- From: Dave Beckett <dave.beckett@bristol.ac.uk>
- Date: Thu, 18 Sep 2003 13:53:43 +0100
- To: "Miles, AJ (Alistair) " <A.J.Miles@rl.ac.uk>
- Cc: public-esw@w3.org, "Wilson, MD (Michael) " <M.D.Wilson@rl.ac.uk>
[going to public list, to be linkable] These comments I made last week I forgot to forward, still apply to the latest draft http://www.w3c.rl.ac.uk/SWAD/thesaurus/tif/deliv81/final.html Sorry for not forwarding till now. Dave --- Figure 1 missing definition of is-indicated-by* - no footnote given for * Term-Oriented Model In a term-oriented model, although it may be tacitly implied that a set of terms stands for some abstract concept, the concept is not reified in the model. Terms which are preferred terms are used as the nodes in the generalisation hierarchy. In this model it is usual to distinbuish between the class of objects that are preferred terms and distinguish^ the class of objects that are non-preferred terms. This model is illustrated in figure [???]. missing figure number 2 Summary of the TIF RDF Schema Figure 3 Thesaurus class isn't related to anything else? Scope-Note (rdf:ID="ScopeNote") Defines a comment on a concept, prescribing the bounds of appropriate usage of the concept. Sub-classes of this class are used to describe specific types of comment. A General-Note refers to [???]. A History-Note refers to [???]. A Hierarchy-Note refers to [???]. A Translation-Note refers to [???]. An Editor-Note refers to [???]. references missing Properties The domain/range values for the properties of the TIF schema are illustrated in figure [???] and also in table 1. figure 4 Figure 4 rather wide figure - 1146x158 has-classification-code (rdf:ID="code") Defines a unique [???] identifier for a Concept. ? is this a reference or just being unsure? is-defined-by (rdf:ID="isDefinedBy") Relates a Concept to the Thesaurus it is a part of. [Use rdfs:isDefinedBy ???] It could be used but rdfs:isDefinedBy remains vague Language Identifiers The Language class and identifiers for all languages defined by ISO639-1 are supplied in a separate file (see Appendix [???]). Appendix V Summary of TIFS RDF Schema "...and UF (use-for)...." never mentioned till now Classes "...absence or presence of the use property..." use-for ? Later I see use/use-for are an inverse pair of properties relating preferred to/from non-preferred terms Deriving TIFS from TIF what is ^<sup>-1</sup> terminology? inverse relationship? RDF Schema or OWL Ontology in appendices [???]. [should the ontology extend descriptions for the same resources as the schema, or use different ones???]. appendices II (RDFS) III (TIF OWL), IV (TIFS OWL) Better say which OWL - OWL DL, OWL Full, OWL Lite Example: ELSST, a Multilingual Thesaurus Refers to March 2002 as a future date - this reads as out of date. Conclusions and Future Work The TIF is built as an extensible framework of classes and properties. In this way it may be extended to express other kinds of semantic relationship between concepts or terms. For example, we could introduce ACK /AF relations between terms for abbreviations and acronyms, such as in the DAML+OIL ontology from DRC, .. reference and citation needed ... or we could express the diagonal relationships related-broader-concept and related-narrower-concept, such as in the ETB thesaurus schema. ... ditto ... We could also model categories, as are found in many thesaurus, as a sub-class of concepts or terms, and the semantically more precise relationships that go with them. Appendices Links to the RDF/S and OWL files would be good. Which OWL version (date) are these? There are problems with # and rdf:ID in these docs. <!ENTITY tif "http://www.w3c.rl.ac.uk/2003/07/31-tif-owl#" > xmlns:tif ="&tif;" xml:base ="&tif;" and later: rdf:ID="Thesaurus" which will give a URI http://www.w3c.rl.ac.uk/2003/07/31-tif-owl##Thesaurus Correct: <!ENTITY tif "http://www.w3c.rl.ac.uk/2003/07/31-tif-owl" > xmlns:tif ="&tif;#" xml:base ="&tif;"
Received on Thursday, 18 September 2003 08:54:44 UTC