Stephen Downes

Knowledge, Learning, Community
A nice comment in today's Semantic Web discussion group pointing to an underlying problem: "If we want RDF-based formats to be interoperable, they can't be extensible, because inevitably some mutually incompatible extensions or vocabularies will arise, or they can't be decentralized since some authority will have to maintain this interoperability." The author, Adam Atlas (?), lays out three approaches: (1) hardcode these similarities in function into parsers, (2) create new ontologies for the express purpose of bridging these similar but incompatible formats, or (3) politely ask the vocabulary authors to add compatibility. None of these, he argues, is workable. In response, John Hardin posted this item, which is a link to a long PowerPoint the Universal Data Element Framework (UDEF). Essentially this is an endorsement of option (2). It is also an extraordinarily bad idea. One wonders, if we have a UDEF, why we would need all those other datadata formats at all. But, of course, the different formats express real differences in opinion, differences that cannot merely be glossed over by a universal translater (language is, after all, culturally specific, and if the Semantic Web is about anything, it's about language). One wishes that the designers of today's Semantic Web had read their history.

Today: 1057 Total: 1063 [Direct link] [Share]

Image from the website


Stephen Downes Stephen Downes, Casselman, Canada
stephen@downes.ca

Copyright 2024
Last Updated: Mar 29, 2024 08:48 a.m.

Canadian Flag Creative Commons License.

Force:yes