Code v Data

...or, in the hip vernacular of the moment, should that be 'Meta' v 'Mutter'?

I just got asked for my learned (?) opinion on whether HTML constituted code or data.

My kneejerk response was 'Data'. After all, HTML is meant to comprise of documentation marked up to look pretty.

Except:

1. HTML is a language, and contains (admittedly static) constructs that require interpretation in order to be used.
2. dynamic behaviour can be provided by the inclusion of applets, javascript etc.

OK, then. Code it is.

So, why did I initially think Data?

Assuming that I have some tentative grip on reality, I can only assume that my confusion arose through:

• the fact that great pains have been made to ensure that HTML is representible as pure ASCII text, and
• the notion that HTML and associated initiatives such as XML strive to separate form and function, with HTML providing the form (ie Data). It hasn't entirely succeeded.

In fact, the web standards are evolving at an incredible rate (HTML -> XHTML?) and I haven't been playing the game long enough (or seriously enough) to catch up with the latest.

Believe it or not, I have worked for 20 years as a software engineer without needing (or even being able to get access to) much exposure to web technology. Until recently, that is.

So, as I said: learned opinion?