[
Lists Home |
Date Index |
Thread Index
]
> It's a point that Uche's in a better position to argue, but once more this
> isn't an either/or choice. It will be possible to get the benefits of XML
> and RDF technologies - that the primary serialization of RDF is XML is a big
> hint.
I think the article of the title of this thread does a huge injustice by
insisting that people use RDF/XML. I have been advocating for about 2 years
now (since I got thorughly fed up with RDF/XML) that:
1) Users should keep on using XML usign good *XML* design techniques, many of
which are well established from SGML days
2) Tools vendors should do a better job of enabling users to create RDF model
views of XML instances, in order to empower them towards (1)
Good XML design, in my experience, leads to the generation of very good RDF
views, as long as you have the tools. We in the 4Suite camp have put quite a
bit of wrk into making good XML/RDF reflection tools available, and I've
always been very surprised that we seem so alone in this effort. I know that
there are many impeovements to be made to 4Suite's doc def architecture, which
either offers simple XPath pattern extracion (good for stuff like extracting
metadata from HTML) or the full power of XSLT (good for more sophisticated
models, but beholden to the unfortunate RDF/XML). I've long thought there
should be something better, and have just lacked the tim to work on it. My
leading thought is that one should be able to annotate RELAX NG ow WXS
schemata with RDF view notations. Eric DV's excellet work with XVIF puts this
tantalizingly within reach, but time is short.
Re: Other matters raised in the thread so far:
Mike, I'm a consultant. If I don't make my clients more productive, I don't
eat. I think that my experience, and the trend of my work proves that RDF
does indeed pay great dividends if used intelligently. Others have reported
the same gains. I'm not sure why you claim not to have seen them when I know
you have. If it suits you to pretend I'm the only one in this camp, even
though you correspond with at least two others in that camp on an almost daily
basis, then I'll leave you to that amusement.
I do regret that there is not enough body of best practice on how to use RDF
so well out there. I have no time to do it: it really needs the sort of
tireless work that Costello applied for WXS. As for the RSS folks, I think
you'd be extremely naive to try to derive any technical point from such a
clearly political matter. I haven't heard any loud protests and revolt
against Crative Comments, DMOZ, RPM/RDF, Musicbrainz, PRISM, or the like.
Most likely cooler heads in those camps realize that syntactic differences
such as whether to add a "rdf:about" attribute are silly grounds for so much
strife.
My argument against RSS 0.92 has always been that is is *bad XML*, regardless
of whether or not it uses RDF. After all, I have no problem extracting RDF
from good XML. I haven't even bothered looking at RSS 2.0, but frankly, given
its provenance, I'm not inclined to waste my time.
As for RDBMS. It's another point that is not as simplistic as its attendant
flames. For some classes of data modeled in RDF, RDBMS is fine. In other
cases, it's a curse (especially when you have a lot of N-ary or transitive
relationships, which are common in real life). This is an old problem with
RDBMS, and is simply an illustration that no model is ideal for all data,
despite the claims of relational purists.
--
Uche Ogbuji Fourthought, Inc.
http://uche.ogbuji.net http://4Suite.org http://fourthought.com
Gems From the [Python/XML] Archives - http://www.xml.com/pub/a/2003/04/09/py-xm
l.html
Introducing N-Triples - http://www-106.ibm.com/developerworks/xml/library/x-thi
nk17/index.html
Use internal references in XML vocabularies - http://www-106.ibm.com/developerw
orks/xml/library/x-tipvocab.html
EXSLT by example - http://www-106.ibm.com/developerworks/library/x-exslt.html
The worry about program wizards - http://www.adtmag.com/article.asp?id=7238
Use rdf:about and rdf:ID effectively in RDF/XML - http://www-106.ibm.com/develo
perworks/xml/library/x-tiprdfai.html
|
