Yes, and hopefully the infrastructure (i.e. the distributed ebXML registry/repository) will soon be in place to enable the core components to be managed on-line, in nearly real-time and with equivalent terms in as many languages as is necessary.

Ed Buchinski

Senior Project Coordinator | Coordonnateur principal de projets

Enterprise Architecture and Standards | Architecture intégrée et normes

Chief Information Officer Branch | Direction du dirigeant principal de l'information

Treasury Board of Canada, Secretariat | Secrétariat du Conseil du Trésor du Canada

Ottawa, Canada K1A 0R5

613-957-2497 | Buchinski.Ed@xxxxxxxxxxxxx | facsimile/télécopieur 613-957-8700

-----Original Message-----
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx]On Behalf Of Peter F Brown
Sent: March 7, 2007 2:55 PM
To: [ontolog-forum]
Subject: Re: [ontolog-forum] [ontology-summit] PLEASE, PLEASE!!

Regarding the UN/CEFACT CCTS: the big strength is not so much the core component themselves but rather the fact that there is a standardized methodology for identifying, creating and managing them….

Peter

Keith:

These are good comments. Answers inline:

“I'm wondering why "one person or a small group" labeling some "thing" is a problem? Is this to say that one person or a small group can't label a thing correctly and comprehensively keeping the best interest of the end-user in mind? If a taxonomy has thesaural properties, or is just a good-old fashioned thesaurus, it should be able to accommodate various view points with synonymy.”

(DN) From my perspective, even the best intentions get tags wrong or fail to comprehend variances or differing points of view. The thesaurus approach is a reasonable way to mitigate this in theory, however from practical experience, it is much harder. The United Nations Trade Data Element Dictionary (UN/TDED) tried the once and for all approach and got very far with it based on a large buy in. The UN/CEFACT CCTS project tried to add the thesaurus angle with a complicated twist – capturing the context in which the differing labels were applied and mapping each variance to the upper level ontology. The jury is still out.

As I said before, I am sitting on the fence. For well framed projects, the one group deciding certainly works well. Examples include the dewey decimal system (spelling??), WSDL, XML Infoset, HTML and others where the semantics of the fixed set of elements were agreed upon and seem to be used with no problems. If the work is taken outside the original frame or scope, that seems to be when issues arise. The UN/CEFACT CCTS work is all in english for example. We had great trouble mapping certain words like ‘payload’ to french. Nevertheless, if the scope is english only speakers, I think they did a good job.

“As for the Slashdot.org "experiment," why generalize about how an ontologist or taxonomist would tag it? It sounds like, if there are, indeed, such unique terms being used to tag something at Slashdot.org that they would elude a person whose vocation is to tag content using the most widely-accepted terms, then you're really saying that "one person or a small group" is tagging things in a completely unique way at Slashdot.org! Isn't that a problem then?”

(DN) the context was the main reason. For example, an ontologist or a taxonomist might tag it based on a directive to identify nouns, verbs and semantics. Some of the /. Crowd tagged news articles with words like “stupid”, “idiots”, “ROTFL” etc. While I cannot concretely justify my statement, I would encourage you and others to take a look for yourselves. It really emphasizes the point.

Cheers

Duane
--
**********************************************************
Sr. Technical Evangelist - Adobe Systems, Inc. *
Chair - OASIS SOA Reference Model Technical Committee *
Blog: http://technoracle.blogspot.com *
Music: http://www.mix2r.com/audio/by/artist/duane_nickull*
**********************************************************