On Tue, Jun 23, 2009 at 9:39 AM, Casey Bisson <[log in to unmask]> wrote:

> The mistake here is presuming that (X)HTML coded data isn't (or can't be)
> data.
>

I think it's a greater mistake to assume that it will be.

We are talking about putting semantics into a structured data source (by
people who are trained in both the data format in particular and the tools
to input data into it), why introduce another data format (out of context, I
might add -- we're only talking about a single tag) which is much more
likely to be adulterated by human input and the fallacies that come with
being human?

A semantic XHTML serialization of data is very different than ad-hoc usage
of HTML for a local need.

-Ross.