- From: Sergey Melnik <melnik@db.stanford.edu>
- Date: Fri, 08 Sep 2000 18:00:20 -0700
- To: Tim Berners-Lee <timbl@w3.org>
- CC: ML RDF-interest <www-rdf-interest@w3c.org>
Tim, FYI, one of my earlier drafts [1] gives some specific recommendations for the design of languages of different kinds including SQL. One of the benefits of URIzation is context-free recognition of referenced objects (e.g. field in a relational table). The DL'00 paper [2] demonstrates mixing of some very different (Web) languages (page 5). Sergey [1] http://www-diglib.stanford.edu/diglib/ginf/WD/ginf-overview/#uaf-sql [2] http://www-db.stanford.edu/~melnik/pub/dl00.pdf Tim Berners-Lee wrote: > > When I try to explain the same thing for the nth time for n>~5 > I write it up in Design Issues. Now I have just done a small > bit > http://www.w3.org/DesignIssues/Webize > on hwt is involved with taking a language and using on the web. > > So if you can't understand why your favorite KR/CG/whatever > langauge hasn't been used for knowledge on the web, apply these > simple instructions and maybe it will! Seriously... > > I just talk about replacing names with URIs and what that entails -- > I would like comments on the question, and points I have missed > about idenifying closed world assumptions in general and disposing > of them. Simple self-defense techniques if you meet them in > your dailiy business... cc me on any mail as I don't keep up with > this list. > > Tim
Received on Friday, 8 September 2000 20:44:38 UTC