Re: MARCXML to Topic Maps implementation!

Hi Patrick,

[...]
> When you say *proper ontology* do you mean:
[...]
> 3) some combination of #1 and #2?

This one. :)

Maybe something like the bibo ontology [1] may help to get #1 right
(OpenLibrary uses that ontology to some extend, I don't know how good
this ontology is or if it is widely used).

The fields are meaningless for me. If you aren't a MARC expert it
would be hard to find a book which was written by author XY.

Ad #2 [more subjects]:
That's probably the hardest point in the mapping but would definitely
the killer feature of Topic Maps above MARC.

But doing this right will take a long time, I think since you need a
human who checks the translation carefully. I doubt that you'll be
able to automate that translation completely, I think it will be a
semi-automatic translation, i.e. the app. could suggest some authors
based on a Wikipedia search but the human has to decide if it *is*
that subject in question.

I know that the project is very young, I just wanted to say, that the
advantage of the topic map over the MARC record is not yet obvious for
me. Maybe I am too impatient, though. ;)

It's true that you (as Topic Mapper) could add more information to the
topic map and merge-in information and add topics etc. which wasn't
possible with MARC, though.

[1] <http://purl.org/ontology/bibo/>

Best regards,
Lars
-- 
Semagia 
<http://www.semagia.com>

<http://www.topicmaps.de/mailinglist/> German Topic Maps mailing list
<http://tinytim.sourceforge.net/> Open Source Topic Maps engine
<http://mappa.semagia.com/> Mappa - Python Topic Maps engine