Dear Joao, of course, you are completely right from the normative point of view.
My argument was motivated by the question how to deal with it in practice if you have term lists or thesauri such as the AAT at hand which do not obey those standards. Whereas new thesauri will (hopefully... but I am in doubt) obey the distinctions required by the definition of SKOS, in particular the one between concepts and terms, those legacy thesauri don't. In order to meet the requirements of SKOS, technically, an intermediate step is necessary. One solution w.r.t. OWL-DL (!) --- which is incidentally the same as one we found out independently --- is described in section 3.3 of the SKOS primer (http://www.w3.org/TR/2008/WD-skos-primer-20080221/). The ex:Painting term is an example of what I was thinking of. My methodological concern was just that rechristening a term to a concept by just putting string quotes around it or writing it in upper case letters is a notational trick, but not a solution. So I am happy that you agree with me to keep the distinction. This is not trivial, because you can find it in the literature, also in papers on SKOS. Getting to concepts for/in application domains requires an extra effort. I agree with you that my suggestion for a more liberal formulation may lead to further misunderstandings, so probably keeping it as strict as in Christian-Emils original wording is the better choice. But what I had in mind --- although it is too technical and too detailed to be reformulated for the "Types" section --- is exactly the solution mentioned above. So, I think, we don't have a real disagreement and need not continue the discussion. Regards, -- Guenther Goerz On 11/6/08, João Oliveira Lima <[email protected]> wrote: > Dear Guenther Goerz, > > I fully agree with you in the several points, like: "concepts" and > "technical terms" are disjoint classes. > > The point is: the "E55 Type" class isA "E28 Conceptual Object". The > CIDOC CRM ontology allows all classes/instances to be associated with > multiple types and names (properties of E1 Entity). In my point of view, the > correct place of "Technical Terms" is under "E75 Conceptual Object > Appellation" class (direct subclass of E44 Appellation). > > The new standards for thesaurus, like the British BS 8723 > (http://www.iskouk.org/presentations/will_21072008.pdf > , see slide #2) and W3C SKOS > (http://www.w3.org/TR/2008/WD-skos-primer-20080829/), have > the "concept" (and not the name, label or technical term) as the basic unit > of organization. > > The BS 8723 (and ISO 2788, also) identifies 3 specializations of > hierarchical relationship (BT/NT) between ThesaurusConcept instances (see > slide #6): BTG/NTG (class/specie), BTP/NTP (whole-part), BTI/NTI > (class-instance). I agree with you that ""narrower/broader term" is not the > same relation as "sub/super-concept" in a concept hierarchy". But, the > BTG/NTG relationship is equivalent to super/sub-Concept relationship. > > Regards, > > Joao Oliveira Lima > > > > > On Thu, Nov 6, 2008 at 8:48 AM, Guenther Goerz <[email protected]> > wrote: > > > Dear Joao, > > > > > > On 11/6/08, João Oliveira Lima <[email protected]> wrote: > > > Dear Guenther and Christian-Emil, > > > > > > About your brief comment, I think that the "E55 Type as concept" is > > > correct. The terms (or Labels) which concepts are knowed should be > modelled > > > in another class. > > > > I fear I don't get your point. So, let me explain in more detail what > > I meant: Referring to the last paragraph in Christian-Emil's text, if > > we introduce the type "artist" we could take this term from a > > thesaurus like the AAT. I would not presuppose that such a technical > > term must have in any case the status of a concept, because then I > > would exclude term lists and thesauri which talk just about technical > > terms. Maybe I have a more rigid use of "concept" because I do not > > regard it as synonymous to "technical term": Technical terms are just > > normalized words as in a controlled language, whereas concepts result > > from an abstraction process. And I think to keep the destinction > > between "concept" and "technical term" is important from a > > methodological point of view. If such a term is embedded in a > > "norrower/broader term" hierarchy in a thesaurus --- as one would > > expect --- one can of course navigate in this hierarchy as well, > > keeping in mind that "narrower/broader term" is not the same relation > > as "sub/super-concept" in a concept hierarchy. > > > > So, w.r.t. the practical use of E55, I would argue to keep the system > > as open as possible from a methodological point of view and not > > exclude to take terms for E55 Types from thesauri of the kind > > mentioned above from the very beginning. > > > > > > > Making an analogy with FRBRoo entities, the "E55 Type" is located > at > > > same level of "F21 Individual Work" (abstract entity, without symbols or > > > names), and the "E44 Appellation" (or Exx Type Appellation") is located > at > > > same level of "F2 Expression" (symbolic entity). > > > > If this was the idea of the authors of FRBRoo I must confess that I do > > not share the --- in my view rather obsolete --- metaphysical > > assumption of abstract entities without symbols or names. > > > > Regards, > > -- Guenther Goerz > > > > > > > > > > > > > > > > Joao Oliveira Lima > > > > > > > > > On Wed, Nov 5, 2008 at 9:59 PM, Guenther Goerz > <[email protected]> > > > wrote: > > > > > > > > > > > A brief comment: In the first paragraph you write that instances of > > > > E55 Type represent concepts. I think this is unnecessarily > > > > restrictive: They can just be terms (e.g. in a thesaurus) --- without > > > > the claim that they must be concepts, i.e. results of an abstraction > > > > step. > > > > > > > > > > > > Cordially, > > > > -- Guenther Goerz > > > > > > > > > > > > ------------------------------------------------------------------------ > > > > Prof. Dr. Guenther Goerz Fon: (+49 9131) 852-8701; -8702 > > > > Univ. Erlangen-Nuernberg Fax: (+49 9131) 852-8986 > > > > Department Informatik 8/KI goerz AT > informatik.uni-erlangen.de > > > > Haberstrasse 2 ggoerz AT csli.stanford.edu > > > > D-91058 ERLANGEN > > > > > > > > http://www8.informatik.uni-erlangen.de/inf8/en/goerz.html > > > > > > > > > > > > > > > > > > > > > > > > On 11/5/08, Christian-Emil Ore <[email protected]> wrote: > > > > > Dear all, > > > > > I attach a draft of a "about types". > > > > > > > > > > It is based on the new scopenote, the orignal text, Martin's new > and > > > the > > > > > comments from Erlangen. > > > > > > > > > > Regards, > > > > > Christian-Emil > > > > > > > > > > > > > > > > > > > > > > > > > > > _______________________________________________ > > > > > Crm-sig mailing list > > > > > [email protected] > > > > > http://lists.ics.forth.gr/mailman/listinfo/crm-sig > > > > > > > > > > > > > > > > > > > _______________________________________________ > > > > Crm-sig mailing list > > > > [email protected] > > > > http://lists.ics.forth.gr/mailman/listinfo/crm-sig > > > > > > > > > > > > > >
