Re: Updated RDFa-in-text/html tests

Hi Philip,

> I agree it's mostly just a pathological edge case - if someone declares a
> prefix as the empty string, and then uses that prefix for RDFa data, they
> can't really expect it to work sensibly. But it's quite common in text/html
> documents to have xmlns:* attributes with no value, e.g.:
>
> ...

That's right. There are many features in IE where the mere presence of
a namespace is enough to enable that feature -- i.e., you don't need
to actually have a value for the namespace. Given the number of
examples you've found, I imagine that Microsoft tools must put these
namespaces into the documents automatically.


> ...
>
> It would be nice if these people could use RDFa (e.g. by
> copying-and-pasting CC licensing data), and still have their page handled
> robustly (i.e. still being able to extract the CC data) despite having this
> kind of bogus markup elsewhere in their pages. Fatal errors would be bad
> (they'd make it hard for someone to incrementally adopt RDFa because they'd
> have to fix all these other issues in their markup first), but anything else
> (ignoring the attribute, undeclaring the prefix, treating it as a relative
> URI, etc) seems reasonable to me.

Yes, definitely.

Regards,

Mark

-- 
Mark Birbeck, webBackplane

mark.birbeck@webBackplane.com

http://webBackplane.com/mark-birbeck

webBackplane is a trading name of Backplane Ltd. (company number
05972288, registered office: 2nd Floor, 69/85 Tabernacle Street,
London, EC2A 4RR)

Received on Friday, 5 June 2009 16:08:44 UTC