RE: metadata vs. "structured data"

On Tue, October 24, 2006 4:36 am, Neil McNaughton wrote:
> It seems to me that RDF's intent is to make master data (as above)
> accessible to other readers/applications that don't know or care about the
> details of BananaML itself? Is this right?

RDF can store any kind of data you like, which I think is made clear in
the normative specs. A Primer should address the most obvious needs of
people getting accustomed to the technology being explained. In my
opinion, there are many people out there who first heard that RDF was
originally developed to store metadata for web resources and then decided
that RDF/XML was too complicated and difficult to integrate with their
most common type of web resource: HTML pages. The increased structural
richness of XHTML2 will have more people storing content in it, and the
rumored greater simplicity of (the otherwise cryptically named)  RDFa will
have more people reconsidering whether to use RDF to store metadata about
that content. The primer should show them that the rumors are true and
that RDFa is a good way for them store metadata about those resources.

I realize that RDFa would be a great way to create XHTML pages of movie
times that are readable by both humans with browsers and by semantic
crawlers that incorporate that data into people's potential schedules. I
haven't  heard of any theater chains looking for ways to store XHTML so
that such data can be easily extracted, though. I do know of major
magazine publishers who want to store easily addressable metadata in XHTML
pages (not necessarily publicly available web pages, but in content that
plays a role in B2B workflows). Helping them out will give RDF a better
foothold in a highly visible industry; I think it's worth it.

Bob

Received on Tuesday, 24 October 2006 13:19:12 UTC