Why doesn’t google.com validate?

Matt's answer:
We have a question from Michael Thingmand in Denmark. Michael asks, “Why doesn’t google.com validate according to the W3C?” Okay, long story, but the historical answer is this. Google looks at the number of bytes that we actually return to users. And we want that to be as small as possible, because every byte matters when you’re serving up hundreds of millions of search requests to users. So, typically we’ve been a little more willing to say things like, oh, we don’t really need that double quote or something like that, or we’ll specify that color in a way that doesn’t validate. And the interesting thing is then there were a few people who said, oh, well here’s a way you can make Google’s pages validate and still use a smaller number of bytes. So that was a very cool and noble effort. But, Google still has to deal with a ton of idiosyncratic browsers, you know everything from WebTV to mobile phones. And so we’re a little more worried about compatibility with all those different types of browsers than we necessarily are about whether we validate according to the W3C. I think it would be fantastic if we did. I would definitely support any Googler who wanted to spend 20 percent time making sure that we validate, because do I think that we could validate and still have a small page size. But it’s important to realize that the vast majority of pages on the web don’t validate. And while it’s great to make valid code, and it can be easier to maintain if you know that all your nesting is in good shape and everything validates, we have to crawl and index and return results on the web even if pages don’t validate. And so we don’t give any sort of boost to pages if they validate. It may be a nice thing to do for your own internal purposes, but it doesn’t get any sort of Google boosting in your rank or anything like that. And the simple reason is that the vast majority of pages on the web don’t validate as it is right now.
by Matt Cutts - Google's Head of Search Quality Team