[tcs-lc] relationships between vernacular names

Nozomi Ytow nozomi at biol.tsukuba.ac.jp
Fri Apr 29 21:29:14 PDT 2005


Roger,

how can I represent relationships between vernacular names in TCS? 
We have five UTF-8 strings to represent cherry in Japanese, whatever
its scientific name is.  I'd like to represent equivalence between
them without referring to scientific names.  Using
"is vernacular of"/"has vernacular" here seems an abuse.

If we add a type to enumeration to represent relationships between
vernacular names, then how the above relationship should be
represented?  We store the relatioship by triplet of
RelationshipAssertionID, TaxonConceptID, and a boolean-ish whether
the TaxonConceptID is FromTaxonCocnept or not; type of 
RelationsihpAssertion and AccordingTo is stored in
RelationshipAssertion data object having the RelationshipAssertionID
(in temrs are TCS equvalent, of course). In above cherry case, it
would be recorded as equivalent to

(vernacular-equivalence, cherry_in_half-width-Katakana, isFrom)
(vernacular-equivalence, cherry_in_full-width-Katakana, isFrom)
(vernacular-equivalence, cherry_in_Hiragana, isFrom)
(vernacular-equivalence, cherry_in_ShimplifiedKanji, isFrom)
(vernacular-equivalence, cherry_in_OriginalKanji, isFrom)

where the second element is ID of TaxonConcept retaining these
reprsentaitons of cherry.  Do I need to create five or ten
(combination of these fives) RelationsihpAssertions to represent
it (and hence GUID for each of them)?  I think it is better to allow
RelationsihpAssertion to have more than one FromTaconCocnept and
ToTaxonCocnept.

Cheers,
JMS
--
Dr. Nozomi "James" Ytow
Institute of Biological Sciences / Gene research center
University of Tsukuba
Tsukuba, Ibaraki 305-8572
Japan


More information about the Tcs-lc mailing list