Google Uses Bad Markup

I think the thing over the Golden Spider entrants has been a bit ridiculous to be honest. While web standards are important, they are no the be all and end all that some people have made out over the past couple of weeks. Personally I find working towards an improved user experience to be more of a worthwhile venture.

Anthony

That maybe true to a point, however some of last year's winners were completely unviewable in Firefox, which hardly improved anyone's experience

I remember seeing an article about that. Apparently, the HTML was chosen very specifically in order to be compact and to look the same in as many browsers as possible.

I believe it would be possible to rewrite in a standards-based way, and be more compact, but I don't think Google would care for hints on how to write their code ;-)

http://www.google.com has 63 errors! and it's not exactly the most complex webpage the world has ever seen!

It's possible to be standards compliant and browser compatible.

As I mentioned over on the forums, the extra size of the document if it were valid would cost Google money considering the bandwidth.

I know an extra byte is nothing to any of us, but with that much traffic it all adds up.

Being unviewable in certain browsers is obviously a big issue but people being over zealous in terms of full validation gets on my nerves.

'sfunny cause it's true.

I used to be one of those bigots. If it wasn't W3C compliant, then it wasn't going near my server.

Then I needed the money, and realised that as long as it works, the client will be happy. If it's 100% correct code, then I'm happy as well, but at least if the client is happy I'll get paid and be satisfied to move on.

100% compliance is for people that have the time for it. If you're not being paid top-dollar to produce your code, then the client is not going to get top-quality.

We do our best with the time that is allotted to us, but sometimes we just don't have enough time to do /the/ best.

As for google - they obviously have the time and resources to do it correctly, but have chosen not to. I'm sure they have some compelling reasons for this, or they would have corrected it long ago (they have some pretty smart people working there).

I'm sorry, but not having a DOCTYPE is unforgivable.

It's the most basic of basics.

I know it's hard to be standards compliant and work in all the popular browsers. sometimes I too run out of time and have to leave one or two validation errors in there.

But I do strive for validation, as much as possible.

Have heard before that one of the reasons Google chooses not to keep the main page standards complient to save in size.
I have no idea now many hits google gets a day, but I'm betting it's in the billions. The few bytes for a multiplied by a a few billion quickly adds up.

Should have read "The few bytes for a DOCTYPE", but Wordpress complained about the angled brackets :)

As someone already said, yep, this is a bandwidth thing. Every browser around will deal with the Google page, so why go adding unnecessary bytes? Note that they're not too generous with line-breaks, either. (You see the same thing with some SOAP libraries; everything that can be left out without causing chaos is).

I think this is a sensible decision on their part, actually, considering that their market can use the site anyway.