HTML Validator and SEO

does invalid code hurt?

[Black_Jackz]

I was just using the HTML Validator to check my pages, to see how many errors appeared. Now, from Netscape and I.E. I get no errors viewing any of the pages, and they load up quickly. I use the CSE HTML Validator, and my pages come back with alot of errors in the code.

I then decided to check some of the other high ranking pages to see if there were clean? Nope, same thing. My question is how important is clean HTML, and which software seems to offer the best results for checking HTML code.

[WibbleWobble]

w3c validator :)

Its not hugely important for SEO, but is much more-so for accessibility, imo. Its all about the content, really. And links.
So many sites come back with invalid html it seems unlikely that any search engine could possibly penalise for errors.

[DrDoc]

Remember that the CSE HTML Validator returns alot more "errors" than what is true.

I'm with WibbleWobble, use the W3C Validator ( [validator.w3.org...] ). Also, there's a difference between errors and mere comments or warnings. :)

[sun818]

I use CSE too because you can use it on your desktop without Internet access. I think its good practice to fix your HTML as much as possible, especially if you use one template page for many parts of your site. :) I'm not going to worry about putting an alt="" in my <img src>, but I do want my tables to contain the proper opening and closing tags.

[tedster]

I agree that no search engine would intentionally penalize invalid code - that would not be in their best interests at all, since their users want access to the best variety and depth of information.

What SEO really does need is the elimination of code problems that keep sections of a page from being indexed by the spider. Such errors occur mostly with inaccurate copy/paste operations. ANY validator should catch those kind of gross errors.

[Mohamed_E]

I started validating to make sure that there are none of the gross errors that Tedster mentions, then gradually started learning enough to make the validator happier. I was never compulsive about it, there is still one "invalid" construct that renders perfectly well in all browsers and that I have no plans to fix. But one error in 83 pages seems acceptable :)

Search engines do not really care about valid HTML, but there are other things in the world of the web than search engines!

[aus_dave]

When I first started using the W3C validator I would get a long list of errors ;).

By fixing the errors, I gradually learned more about HTML and how you can't rely on a program like Dreamweaver to create your pages without some intervention and editing.

Now I validate all new pages after uploading and enjoy seeing the 'success' page instead of errors.

It might not mean much for SEO but validating code has taught me as much about creating web sites as hanging around here does :).

[tedster]

A Related Story:
A prospective client of ours went to a site we designed to check out our work on their mobile phone. Now we had developed the site several years earlier, when small screen displays were not on our radar, and we had some trepidation about how the site would look.

The site looked really good on the hand-held screen. Our prospect was quite impressed by this, and we got the contract.

The main reason behind that success story was that our code was valid. Small screen browsers are very lightweight, and they cannot forgive as many HTML errors as their big brothers can.

[DrDoc]

I use CSE because you can use it on your desktop without Internet access.

I do that with W3C's validator ;)

[gsx]

... and Netscape 4 does a similar job - but forgets to tell you what the error is ;)