it explains some of the confusion. Semantic web is a web of meaningfully linked data. That is the aspiration and there is a great deal of healthy argument about how to best achieve that aspiration. But it is important to keep in mind the evolving nature of the various approaches. What we refer to as Semantic web today may be very different from what we refer to it as tomorrow.

Bastian Albers said on Jul 24, 2009:

I still think the main idea of the semantic web makes a lot of sense: making machine-readable data that is connected to the site available while keeping the page human-friendly. But i have not seen a single good argument, why you should couple html and data tightly and not just de-couple it like you do with css and html. You would just need to spice your site up with some connected semantic file in the head and you would be fine without having to mess up your code. I do not see that not happening - the only question is how big it will get and what you can actually achieve doing it.

brian fidler said on Jul 29, 2009:

Well it appears the Sedative Web is alive and well ;)

Maarten  said on Aug 03, 2009:

Sergey makes a good point: the debate has nothing to do with the syntax (XML-based, JSON, ...).

*What* (for me at least!) the idea of "the semantic web" (I like the terms "web of data" and "linked data" better) is, is using the architecture of the web (links, identifiers, syntax, ...) to not only link pages with info (that you can traverse, read, link to), but to link the data that is in those pages.

*How* is a discussion for a different venue--just stop focusing on the quite irrelevant this-versus-that syntax-discussion. But you have a point re: clarifying the idea behind it.

*What can I do with it*? Some suggested recent reading/examples:

- (A lot of practical posts the last two weeks re: RDF + Google Visualisations)

- Nice intro for webdesigners

- (Mashup/browser-thingy, launced last week: "Live views of the web of data")

- (further reading)

Greg Reimer said on Aug 03, 2009:

While it may be fun to kick the dog when it's down, X(HT)ML is far from dead, for the simple reason that there currently exists a massive legacy of developer mind-share and infrastructure devoted to XML and all its quirks. In an ironic twist of fate, HTML5's "pave the cowpaths" pragma ensures the continued existence of XHTML well into the future. With time, perhaps the tag soup parsing model and things like JSON will completely displace XML, but it may be a *long* time before that happens, and who knows what will crop up in the meantime.

endris  said on Aug 05, 2009:

I think, you have to add to article's title "joke" or "funny story", because is no facts in it, some readers can make wrong conclusion in their heads :)

And why it is used term "semantic web" I can't understand. Nothing about it in article.
Semantic web is not about data formats (html, xhtml, xml, json and so on), it's about identifying data meaning.

Igor Goldkind said on Aug 05, 2009:

No: the Semantic Web is a set of formats and languages that find and analyze data on the World Wide Web, allowing consumers and businesses to understand all kinds of useful online information.

Anoop John said on Aug 06, 2009:

That was a very nice story. I normally don't get time to read through full blog posts but I liked the presentation style and read through :-).

Personally I feel that semantic web is not dead yet. I would say it is only RESTing. As the number of plays going around these theaters increase more people would just want to get the information behind the plays than the complete play themselves.

Parag Shah said on Sep 01, 2009:

I enjoyed reading your story very much.

HTML5 and Javascript can in many ways standardize the way rich web ui's are created. So the demise of XHTML may be followed by those of some rich client UI technologies for the web.

Michael Cheng said on Sep 11, 2009:

Ha ha... i like this.

VM  said on Dec 19, 2009:

Loved the article.

My Opinion:
"Semantic Web" will probably live on to torture a lot of innocent programmers. It will slow down the progress of technology. But it might perk up the economy, and many people might be able to make a good living out of it. :)

If it was up to me, I would disband W3C and start over. Let the people who actually use the standard create the new one. And keep all the theoreticians locked up somewhere. So the rest of us can stay sane and enjoy life. :) :)

This post is too old. We do not allow comments here anymore in order to fight spam. If you have real feedback or questions for the post, please contact us.