On Mon, 28 Jul 2003, Andy Powell wrote:
> It would be interesting to do a mapping between qualified DC and LOM -
> which I think would show a much more balanced two-way mapping with less
> loss of information and more reliable transformations.
Duh... I am thick! It would help if I read to the end of the document
before commenting. I now see that you have done the qualified DC mapping.
I'll try and put together my thoughts on this. One immediate comment is
that you probably need to think about a slightly more complex mapping
because DC and LOM use 'properties' and 'encoding schemes/sources' in
slightly different ways. For example DC uses the combination of
dc:format scheme="dcterms:IMT"
for what LOM calls
technical.format
(I think! Again, I should really add this to my XSLT transformations).
I think that it is also the case that you need to explicitly show both way
mappings - i.e. you need to show dc->lom and lom->dc separately.
Personally, I would do this for 'qualified DC'->LOM and LOM->'qualified
DC' and state how you move from this to the simple DC case. I.e. treat
all the DC elements, element qualifiers and encoding schemes together.
Andy
--
Distributed Systems, UKOLN, University of Bath, Bath, BA2 7AY, UK
http://www.ukoln.ac.uk/ukoln/staff/a.powell +44 1225 383933
Resource Discovery Network http://www.rdn.ac.uk/
|