[topicmapmail] tagging vs topics

[Tony Coates]

> -----Original Message-----
> From: Ann M Wrightson [mailto:ann.wrightson@alphaxml.com]=20

> - use the original ISO 13250 SGML Architectures concept (now=20
> known as HyTM) to map (appropriate structures of) elements to=20
> topic map components. (Do I remember a demonstration of this=20
> with NewsML?) Needs a reasonably good semantic match; and=20
> this is one of the ur-visions for 13250.

I'm sure anyone who subscribes to this list could handle this, but what
about the 99.99% of the world which doesn't?  Which tool(s) do they use?
They won't be able to roll their own.  While Topic Maps + AF:NG might be
a theoretical solution, it's a geek^2 (geek-squared) solution, and hence
impractical.  A solution with at most one minority technology
(henceforth geek^1) can become popular if it solves the right problem,
but geek^2, geek^3, and higher order solutions usually have too steep a
learning curve to justify the benefits that they offer.

The advantage of being able to embed your topic map information in your
document (or your document in your topic map) is that you are only
dealing with just the one concept.  Of course, you could have one topic
map for the "real" topics, and a second AF:NG topic map to encode the
mappings from the documents to the "real" topic map, but you can see how
it starts to sound a bit messy (unless you have a tool that really makes
this kind of workflow straightforward).

Anyway, the same applies to the other suggestions that Ann made.  As I
read them, each is just another way of making the mapping from document
concepts to topic map concepts more-or-less "the user's problem", but
it's this problem that users want to have solved (and not just
transformed a bit).  Topic Maps need to be seen to solve a problem, and
the whole of a problem, not just provide a partial solution that may
never be made whole.

	Cheers,
		Tony.
=3D=3D=3D=3D=3D=3D=3D=3D
Anthony B. Coates
XML & Search Architect
Chief Technology Office
Reuters Plc, London.
Tony.Coates@reuters.com
=3D=3D=3D=3D=3D=3D=3D=3D


------------------------------------------------------------- ---
        Visit our Internet site at http://www.reuters.com

Any views expressed in this message are those of  the  individual
sender,  except  where  the sender specifically states them to be
the views of Reuters Ltd.