> John Cowan wrote:
> >
> >...
> >
> > No flame here. But you could ask our customers who are paying a 10%
> > premium for XHTML news with RDF-format SNOMED, vs. those who just get
> > the XHTML, whether *they* consider it worthwhile. I grant that this
> > is not really a proper comparison, as I doubt whether any of them are
> > using general RDF tools.
>
> This is at the heart of what will make RDF succeed or fail. Today people
> use XML data that "happens" to be RDF compatible in exactly the same way
> that they do regular XML data. RDF will start to succeed if and when
> tools arise and gain currency that take advantage of the RDF-ness of the
> data. I am thinking in particular about:
>
> * query engines
> * validators
> * data binding APIs
>
> The Fourthought guys convinced me that they were making practical use of
> RDF precisely because they have a bunch of RDF-level tools that they
> use. I think about these tools as being roughly analogous to SQL, except
> that the data source is graph-structured XML rather than a relational
> database. In that sense I don't see anything fanciful or AI-ish about it
> at all. They should talk about it as the "neo-relational web".
Thanks, Paul. You nailed it. At Fourthought, we do practical things with
RDF. Period. I personally am mostly curious about the Semantic Web, but it
is not what impels me to use RDF. We have used RDF technologies and 4Suite to
implement projects for clients with remarkable efficiency. I can disclose
that one of these clients, the Global Knowledge Engineering division of Sun
Microsystems, ended up with a very practical and valuable extensible metadata
repository based on RDF, including viewing and management applications. One
senior observer of the project noted that he had seen another such project in
another company take ten-fold of the time and ten-thousand-fold of the budget
to create the same thing using proprietary technology. This is no
exaggeration.
What we did was more a matter of extensible and generic database structures
and relational models than any fanciness with semantic networks. This is the
heart of the value I see in RDF. It was nice to see the skepticism of a few
hard-core RDBMS and OO folks melt away over the few months' life of the
project, as well.
To illustrate exactly how in the Sun project we addressed the three very
accurate bullet items you put up:
> * query engines
We ended up inventing Versa (though we are running it as an open community
standard) precisely because all the existing RDF query languages and engines
seemed very academic, and seemed to failed to take advantage of the very power
of RDF. In 4Suite we have Versa strongly integrated into XSLT (which is how
the Sun project used it), Python and HTTP, because just like SQL, RDF query is
not very valuable in isolation from other technologies.
> * validators
We ended up using a mix of RDFS and DAML+OIL for this. Note that we do not
use much of the inferencing capabilities of DAML+OIL, just the improved
constraint language. It turned out that this met our needs well enough,
though both RDFS and DAML+OIL need a bit more cooking before they're ready for
very broad use.
> * data binding APIs
We didn't need to do very much of this in the Sun project because RDF was the
native format. The main data binding was the binding of Versa query results
to XML for processing within XSLT.
For other projects, we have developed what we call "document definitions" (a
confusing name, I know) which can use XPath or XSLT to define a data binding
between XML and RDF. We have this because we recognize that not everyone will
be as willing as Sun was to make RDF the native DBMS. In fact, I am also
working on a similarly practical RDBMS to RDF mapping.
Anyway, let's stop confusing RDF with the Semantic Web, and let's sto using
wither as a straw man substitute for well-meaning people who have well-meaning
concerns about Web services. I've been involved in Web services since late
2000, and I see both it and the Semantic Web as complementary technologies,
both of which can make important contributions to the current Web. Despite
its rap, as Joshua points out, people are doing practical Semantic Web things,
RDF or no, just as people are doing practical Web Services things.
--
Uche Ogbuji Fourthought, Inc.
http://fourthought.com
http://4Suite.org http://uche.ogbuji.net
Track chair, XML/Web Services One (San Jose, Boston):
http://www.xmlconference.com/
RDF Query using Versa - http://www-106.ibm.com/developerworks/xml/library/x-thi
nk10/index.html
WSDL and the Wild, Wild West - http://adtmag.com/article.asp?id=6004
XML, The Model Driven Architecture, and RDF @ XML Europe -
http://www.xmleurope.com/2002/kttrack.asp#themodel
