Re: @rel syntax in RDFa (relevant to ISSUE-60 discussion), was: Using XMLNS in link/@rel

Maciej Stachowiak wrote:
> 
> Before, you said that you're not trying to push an RDFa designed for XML
> only into HTML as a done deal. But your comments here and elsewhere sure
> make it seem that way. Earlier you described Henri's proposals as
> "change for the sake of change" and referred to "an HTML world view
> which is empirically incorrect". These do not seem like the hallmarks of
> being open to feedback and change.

Which technical arguments of Henri's did I not respond to in technical
terms?

> But a comparison of the technical merits of using @profile vs.
> backing off of CURIEs would be more constructive than your current
> approach, where you seem to be rejecting anything but your preferred
> approach out of hand.

"Out of hand?" I've delved into multiple specs, given technical
examples, and even written code to prove that Henri's assumptions were
incorrect. I've also explained how we're exploring the use of @prefix
specifically to address Henri's concern. I've spent an untold number of
hours debating this issue when Henri isn't even interested in *using* RDFa.

Maybe you're assuming that there is no downside to a new syntax for RDFa
in HTML, so why don't we just remove a feature because it displeases
some, right? Unfortunately, there *is* a great downside, because we
already have implementors such as Yahoo, Creative Commons, and a number
of UK government agencies. I and other RDFa task force members are in
regular contact with these folks, and we know that the cost of change
for them would be quite high.

So yes, all things being roughly equal, I think we should default to the
existing syntax. Personal taste is, in particular, *not* a good reason
to change anything.

But, if there's a good technical reason to adopt a different syntax, I'm
open to it, case in point with @prefix (even though the case is fairly
light, in my opinion.)

Which issues do you think are good technical reasons to adopt an RDFa
syntax in HTML5 that differs from its current syntax in XHTML1.1?

-Ben

Received on Thursday, 5 March 2009 01:04:01 UTC