Not this again

March 6, 2009

For some reason, Jeff Atwood periodically likes to return to the great HTML validate vs. don't validate debate. As it happens, I agree with him: getting all your site's pages to validate is largely pointless. And while validation of XHTML is essential—or will be, when application/xhtml works in all browsers—XHTML also sucks. In fact, it sucks for that very reason.

Take it away Tim Berners Lee:

"Some things are clearer with hindsight of several years. It is necessary to evolve HTML incrementally. The attempt to get the world to switch to XML, including quotes around attribute values and slashes in empty tags and namespaces all at once didn't work. The large HTML-generating public did not move, largely because the browsers didn't complain. "

To put it more bluntly: XHTML has been a failure.

Predictably, a flood of programmers all appeared in Jeff's comments section to take him to task for being so lazy. Look, the fault tolerance of HTML is a feature, not a bug. It's why the web is what it is today. If HTML validation in browsers had been as strict as a compiler is on a programming langugae, HTML wouldn't have been widely adopted by programmer and non-programmer alike, and the web never would have become what it is today. And for goodness sakes, it's not a programming langugage. HTML is a markup language. There is no equivalence.

I fully acknowledge that handling invalid SGML is hard, and yes, many browsers are bad at it. (That's one thing HTML 5 aim to fix.) But you have to take the bad with the good and the good provided by fault tolerance far outweighs the bad.

Now, it's probably not fair to expect us programmers to take into account normal human flaws, but it's very annoying to hear people lecture Jeff about the importance of perfectly valid markup. As far as I'm concerned, I want my HTML to be valid enough to render consistently in the browsers I'm targeting. Beyond that, I've got better things to do. If you want to waste your time, go ahead, but don't push your hangups on me, man.

Tagged with: .

Leave a comment:

Comments are closed for this entry. There are 3 comments for this post:

name:
Russell Heimlich
date:
Mar 6, 2009 08:45 p.m.
comment:
Validating HTML is the equivalent of jocks comparing the size of their junk, it's only relevant when other jocks are around. And if you plan to run ads on your site, you can forget having a perfectly validated site. For some reason ad code hasn't evolved since 1999 when it was actually a challenge.
name:
James Stevenson
date:
Mar 8, 2009 01:05 p.m.
comment:
Ha ha, agreed.
name:
DZ
date:
Mar 23, 2009 10:41 a.m.
comment:
I tend to agree, but to play devil's advocate for a bit, validating markup as you go reduces the chance for a snowball effect later on. Especially with javascript -- if markup accidentally puts a browser into quirks mode, or if the engine incorrectly autocorrect invalid code, you're relying on implicit browser logic rather than explicit code logic to define behavior.