Re: [xtm-wg] Re: The Future of TopicMaps.Org

[Murray Altheim <altheim@eng.sun.com>]

Jean Delahousse wrote:
> 
> Murray,
> 
> I think you are right "Semantic Web" is a bag.
> 
> What defines Semantic Web is what you find in the bag today, that will give
> you an idea of what Semantic Web adress as technical subjects.
> 
> It's what have been done by the European Commission who is getting ready to
> help semantic web technologies.

Maybe I can earn that post in Barcelona after all...  :-)

> I went to that first presentation of "semantic web" works and projects last
> november and here are some of the issues adressed :
> - automatic metadata over content
> - automatic metadata over multimedia document
> - classification and navigation over content using metadata
> - Organization of subjects (TM / RDF)
> - Ontology tools
> - Multilingual ontology (more or less universal)
> - Navigation tools (text, graphical...) over text or multimedia documents...
> - ....
> 
> If you look closely and try to understand the articulation between those
> differents layers of projects, I think one can get a pretty clear idea of
> what is "semantic web".

I'm much more comfortable with the idea that *a* semantic web is being
co-developed by a lot of communities, using many different technologies,
with some overall gameplan but no dictatorial (ie., monopolistic) control
over that design than I am with the idea that one person has an idea of
something called "The Semantic Web(c)" that is developed according to
one vision, one set of proprietary plans, one team. That team wouldn't 
likely operate in the public interest and would be driven by a lot of 
money, and would also likely represent only a small portion of the thinking 
and development that is already going on. I'm also worried about the 
"not-invented-here" syndrome, which is marginally acceptable when we're
talking about "reinventing" existing technologies, but I'd hate to see
that applied to something so important as representing the thoughts
and ideas of humanity, especially as we move forward with computer 
systems that will operate on those ideas.

Not to pick on the Scientific American article, but the *very* first
thing I looked at was the diagram on page 39. The description on the
facing page says 

  "Elaborate, precise automated searches will be possible when
   semantics are widespread on the Web. Here a search program 
   correctly locates a person based on an assortment of partially
   remembered knowledge..."

There's about a half dozen errors in that sentence. "Semantics" are
not ever going to be widespread on the Web, information is (now I
understand why Sam Hunting hates that word). Computers don't 
"remember" anything, they store it, and they don't store "knowledge"
they store data. And the search doesn't locate a person, it locates
some information about a person.

I realize this is simply a magazine article, but these kinds of 
errors seem so prevalent (the latter is a design error in RDF) that
I worry that we have an uphill battle in trying to design things
correctly. I think we're going to really rely on our "elders" in
the field to point out these types of category errors. And not to
pick on the W3C here either. Their ability to bring people together 
based on their visibility can't be overstated. I hope they act
responsibly given that power.

Okay, so I've stated my fear. What might be my vision? 

I think there are *at least* two communities that need each other right
now, those who have long directed research into the study of knowledge
representation, and those interested in designing and implementing
computer systems to represent human knowledge. 

I see quite a number of activities, and it's unlikely that we'll see
coordination between them all (due to philosophical, political, language
and other differences), but we can still try to glean from them useful
design ideas. I don't see that coordination occurring necessarily in 
any specific vendor forum, rather between the gamut of *individuals*
that compose those various communities. We need to reach out to everyone
to bring these people together, to at the very least gain their feedback
in locating when we've strayed a bit off the path. We'll of course 
receive conflicting ideas given that these communities don't all agree
with each other. I hope whatever fora are chosen for this communication
doesn't exclude anyone (by virtue of membership, cost, etc.).

And in the end, it won't be perfect, but that's okay. I agree that we 
can't really see where this will lead, but I'm never comfortable stepping
entirely into the dark. Even a little bit of light (ie., direction) is a
good thing. I've not seen anything that begins to flesh out an architecture
for this sort of thing, perhaps save Doug Englebart's OHS or Ted Nelson's
Xanadu (not that I think they are particularly comparable). That architecture
is what I'm trying to point out is missing.

> Let's compare the work done in the last 15 years to allowed computers to
> communicate with each other through worldwide networks and the semantic web
> initiative. I don't think anyone had a clear schema of what will be the
> final result of all the work done on very different level to allow computer
> to exchange so easily, but all the partial initiatives and a very general
> schema like "lets all those lonely computers connected to each other" gave a
> good result.
> 
> I think the same kind of think will happen for the Semantic Web : a schema
> like "lets all those documents organized, linked, sharabled and processable
> by smart soft and by humans" should be enough to get somethink quite
> interesting and usefull.

Yes. We've assembled some of the best and brightest in the "extended room,"
so I expect we'll see real results, but we can gain some measure of economy 
if we have some idea where we're headed. Above I mentioned the idea of 
listening to our "elders", by which I don't want to imply any advanced
age but rather that we listen carefully to those who've tread these paths 
before (the same advice I'd give anyone on any path). An avant garde still
must intercommunicate, and in fact the most progress we've made as a race 
has come about when there was the most intercommunication. So if history 
is any indication we're likely to be successful; we've certainly improved 
our ability technically to intercommunicate, at the very least. Hopefully
we have gained in our ability to also communicate.

Now I've begun to ramble. Must be my stomach calling... 

Murray

...........................................................................
Murray Altheim                            <mailto:altheim&#x40;eng.sun.com>
XML Technology Center
Sun Microsystems, Inc., MS MPK17-102, 1601 Willow Rd., Menlo Park, CA 94025

      In the evening
      The rice leaves in the garden
      Rustle in the autumn wind
      That blows through my reed hut.  -- Minamoto no Tsunenobu

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

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

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