W3C home > Mailing lists > Public > www-rdf-interest@w3.org > October 2004

Re: Troublesome relations

From: Adrian Walker <adrianw@snet.net>
Date: Tue, 19 Oct 2004 08:17:39 -0400
Message-Id: <5.0.2.1.2.20041019080730.01b83cb0@pop.snet.net>
To: Danny Ayers <danny.ayers@gmail.com>
Cc: www-rdf-interest@w3.org

Danny --

I don't see much of a conflict,  {relational,RDF} religious dogma aside.

To oversimplify a bit, a triples form of RDF can be treated as a ternary 
relation.  There's quite a bit of work that one can find via Google on 
storing and indexing such triples usefully in Oracle and the like.

One concern though, is that writing queries, and understanding what they 
are returning, can be much harder if we insist on storing everything at low 
arity.

There's an online example at our site**, called RDFQueryLangComparison1, 
that hopefully illustrates the last point.

                                 Cheers,  -- Adrian



                                         INTERNET BUSINESS LOGIC (R)

                                            ** www.reengineeringllc.com

Dr. Adrian Walker
Reengineering LLC
PO Box 1412
Bristol
CT 06011-1412 USA

Phone: USA 860 583 9677
Cell:    USA  860 830 2085
Fax:    USA  860 314 1029







At 01:34 PM 10/19/2004 +0200, you wrote:

>I've recently been reading around crossover between the RDF model and
>Codd's relational database model. Jay Fienberg brought to my attention
>[1] to a couple of comments from C. J. Date and F. Pascal (the
>legendary database debunkers): On Metadata, RDF and Relational
>Representation [2] and On Relational Binary Database Design [3]. Their
>line in both is essentially that the RDB is the One True Model. The
>comments in [3] don't seem unreasonable (however debatable), but [2]
>is a longish query with a terse response from Date:
>
>[[
>This approach is the old argument that all relvars should be binary in
>a different guise! Thus, a cogent counterargument is:  How do you deal
>with irreducible n-order predicates for n <> 2?
>]]
>
>Problem I'm having is coming up with practical examples of such
>predicates, or how this might impact on RDF modelling. (I could well
>be misinterpreting Date's statement, as Googling "all relvars should
>be binary" doesn't exactly give evidence of an old argument, rather it
>gives [2] as the first hit and expanding to 'relation variables'
>didn't help any).
>
>The nearest I can get is in wondering perhaps if this is isomorphic
>with the way RDF predicates can't be treated as instances for
>per-usage qualification, you can't really go from predPQ(a, b, c) to
>predP(a, b), predQ(predP, c).
>
>/notation undoubtedly dodgy, hope you can read my mind/
>
>Can anyone please shed light?
>
>Cheeers,
>Danny.
>
>[1] http://icite.net/blog/200410/data_articles.html
>[2] http://www.dbdebunk.com/page/page/1114998.htm
>[3] http://www.dbdebunk.com/page/page/1147347.htm
>
>--
>
>http://dannyayers.com
Received on Tuesday, 19 October 2004 12:17:44 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:52:10 GMT