Is machine readability a valid point for xhtml if it does not validate?

Question

I often run across developers that insist on using the XHTML doctype and when I ask why they common response is that its "machine readable". Once the project is underway the markup does not validate.

Now that their markup does not validate...is machine readability valid anymore?

I assume that if it does not validate it can't be processed as XML and cant be queried using xpath.

Answer 1

Being well-formed XML and valid XHTML are two different issues. But anyway, plain HTML is perfectly machine readable if it is well-formed and valid. The only difference is that there are more, better tools and libraries for working with XML content than SGML content. Certainly I find it easier to genrerate valid XHTML than valid HTML, but there's no real excuse for generating invalid documents of either type.

Answer 2

If you screw things up, you screw things up. If they tell you the pros about XHTML and are not delivering XHTML, but something else it will simply not be an easy job and it highly depends on how NOT Xhtml/html compatible their product is. However depending on your environment and your use you should consider using tidyhtml.

Answer 3

Developers probably won't have to process the finished markup sent to the clients, because they can connect into any stage of the preprocessing. Therefore any error later in the chain is unlikely to get caught, and will be served to the users until there's a visual bug or someone tries to parse / validate at some later stage during / after preprocessing. Validation can catch such bugs preemptively, but have you ever heard of a workplace where "preemptive" is not a buzzword?