However, one day I faced reality. Met a light, so to speak. I read an article by a very wise guy, that basically said: We make websites to help the customer do their thing better. Now to validate in W3C. I was, like, WHAT?! I thought W3C validator was for the customer, and not the developer. I came to realize, however, that everything comes at a price. XHTML is by nature strict. X stands for XML, so to have valid XML, everything must be validatable. So without validation pass, XHTML made no sense.
I also got in contact with reality when I started validating big sites like google, microsoft and – hold your horses – the validator itself. And they all turned out invalid. Surprising, eh? I sent a mail to the developers of the validator, where they replied something like this: “eh, hehe, hihi. didn’t see that coming – oh the irony”.
The motivation of the validator is to make sure the sites follow W3C’s recipe for HTML5, XHTML, HTML 4.01 etc. The thing though is that many times it’s impossible to validate and do cool things at the same time. Many plugins and cross browser compatibility fixes actually break the validation.
Another fun motivation for validation is SEO and semantics. Semantics is a word that describes the addition of meaning to content, so that search engines, screen readers and other tools without (well functioning) human eyes can understand the meaning and importance of the content. However, it turns out that search engines don’t really care about validation. It doesn’t mean that they don’t care about semantics, but not validation. Read this article to gain more insight. This topic is however religion. And you will find many SEO expert calling this heresy.
So, to conclude this essay: Validation is mostly never relevant to user experience. I will run the validator and fix relevant errors, but I will not make the site validate.
Some sites that do not validate:
- Google.com (57 errors)
- Microsoft.com (517 errors)
- Vg.no (332 errors)
- Paypal.com (37 errors)