xml-dev - RE: local, global (was various ontology, RDF, topic maps)

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

Take a look at the dot.com fallout.  People 
will believe and invest and will lose their 
shirts and others will take their shirts and 
build big pink houses on the hillsides outside 
San Jose with them.  When HTML said SGML 
was evil, some took that to heart.  When 
others said HyTime was evil, some took that 
to heart.  Both times, the sayers were wrong. 
It took experience and a lot of hard work 
for people to find the baby in the bathwater, 
but they have and there are remarkably good 
systems coming out as a result that have little 
in common with the original HTMLOverAll systems. 
They look like... SGML + hypermedia circa 1989 
with better graphics.  So far so good. 

That is what we are doing here with the semantic 
web.  I started out with the "the semantic web 
is a crock" position knowing full well that the underlying 
tech and concepts do work in a limited fashion 
because, being an old guy now, I was a young 
turk when case grammars, AI, expert systems 
and all that were discussed last time.  They 
do work, in a limited fashion.  But I took 
the very pessimistic approach precisely because 
of the "idealistic" view you talk about.  The 
WWW and the W3C hyped a lot of people's lifesaving's 
into other pockets.  Now we are more cautious 
about such visions, but we must also look at the 
tech itself and ask what good we can do with it. 
As others have mentioned, some of the search 
engines are a LOT better these days.  When 
I sit with my son to do a book report on the 
Enola Gay and type that in, I don't see nearly 
as many superstitious unwanted hits as we once 
did.  That's progress.

What we are doing IS trying to understand how 
local ontologies (all markets are buyer's markets 
regardless of supply and demand) can interoperate. 
There are some very hard social problems as well 
as technical problems.  We aren't shying away from 
those in this forum.  We may not solve them, but 
we acknowledge them and try to advise others about 
them.  An ontology is an opinion of sorts.  As such, 
regard the source, but moreover, observe and test 
the source, forgive, but don't forget.  The rest 
is Tit for Tat.  To be sure, what the system 
doesn't see, doesn't exist in that system.  We 
have to be mindful of that.  ANY hypermedia 
system can make information disappear by ignoring 
an address, can make any concept appear to be 
the work of another by frequency of attribution, can alter 
the course of events by refusing to acknowledge a 
submission.  All true and precisely why we 
use these email lists so assiduously to proof 
the visions of those with big podiums.  We 
build that foundation of agreement by open 
debate on the issues, begging those that have 
other work to do to please ignore these posts 
until they need the information for their work. 

A universal semantic is a golem.  We not only 
won't build it, we don't know how to build it.  The aleph 
is not ours to mount and mud just doesn't cohere 
in the rain of continuous hits on a server. :-)

Len 
http://www.mp3.com/LenBullard

Ekam sat.h, Vipraah bahudhaa vadanti.
Daamyata. Datta. Dayadhvam.h


-----Original Message-----
From: Simon St.Laurent [mailto:simonstl@simonstl.com]

I find it remarkable that people talk about 'agreement' as if it weren't a
radical concept in itself, and that they actually seem to believe that
maintaining large sets of agreements about meaning is both possible and
beneficial.  I'm heartened by discussion of the contingencies involved in
such projects, but still find the foundations far less solid than large
groups of people seem to believe.