[topicmapmail] Announcement of XML Schema for ISO 13250 Topic Maps

[David RR Webber]

Message text written by Patrick Durusau
>The more complex a topicmap becomes the more difficult it will be for
administrators to trouble shoot problems by manual examination of the
underlying
topic map. I suspect specialized tools will be developed to implement, test
and
maintain topicmaps, particularly those that have any significant size. Not
quite
the Indiana Jones type image we might like to project but certainly the
reality
of working with complex information systems.
<<<<<<<<<<<<<<

Patrick,

This all reminds me of a job tracking system I implemented for a 
major telephone company some years ago.

The VP's assigned it top business priority as critical to the 
company's marketing of products - I had exclusive access to machine
resources at the high levels of system priority (and I abused that
to the hilt running more jobs in a month than a whole department!!).

The field information analysts delivered the spec's for the intervals
and events to track.  There were five other supporting developers.

Two months later the first live data came off the system.  The VP's
and the field managers looked at the printed charts and the
on-screen tables.  "That's impossible - doesn't make any sense"
where the general comments.  Because not all products followed
all the 30 data points - the statistics - while mathematically completely
correct - were meaningless.

This is the trap and the issue that Martin is envisioning.  While
biblical uses of TM's are interesting - the killer app' is clearly
business analysis.   Hard experience teaches us - that unless
the end users can exactly understand the information path - and can
verify and authenticate it - then all you have is meaningless
statistics.

Let's not let the theory blind us from building practical solutions.
That's why I go back to the requirements every time.

In the telephone example above - the requirements got screwed
up - they realized that afterwards.  If they had listened to the 
first tier users and management - it would have worked. 
Instead three more tiers of managers 'refined' the requirements
believing they could make silk out of a sows ear.

Let's build topicmaps mechanisms that can solve real business
problems quickly and consistently.  It might not be so intellectually
satisfying - but it will work - and it will fund the next level of 
research after that.  Sorry to be so pragmatic for a New Year!

Thanks, DW.