Re: [xtm-wg] How can a large chunk of documents and data can be structured

["Bernard Vatant" <universimmedia@wanadoo.fr>]

Steve's relevant advices and remarks I want to second.

Since knowledge engineers, librarians or domain experts are not so far for
most of them, topic maps hackers, critical coordination tasks are pending in
TM development, the most critical of all being indeed to rewrite accurately
the TM domain ontology in terms of relevant association and role types, and
tune all the actors around that rewriting, from ontologists to developers
and interface designers. Being in charge of such a coordination in some
Mondeca projects, I can confirm it's a challenge to impulse and maintain
this consistent "topic map spirit" all over the actors team. What are the
relevant competences and background to achieve this coordination task is an
open question. I won't be as arrogant as to recommend a profile :))
But it has to be done and explicitly untrusted to someone - full time job.

We are at the most difficult stage, trying to impulse a bootstrapping
process. That needs patience and pedagogy. We need tutorials, and Steve's
one in Montreal is indeed a good initiative. And we need certainly more
permanent technical and "ontological" support and tools, to be developed
e.g. in OASIS technical committees frame (see other thread impulsed by
Murray), to help newcomers to the technology not to loose time rediscovering
all the problems we had hard time figuring out and beginning to solve.

Regards

Bernard

-------------------------------------------------
bernard.vatant@mondeca.com
Mondeca - "Making Sense of Content"
www.mondeca.com
-------------------------------------------------
[Steve Pepper wrote]


> At 01:16 18.07.2001 -0700, Anil Kumar wrote:
> >            I am facing a problem of how to make a
> >topicmap out of a large chunk of data and documents
> >available to me. As a developer I dont know which data
> >for eg. of a topic in library/book is related to
> >others and how they are related and what are to be the
> >possible relationship between them.
>
> As a developer it's not your job to know the semantics of
> the data you are working with! For a project like this to
> be successful, there *must* be collaboration between
> documentalists (indexers, librarians, etc.), developers,
> subject matter experts, and knowledge engineers.
>
> The involvement of subject matter experts is critical.
> A topic map represents a portion of the knowledge in the
> domain covered by the information resources, and without
> subject matter expertise, the result will be next to
> useless.
>
> Note also that the team should also include topic map
> expertise and experience in knowledge engineering.
>
> Conventional cataloging know-how is not enough, and may even
> be an impediment. Librarians need to overcome their ingrained
> habits of viewing everything in terms of subject area
> categories. This is not because topic maps can't support that
> kind of categorisation -- they can.
>
> The problem is that people with a library science background
> tend to think too much in terms of hierarchies composed of
> simple "topic/subtopic" relationships. Topic maps allow you
> to express much richer relationship types, and it takes some
> experience to tease these out of the data and the experts!
>
> >           Does all the above work need to be done
> >before making a topicmap/xtm file or how do we go
> >about to make a topicmap/xtm file if we are given a
> >variety of documents,data and files.
>
> In my experience, the more analysis, design, and modelling
> of the information domain that you do up front, the quicker
> you arrive at a useful end-result. Consult the documents,
> data, and files as part of that process, of course. But
> don't start doing development work until your topic map
> "ontology" (essentially, the set of topic types, association
> types, and occurrence types) has been worked out.
>
> By the way (plug), I am covering all of this in my tutorial
> on "Modelling Topic Maps" at the Extreme Markup conference in
> Montreal on Monday August 13th. This is the blurb:
>
>    This highly interactive tutorial is for people who already
>    know that topic maps are something for them, who understand
>    the basic concepts, and who are ready to create their first
>    real topic map application or prototype. It focuses
>    exclusively on issues of analysis and modelling, and teaches
>    a methodology for designing topic map ontologies. Whether
>    your application relates to information management, knowledge
>    management, or application integration, this tutorial will
>    provide you with a head start and help you avoid the most
>    common traps and pitfalls.
>
> For more details, see
>
>   http://www.extrememarkup.net/extreme/2001/monday.htm#pepper
>
> I hope to see you there!
>
> Best regards,
>
> Steve
>
> --
> Steve Pepper, Chief Executive Officer <pepper@ontopia.net>
> Convenor, ISO/IEC JTC1/SC34/WG3  Editor, XTM (XML Topic Maps)
> Ontopia AS, Waldemar Thranes gt. 98, N-0175 Oslo, Norway.
> http://www.ontopia.net/ phone: +47-23233080 GSM: +47-90827246
>
>
> To Post a message, send it to:   xtm-wg@yahooGroups.com
>
> To Unsubscribe, send a blank message to:
xtm-wg-unsubscribe@yahooGroups.com
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/


To Post a message, send it to:   xtm-wg@yahooGroups.com

To Unsubscribe, send a blank message to: xtm-wg-unsubscribe@yahooGroups.com 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/