That’s one of the major findings of MAMA 2008, which states that only 4.13% of the code they’ve indexed and analyzed complies with W3C standards.

MAMA ran every single URL it analyzed through the W3C validator; the validator’s SOAP response contains a binary true/false result of the validation. A “true” value is considered a successful validation. [link]

Should this be of major concern? I’d say “probably”, because the reasons for being concerned are mixed.

You should be concerned if you’re building sites that are meant to take full advantage of the benefits that well-marked-up coding provides. For instance, documents with an accurate !DOCTYPE declaration are rendered in a different (and more universal) mode than documents without the declaration (ie, “quirks mode”).

However, a reason to not be overly concerned with this statistic is the finickyness of the W3C validator. Most mainstream browsers are capable of compensating for the odd mistake in human-created code. When Opera says “4.13%”, they’re talking about 100% validations (boolean true/false). If you forget to close a document with , it could be considered a violation, hence a nonvalidation - but it won’t really affect the user experience.

Same thing goes with not ending empty elements with />, or a & in a hyperlink, minor little nitpicky flaws that don’t have major impact on document rendering/functionality at all.

It’s more likely that, on average, a large portion of the Web’s code validates at acceptable levels, which would clock in at a far more realistic number than “95.87% invalid”.

But now, you at least know not to stress if your code fails at the Validator - 95% of the Internet’s got your back ;)