Re: Late, but I have reservations.

Gannon wrote:
> Second, some of my issues may have been dealt with already: including
> 1) the syntax incompatibility with DCMI <meta> and <link>, and 2) the
> same incompatibility vis-a-vis GRDDL.

As Toby mentioned, this is not an incompatibility with DC, it's simply a 
different way to express metadata. In addition, if I remember correctly, 
Dublin Core is considering changing their XHTML recommendation to use RDFa.

Regarding GRDDL, I don't think your comment makes sense: GRDDL is a 
generic transformation mechanism. RDFa can be interpreted using GRDDL, 
or other means of expressing data can be interpreted using GRDDL. So I 
don't see what the incompatibility, or even inconsistency, is.

> Third, to use the RDFa attributes within the <head> element *at all*,
> regardless of syntax, might lead to less clarity due to the nature of
> meta data.

Less clarity for whom? There exist plenty of mechanisms for expressing 
human-invisible metadata already. <link rel="alternate"> is one of them. 
Or simply publishing RDF/XML.

> http://www.RUSTPrivacy.org/PIIandRDFa.pdf

In general, I think your interpretation of RDFa is confused. RDFa is 
about expressing structure for data that is already being published as 
HTML. There cannot be any deep privacy or policy implications to RDFa, 
since the expression of the raw data can already be done using existing 
technology. RDFa only helps to connect the human-readable to the 
machine-readable. If a publisher wishes to keep those two far apart, 
then there are many other ways to accomplish this than RDFa.

-Ben

Received on Thursday, 24 July 2008 04:59:32 UTC