[topicmapmail] the praeteritio explanation

[Alexander Johannesen]

On Mon, Jan 11, 2010 at 00:33, Michel Biezunski <mb at infoloom.com> wrote:
> The way I see things now, after several years of experience with real topic
> maps applications, is that the original model was working for a number
> of applications, but was too limited to capture situations where topics were
> not just nodes but could really be also associations between
> topics, or even occurrences or names. I consider connectivity
> as a way to create meaning by designing views.

When I was learning TM waaaay back, I latched on to the notion that
"everything is a topic", and by that I thought *everything* was a
topic, like names, associations, types, occurrences, everything. In
that sense I struggled with the TMDM because it was an interpretation
of that notion. It wasn't until I read (and understood?) the Topic
Maps Reference Model that everything clicked into place for me. And I
created a recursive key-value property store that had special indexes
that were TMDM compliant but in itself was a TMRM implementation.

> The TMDM model is
> perfectly compatible with this model

Well, I can represent a TMDM implementation through a more TMRM model,
of course, but is it the TMDM we want or is there something more
interesting going on in the TMRM? For me, it's the latter ; all
promotion ever pushed in the TM world has been the TMDM, all
tutorials, articles, books, everything. But the real juice, when you
look beneath the mask of TMDM is the true exciting bits. Whenever I've
taught TM to people, it isn't until I show the TMRM that people get
truly excited. The TMDM is a great model for some things, but I'm also
constantly hitting its constraints and jump into TMRM (and no, it's
not just because I'm stupid and don't know any better), constraints
that a lot of people find a bit puzzling and hard to get around.

>, but it's not the only possible representation
> of topic-aware information.

Amen.

> I think I may have thought the same thing all along, but maybe I was
> eager to get something standardized in a way that was going to be a
> first step. I also thought that things evolve constantly, even if for
> political reasons, we had to stabilize and standardize our thinking a certain
> way at a certain time. I certainly didn't expect that the original model
> was going to stick for so long, and regarded as not changeable.

Bit of a conundrum, there, really. I think perhaps that people thought
that getting the TMDM thing right was hard enough (and people reading
that standard still struggle for years the get all the subtleties of
it. Heck, *I* still struggle from time to time squeezing my reality
into it) and hence, let's use this one that we've spent so much time
and money and effort on, and squeeze everything the future throws at
us into it.

> i consider
> the fact that the model is starting to show its age as a promising sign
> for the future.

I couldn't agree more, and I'm almost afraid of asking what that
future might be ...


Regards,

Alex
-- 
 Project Wrangler, SOA, Information Alchemist, UX, RESTafarian, Topic Maps
--- http://shelter.nu/blog/ ----------------------------------------------
------------------ http://www.google.com/profiles/alexander.johannesen ---