[
Lists Home |
Date Index |
Thread Index
]
On Jun 2, 2004, at 5:12 PM, Bullard, Claude L (Len) wrote:
>
> The part I did pay attention to was his position on where
> the semantic web is or will be, and on that, I agree. It
> occurs to me that the Semantic Web may be Berners-Lee's
> SGML/HyTime:
<grin>
> something that gets long in the tooth and
> awkward over time as it awaits a market, then when a
> global need for such comes along, someone strips it down to the
> basic necessary parts and succeeds wildly. Google is
> succeeding wildly based on what humans need. What humans
> need the machines to do with a semantic web seems to be
> the missing ingredient.
>
Oddly enough, ERH has written off the Semantic Web as hype at just
about the time I'm starting to take it seriously. Not that I disagree
with anything in his posting from the WWW conference, just that I'm
starting to see what I think were two missing ingredients start to
emerge.
One is a use case, such as Len alludes to: Google on the public Web
is good enough for most purposes that the Semantic Web was envisioned
for a few years ago, so why bother? The emerging answer is that
semantic integration within organizations is becoming feasible: There
is a need for all sorts of enterprise integration -- of data, of
services, of applications, of information -- and XML, TCP/IP, HTTP,
etc. have finally created the plumbing to allow this at a mechanical
level. But Google-like technologies only cut the search space down to
a size that a human can manage, they can't do this well enough for a
machine to process effectively. Businesses with automated processes
that need to know exactly which person, place, or thing they are
dealing with need something more reliable than heuristic matches
between information in different databases, they need provably correct
inferences about their identity. So, there is a real use case in
enterprise integration for ontologies that precisely define the meaning
of data in terms of other data, and for inferencers that can use this
information in a non-trivial way.
The other previously missing ingredient is that real organizations have
at least something approximating an implicit ontology in their database
schema, standard operating procedures, official vocabularies, etc. It
is at least arguable that the technologies that have emerged from the
semantic web efforts allow all this diverse stuff to be pulled together
in a useful way -- ontology editors, inferencing engines, semantic
metadata repositories, etc. I'm seeing real success stories in my day
job, and a coherent story is starting to be told by a number of
vendors, analysts, etc. (I just came across a bunch of stuff at
http://www.topquadrant.com/tq_white_papers.htm today that summarizes
the situation pretty clearly).
So, I don't think that the Semantic Web will replace Google on the
public web any more than SOAP/WSDL will replace RESTful interfaces at
Amazon.com, but in the more organized and managed environments within
enterprises I see these technologies having a real impact. Maybe the
Semantic Web vision will end up like other overly ambitions visions
such as AI [aim for HAL, get Google] or the space program [aim for the
stars, get GPS ] -- the technologies quietly get work done long after
the guiding vision is a quaint anachronism.
I do agree with the implication that it will be some stripped-down
essential subset of the semantic web technologies rather than the
complete corpus of academese that survives in the real world, even the
more controlled real world behind enterprise firewalls. Unfortunately,
I have no idea which subset that will be :-)
|
