Original thread here:
http://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0603&L=dc-rdf-taskforce&T=0&P=4334
ons 2006-03-29 klockan 22:57 +0200 skrev Thomas Baker:
> On the basis of Mikael's drafts
>
> http://dublincore.org/architecturewiki/DCRDFTaskforce/DCRDFExecutiveSummary
> http://dublincore.org/architecturewiki/DCRDFTaskforce/DCRDFExecutiveSummary2
>
> I have added another:
>
> http://dublincore.org/architecturewiki/DCRDFTaskforce/DCRDFExecutiveSummary3
>
> If we could finalize this basic explanation by next Monday,
> it could be included in the documentation packet for the DCMI
> Board of Trustees meeting in late April.
Apart from the fact that this is a fair bit longer than I was imagining,
it's a fairly good summary :-)
A few important comments:
1.
"The DCMI Abstract Model now provides a reference model against which
particular DC encoding guidelines can be compared."
I think we need to say:
"The DCMI Abstract Model now provides a reference model based on which
particular DC encoding guidelines can be defined."
2.
No point in mentioning my name, really.
3.
"The two legacy specifications differ with regard to whether a
dc:creator or dc:contributor is a person (i.e., an entity), or a name
(i.e., a string). In "DC-Simple-in-RDF", a dc:creator is a name:"
We need to make sure that the issue is not limited to
creator/contributor.
4.
Figure 2 needs to be augmented with a type relationship like in the text
version above. The type-less version is actually potentially problematic
as well, so please make sure there's an rdf:type triple there.
5.
"The range "Literal" would be reserved for metadata terms which are
typically associated with value strings, such as dc:title."
Actually, we have not committed to restricting dc:title to literal
values at this point. In fact, we are considering leaving the range open
to allow for, for example, variants of the "same" name described by a
single statement using multiple value strings (with different languages,
for example).
6.
"Declaring "Agent" as the range of dc:creator will mean that inferencing
applications will expect to treat dc:creator as an entity."
rather
"Declaring "Agent" as the range of dc:creator will mean that inferencing
applications will expect to treat the value of the dc:creator property
as an entity."
7.
"to ensure the long-term viability of Dublin Core in RDF."
Note that the issue is limited to RDF. The other DC syntaxes are *not*
affected by the issue.
(Except possibly some Vocabulary Encoding Schemes will need to be made
sub-classes of the range of the respective property?)
/Mikael
>
> Tom
>
--
Plus ça change, plus c'est la même chose
|