W3C home > Mailing lists > Public > www-archive@w3.org > December 2001

RE: WebOnt Requirements - I18N

From: Pat Hayes <phayes@ai.uwf.edu>
Date: Wed, 12 Dec 2001 11:30:58 -0600
Message-Id: <p0510100ab83d4218badb@[]>
To: "Jeremy Carroll" <jjc@hplb.hpl.hp.com>
Cc: <hendler@cs.umd.edu>, "Jeff Heflin" <heflin@cse.lehigh.edu>, <herman.ter.horst@philips.com>, <dlm@ksl.stanford.edu>, <phayes@ai.uwf.edu>, <jos.deroo.jd@belgium.agfa.com>, <ned.smith@intel.com>, <www-archive@w3.org>
>>  >R11. Internationalization
>>  I think this is out of scope unless very focused on ontology-specific
>>  issues of internationalization - W3C has other folks
>>  w/internationalization as a specific focus and we don't want to stray
>>  into their area.
>With respect, the I18N-wg will disagree.
>The charmod wd will probably come to rec before we do, and we will be bound
>by the constraints that charmod places on W3C recommendations.
>I do not believe these will be heavy, but we will need to give some account
>of i18n and the webont language. In fact, a quick scan suggests that all the
>current suggestions for requirements for specificiations (marked in S) are
>ones that should be met at a lower layer (e.g. XML or RDF).
>The core issue for webont and I18N, as I see it at the moment, is that in an
>ontology that has a property named in american english "color", there is no
>way currently in any of the lower layers (XML, RDF) to express that this is
>american english, and that the british english is "colour", and the italian
>"colore". This is because the property names are the fragment IDs of URI
>references, and there is no way in the current stack of attaching a language
>tag to URIs, URI references or fragment IDs.

This is not a problem, because these linguistic differences should 
not be in the ontology. Terms in an ontology are not natural-language 
words. They are more like identifiers in program text. If I write 
some Java and use 'color' as an identifier, I want that character 
string to stay exactly the same as I wrote it, even if the reader is 
Italian or English. Here's another way to put it: any uriref in a 
piece of WOL is not in English, or Italian or anything else: it is in 
WOL, and it should stay in WOL exactly the way it was written.

>I note that in some cases where the same word means different things in
>different languages (e.g. "shopping" in English & Italian) this could become
>a significant problem.

Only if someone were to make the basic error of thinking that the 
terms used in the ontology had the same meaning that they would have 
if they were words in English or Italian.

>We may well need to have a fairly significant effort to explain how
>ontologies can be internationalised.

They should not be, in the sense you are here meaning. A good 
exercise for anyone dealing with a formal ontology is to replace 
every 'word' by a meaningless string, and then try to understand the 
assertions in the ontology. Because the ontology means *exactly the 
same*  as far as any software is concerned.

IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
Received on Wednesday, 12 December 2001 12:31:06 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:42:03 UTC