Misunderstanding markup

Well said!

Spot on with these comments, I’ve just commented on Zeldmans post that you should choose the right doctype for the job so it’s good to hear someone else thinking on the same lines.

On the ball as usual

I definitely prefer the strictness as per XHTML spec, so I guess I’ll be using XHTML5 — HTML 5 with optional strict style markup would just be odd.

On the ball as usual

I definitely prefer the strictness as per XHTML spec, so I guess I’ll be using XHTML5 — HTML 5 with optional strict style markup would just be odd.

You can have the same strict syntax in HTML. And using HTML you don’t have to rely on the lack of proper SGML implementation in your browser. That browsers can digest your broken HTML (this is what XHTML document is for HTML processor) does not mean that you should feed them with it. Zeldman should really quit that marketing game and stick to standards.

Thank you, possibly the first post on the HTML5 / XHTML2 saga that has made some sense.

So where’s the panic?

Thanks for explaining this.

XHTML, like Doctor Bruce, is how I learned. Good to hear I can continue to do what I do, and that HTML5 will also allow me to continue to do what I do.

The confusion surrounding the ‘death’ of XHTML has had a "the sky is falling" vibe recently, thankfully it seems, clear heads will prevail.

Yes, ditto: http://twitter.com/codedread/status/2442135522

Thank you for this writeup. Of the only small bit of reading on this subject I’ve done, you explained it very well and concisely. Looking forward to the future with more excitement now.

Sweet post. Personally I prefer the xHTML syntax, but I’m not going drone on about it. It’s bit like the whole toilet paper under or over preference.

Yeah. I don’t care what anybody says, but HTML’s traditional lax rules encourage badly formed code, and I believe that time will show that to be the truth.

I’ve been digging around for the life of my trying to find an absolute answer: They say the HTML5 spec allows trailing slashes on self-closing elements, but that they should be served as XML then. But is an HTML5 document with trailing slashes served as text/html invalid?

“XHTML 2 is dead. Long live XHTML …as HTML 5.”

A perfect summarization, thank you for taking the time to spell this out. I will push forward with the XHTML syntax in HTML 5.

Thanks so much for clearing the air a bit with this post Jeremy. An understanding of the relationships between the different specs is an extremely important piece of the conversation that has not been very prevalent until Zeldman announced that XHTML was dead, the comments started rolling in, and everybody who understands these things perfectly started rolling their eyes, disgusted at the amount of ignorance out there in the developer community.

Unfortunately, most developers that I’ve had the opportunity to work with only take time to get a very shallow understanding of the technologies they are working with, the core of course being HTML, probably because the consequences are not very severe if they don’t fully understand and implement the standard (most web clients are very forgiving). It’s perfectly reasonable to interpret the incremental version number (XHTML 2) as being an evolution of the XHTML 1 spec instead of a "fresh start". After all, that’s how it goes with software and people are used to that.

well done. this is one of the clearest explanations i’ve read so far. it has all seemed like a lot of chest-pounding and hair-pulling over something not very surprising or upsetting, so it’s good to hear some grownup thoughts on the matter from yourself, zeldman, and others. thanks for bringing things back down to earth.

Henri may have been being snarky, but the quote isn’t saying anything about "formats" from what I can tell. He’s saying that for most people who author web content, XHTML is just a buzz word. Most people serve "XHTML" with "Content-Type: text/html" which removes any benefits over HTML 4 (e.g. inclusion of other XML bits, like MathML, directly in the XHTML markup) and I don’t care to guess how many XHTML authors know how to configure a server to override the default of serving XHTML as HTML. To modify your analogy, it’s like pretending to be a Thai restaurant but only serving McDonald’s hamburgers.

The "requirement" for strict syntax is only enforced if XHTML is served as XML, and browsers parse XHTML served as text/html as HTML. The same ideologic use of strict syntax people have for XHTML can be applied to HTML. Just because you can use loose syntax doesn’t mean you HAVE to use loose syntax, and it’s best if you don’t because (until HTML 5) error recovery (e.g. from unclosed paragraph tags) isn’t always consistent across browsers because the W3C didn’t define how errors should be handled in HTML. The strict syntax in XHTML served as text/html is self-imposed just like HTML (because the browsers treat it like HTML).

The argument for my decision not to adopt XHTML boiled down to this: If you’re serving XHTML as malformed HTML (XHTML’s self-closing elements, for example, are treated as having invalid attributes until HTML 5 is implemented and they are treated as self-closing tags like in XML) and you can code HTML with strict syntax, why not just use HTML?

Serving XHTML as text/html is a stop gap that has lasted 9 years while web developers haven’t gotten to use the "X" in "XHTML." The promise in real XHTML is great, but it’s been held back by lack of support in IE (and probably the lack of web servers that check to see if a document is XHTML or HTML before they pick the Content-type to send to the browser for .html files). The real fate of XHTML, as far as I’m concerned, is whether a complete HTML 5 implementation will require real XHTML 5 support (i.e. XHTML served as XML), and whether the IE team plans to support XHTML 5. If HTML 5 can force IE to support real XHTML, I think XHTML can be revived as the best tool for most web designers needs. If not, I still see no reason to pick XHTML over HTML.

Thanks for explicitly mentioning that XHTML isn’t dead. Few of the well known web design celeb types are saying anything about XHTML 5, and posting comments saying that HTML 5 has an XHTML serialization doesn’t get the same attention.

Thank you for writing this.

Unfortunately, seems like many people forget or aren’t aware that there really are only 4 big things that differ between HTML4 and XHTML1. There’s really no excuse not to do those 4 things.

THANK you. I have read Jeffrey Zeldman’s post, and I’ve read the Unofficial Q&A by Henri Sivone. After all the comments on Mr Zeldman’s post, I was completely confused. All I know is that I learned XHTML because my OU and University of Exeter distance learning courses said to use it. Then I read about the mime-type issue, so in my limited understanding, I thought maybe I should start writing HTML. Then when all this started about XHTML2, I’m like, for shit’s sake, all I want to do is write to standards, but if all these people who know so much more than me are arguing about it, what chance do I have of knowing which to use?

Your post answered my questions. Again, THANK you.

Does this mean XHTML freed your mind or not?

http://adactio.com/articles/1109/

thanks for clearing this up! Now I can sleep again… ;-)

The more I read about this, the more I feel there’s a lot of wailing and gnashing of teeth over nothing.

I’m glad we (the web development community at large) are discussing this and are, for the most part, pretty happy about it. But I have to think we should be able to get over it pretty quickly.

I know of very few web sites that implemented XHTML correctly. I know of thousands of websites that are written using Frontpage etc. that don’t even use correct HTML.

I know personally the next site I build will be written in HTML5 using the skills I learned from using XHTML.

One ring to rule them all, and let that ring be HTML5.

"I have my own catch-all term for the combination of HTML, CSS, and JavaScript; I call it web development." - Well said.

Great post! I, for one, will continue to mark up my code with XHTML syntax. I can’t stand the sloppiness of "old" HTML, but I love the flexibility to do it either way in HTML5. I guess I fall into the category of developers who will be using X/HTML5 and I look forward to all the shiny newness that it will bring. Now, if we can just get them to move a tiny bit faster, so we don’t lose our standards 13 years from now.

Hi, I’m new to this field and I would like to learn HTML.

If I understand correctly, most future web sites will be written in HTML5 (served using the text/html MIME type) and just a few sites will use XHTML5 (served using the application/xhtml+xml MIME type). As a consequence, I will focus on HTML5.

You say that HTML5 can use both the HTML syntax and the stricter XHTML syntax, but since I’m starting from scratch I suppose I could safely ignore the XHTML syntax and focus on the HTML syntax (HTML 4.1/5). Why bother about XHTML syntax?

Unfortunately all modern introductory book/online resources about HTML are focused on the XHTML syntax. Where I could learn the pure HTML syntax without any reference to the useless and distracting (for my purpouses) XHTML syntax?

The confusion and misunderstanding is a result of reality distortion field that XHTML supporters created through recent years. XHTML served with text/html was always a lie that these developers wanted to believe.

Great article Jeremy. I think this will clear things up for many people. HTML5 will be here and in common use within a couple of years in my view, forget 2022!! and so it is time to start getting familiar with it. Using it in its purest form will probably involve rethinking how we currently markup our pages (that is the point after all) but until then as you say it is very easy to serve the page with an HTML5 doctype. We have to start somewhere right? lol

This was so very concise and helpful. Thanks for cutting through all the confusion and clutter of these last few days.

Very nice article. I also agree that stricter rules produce nicer, easier to read, and more uniform markup. That is most of the reason I use xhtml. xHTML is very backwards compatible (as with v.1), except if it is served as real xhtml. (Internet Explorer, I’m looking at you!) Otherwise, there is very little difference, and it will only further confuse people if there is a large divide between xHTML and HTML.

I’ve always been a big proponent of web standards.  Use semantic markup and validate your code are two things I always adhere too.

That’s why, I guess, I’m none too engaged by the wailing and gnashing of teeth around the current HTML5/XHTML2 conversation taking place across the web at the moment. Because whatever flavour of mark-up you use, you should always do those two things and there are still a ton of developers who don’t.

I switched from HTML4 to XHTML some time ago because I felt it gave me the structure I wanted in my coding.  It was virtually always served using a text/html MIME type so it was pretty much HTML4, but with XML syntax.  It just seemed cleaner and more structured and helped hone my trade.

There are still very few websites out there that are using XHTML properly (with an application/xhtml+xml MIME type) compared to the thousands of sites created by WYSIWYG tools that are barely proper HTML, let alone semantic or validated.  I would rather concentrate on cleaning those up and educating developers to code sematically and validate their code correctly going forward, before lameting XHTML2, which seemed on a hiding to nowhere from the get go.

The next freelance web site I build will probably be in HTML4, just to brush up. All subsequent ones will be HTML5.

Jeremy Keith has a great post outling the basics you need to know, as does Jeffrey Zeldman.  Both posts have links to various other sources if you want to dig deeper.

Technorati tags: html5, xhtml2, standards, web

An Englishman abroad - Been working with the web since the 90’s, main focus now is UX/UI. Moved to Cleveland, Ohio in 2001. View all posts by davidmead

Thanks for clearing that up Jeremy and nice to read a balanced - emotion reduced account for the first time this week.

Thank you for the clear explanation.

Great explanation, especially the analogies. There is so much misinformation about this topic, so it’s great to see a pragmatic approach in this world of extremes.

Ah, great post, thanks for writing it and clearing it up for the majority, and you explained it better than most of similar explanations I’ve read so far (including myself :p )

XHTML 2 is dead. Long live XHTML …as HTML 5.

Well said. ^_^

Good article - I’m glad you cleared that up.

I think most people have gotten used to and prefer the XHTML syntax now.

Great explanation. Thanks.

Thanks for this very clear explanation, Jeremy.

Thank you for explaining this in plain English, Jeremy.

Great article! And thanks for adding the Zeldman link. Also, I took your advice and simply changed the doctype on my XHTML Strict code, and it validated for HTML 5, hooray!

Well written. Its intereting in Zeldman’s comments to see people like Rimantas come back out of the woodwork in this conversation with some kind of renewed vigour that their doomsayer obsessiveness about XHTML being wrong is somehow finally proven. That guy was accosting me years ago with the very same diatribe of techno-babble, quite unprovoked as well.

I think their obsessive ranting shows more about a lack of work-life balance than it does about the reality of web development. I like the way you have succinctly put each statemtent by the way, so it will be interesting is they now come over to argue otherwise.

Yes its a good thing. End of story. :)

I’m trying to understand the different points of view on this issue. Your post seems pretty much snark-free. Henri’s Q&A seemed clear to me, but I don’t understand all the issues.

It seems the point of contention is not HTML 5, but the usefulness of XHTML 1.1 over 1.0. Would you say that is fair?

quote"Java is to JavaScript as ham is to hamster" -Apparently you’ve never had a hamster and bacon club sandwich.

Personally, I’m happy to hear that xhtml 2 is dead. It was needless in the first place. Too many people trying to put their mark on unnecessary standards gets us into needless messes.

You summed it so well, Jeremy! :-)

And I, for one, prefer writing XHTML syntax, too: it’s simpler for me, a bit more strict than HTML syntax, and works well in all browsers that can read & interpret HTML.

So I don’t understand from where came all of a sudden this "Didn’t we tell you, XHTML is dead?" thing that I hear now here and there, all the time, especially after W3C announced that XHTML won’t be developed anymore… :-/

XHTML 1.0 is a valid standard. XHTML 5 is developed. HTML 5 with XHTML syntax is perfectly valid, too. So, I don’t understand… :(

@cilice @adactio That is a rather apt description!

Java is to JavaScript as ham is to hamster adactio.com/journal/1595 Misunderstanding markup - July 7th, 2009

Just a friendly reminder that Java is to JavaScript as ham is to hamster. - quote from @adactio: adactio.com/journal/1595

“JavaScript is not Java – JavaScript is to Java as hamster is to ham.” adactio.com/journal/1595 (@adactio, July 7th 2009)

@arberesford source for “Java is to JavaScript as ham is to hamster”: adactio.com/journal/1595

…Java is to JavaScript as ham is to hamster.adactio.com/journal/1595