AustLit, W3C comments and updates
Just to keep things rolling along and so you don’t miss anything, there are two recent comments I want to point out:
In a comment on Dan Brickely and the W3C, Karl Dubost from the W3C said, “It would be cool if you could explain how FRBR could be used in details for example to manage an album photo. That would be a concrete use case and would help people to develop the tools if all requirements are defined.” He included a couple of links to some W3C discussion about this last February. I’d never thought about applying FRBR to a photo album, but it’s a great challenge. Add a comment if you have one.
And in a comment on AustLit, Kerry Kilner, AustLit’s executive manager, left a comment saying anyone interested in FRBR is welcome to ask for the password for their permanent guest account. That’s very generous, and I’m certainly going to ask for it. You can get in touch with him through AustLit.
Kent Fitch, also from AustLit, was kind enough to send me an XML dump of how Souls in the Great Machine is represented in their system, and it’s wild viewing. It goes over twenty levels deep in some places, and has lots of <work>, <expression>, and <manifestation> tags. I’ll slurp it up into a Perl XML parser so I can examine it more. I like reading XML, but I have my limits.
I’ve been working on RDF photo annotation, and an FRBR OWL ontology. My iPhoto RDF exporter tries to do a bit of FRBR-like modelling: unlike most efforts, it doesn’t say that a JPEG file depicts a person, but that a JPEG file is an image file for a photograph that depicts a person (and so on). This allows multiple instance files to be associated with the same image.
Furthermore, I intend to model derivative images (crops etc.).
The schema and tool are up on my site.
Comment by Rich — 15 July 2005 @ 12:13 pmIn 2002 Stephen L. Abrams was working on a refernce model for digital library
Comment by Patrick Le Boeuf — 20 July 2005 @ 7:40 amobjects, that was derived from FRBR and that contained an example based on a
digital photograph of a painting:
ABRAMS, Stephen. A reference model for digital library objects [on line].
[Cambridge, MA]: [Harvard University Library], 2002/July/16 [cited 5 November
2002]. Available from the Internet: http://hul.harvard.edu/~stephen/Model.doc
That interesting document does not seem to have been modified since 2002.
I’m not sure FRBR is totally appropriate in order to deal with photographs.
In my opinion, it would be wiser to model the specific problematics of
photographs separately, and to check afterwards how the resulting model could be
articulated with FRBR.
Besides, there is a working group the aim of which is “harmonization” between
FRBR and CIDOC CRM. CIDOC CRM is certainly more relevant than FRBR to account
for photographs. That working group is currently striving to “translate” FRBR
into the object-oriented formalism that served to define CIDOC CRM. The resulting
OO FRBR will look more like an actual ontology than the current ER FRBR, and it
will certainly be more helpful to use that tool in connection with RDF than the
ER FRBR. Unfortunately, the documents produced by the Working Group on FRBR/CRM
harmonization have not been made publicly available yet, as it is a work in
progress and the results are not yet mature for public display. But anyone who
is interested can contact me.
Just that you know, we continue. See the discussion starting at.
http://lists.w3.org/Archives/Public/semantic-web/2005Sep/0002.html
Comment by karl — 7 September 2005 @ 12:10 am