Thanks for your reply Ian. I'm familiar with the options you suggest. What
I'm getting at is where does that coding / ORACLE text tool reside in a
truly 'interoperable' environment? For example in our databases we have the
coding built in to allow searches of various types using our own thesauri
(for example a thesaurus of archaeological monument types). Since this is
accessed using a web browser, those who have access to the database (ie our
staff, but theoretically anyone on the internet) can use our in-house
front-end to do the searches they want. However, this doesn't make our
database 'interoperable' with another database that uses a different
thesaurus (or a simple flat list or any other sort of semantic conrtrol or
just free text) to record the same concept.

Maybe there isn't a problem here, but I don't have a grasp of the solution
yet if there is one!

Ed

-----Original Message-----
From: Ian Davis [mailto:[log in to unmask]]
Sent: 30 June 2000 12:16
To: [log in to unmask]
Subject: RE: 'Interoperability' explained...?


I might be missing something here but...

How do you build in the intelligence that is captured in the structure of
the thesaurus such that a search on a non-preferred term automatically (ie
without user
intervention) identifies the preferred term used in a thesaurus and
retrieves using that term?
You would code it so a search using a non-preferred term looked for the
existence of that term across the whole thesaurus - this would include
authorized terms and synonyms.  The non-preferred term would be found as a
synonym for the preferred term.  You would then take the authorized term
linked to the synonym, perform a search on the authorized term and display
the results.

Or, similarly, how do you allow for a search on a term plus its narrow and
related terms?
You buy and integrate a package like the Oracle context cartridge which
allows you to allow for searches on text using your thesaurus.  So users can
search for 'term' and return occurances of that term plus occuranaces of all
the terms narrower to the term entered.
Or you develop your own thesaurus and write you own code to do the same
thing.  You would have to take a term and perform a thesaurus search for
that term.  If the term is found as an authorized term you would either look
for terms that were related to it in a broader or narrower sense.  Once you
had found all the NT or BT terms you would performed AND searches on each of
the BT or NT terms plus the original term and display the results.

Retrieval of search terms plus synonyms or NT's or BT's could be either
automatic or could be turned on and off by users.

Ian Davis
Keywording Manager
Photonica
www.photonica.co.uk

> -----Original Message-----
> From:	Lee, Edmund [SMTP:[log in to unmask]]
> Sent:	30 June 2000 11:50
> To:	[log in to unmask]
> Subject:	RE: 'Interoperability' explained...?
> 
> Thanks Paul for two very useful articles. Two very brief comments -
> 
> First of all the metaphor of plumbing for technical interoperability - it
> certainly doesn't hide the horror of it all from me: if my (limited)
> experience with plumbing is anything to go by give me computers any day -
> at
> least they don't mess up the carpet and bring down the ceiling when they
> go
> wrong:-)
> 
> More seriously, the issue of semantic interoperability is excercising me
> at
> the moment. I can see how a first objective would be to agree a set
> terminolgy for describing a record or resource (e.g. a term from the
> English
> Heritage thesaurus of Monument Types - thanks for the plug in your article
> by the way). However, how (or perhaps where is the question) do you build
> in
> the intelligence that is captured in the structure of the thesaurus such
> that a search on a non-preferred term automatically (ie without user
> intervention) identifies the preferred term used in a thesaurus and
> retrieves using that term? Or, similarly, how do you allow for a search on
> a
> term plus its narrow and related terms?
> 
> I guess my concern, briefly, is that metadata initiatives auch as DC, and
> interoperability standards such as XML seem to be developing independently
> of the need for semantic interoperbility that is vital if we are to
> realise
> the vision that you have set out.
> 
> Edmund Lee
> English Heritage Data Standards Unit
> 
> 
> 
> -----Original Message-----
> From: P.Miller [mailto:[log in to unmask]]
> Sent: 27 June 2000 11:38
> To: [log in to unmask]
> Subject: 'Interoperability' explained...?
> 
> 
> All,
> 
> I've got a short piece in the latest Ariadne, which attempts to reinforce 
> the idea that Interoperability is more than just technology...
> 
> I'd welcome comments/discussion...  ;-)
> 
> http://www.ariadne.ac.uk/issue24/interoperability/
> 
> Also in the same issue, I've a piece on identifiers, for anyone 
> interested in DOIs and the like...
> 
> http://www.ariadne.ac.uk/issue24/metadata/
> 
> Paul
> 
>   -- dr. paul miller - interoperability focus - [log in to unmask] --
>      u. k. office for library and information networking  (ukoln)
>      tel: +44 (0)1482 466890                 fax: +44 (0)1482 465531
>   ---------------------------- http://www.ukoln.ac.uk/interop-focus/ --
> 


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%