Re: SUO: ontology areas for review
> LinguisticExpression
> <http://ontology.teknowledge.com:8080/rsigma/SKB.jsp?req=SC&skb=SUMO&id=127>
> <http://ontology.teknowledge.com:8080/rsigma/SKB.jsp?req=SC&skb=SUMO&id=127>
>
>
>
> For each area, we'd like to know
> - what terms do you think are missing and should be added?
> - what additional axioms should be added to flesh out the semantics
> of the existing terms?
> - are any axioms incorrect, or overly specific?
OK....
why are ConstructedLanguage and ComputerLanguage not in a subclass relation:
or is ConstructedLanguage restricted to a constructed HumanLanguage (which
is cute and Esperanto-friendly, but maybe not true?)
how can Sentence be a subclass of Clause? A sentence can consist of a whole
bunch of clauses and a few other things too.... different modes such as
spoken
(does this have sentences? where are intonational phrases?) and written
are not distinguished... Maybe
a confusion between orthographic sentences (which can be weird) and
the notion in the SUMO of sentence as that unit which can have a truth
value?
... which is in any case not captured in "A syntactically well-formed
formula of a &%Language."
Standard problems with the notion of sentences as parts of texts.... even
if this is a mereological part relation... as texts are expressed
through sentences (sometimes) rather than being
built up of them. Is the "realization" relationship a mereological part?
(That is, words may be built up of sequences of phonemes, but they
express/realize sequences of morphemes: morphemes do not have
phonemes as parts, etc.(if one still believes in such things of course)).
Text: ( documentation
<http://ontology.teknowledge.com:8080/rsigma/SKB.jsp?req=SC&skb=SUMO&id=32>
Text
<http://ontology.teknowledge.com:8080/rsigma/SKB.jsp?req=SC&skb=SUMO&id=661>
"A &%Phrase, &%Sentence or set of &%Sentences that perform a specific
function related to &%Communication, e.g. express a discourse about a
particular topic.")
Does a "newspaper" fit here? Or an encyclopedia? Presumably not;
so would they be collections of texts? But surely these are not
"upper".... (but why is "article"?).
Makes me wonder what this bit of the ontology is meant to be doing... what
work will it do? Which goes back to questions raised on this list about this
area again.
Any construction in this area is going to be theory-laden: perhaps a further
argument against the one-top-ontology-fits-all? Although I belong to the
"nice its there" camp rather than the "no use trying" one! :-)
John Bateman.