W3C home > Mailing lists > Public > www-rdf-interest@w3.org > August 2004

Re: [seweb-list] A discussion: Is semantic web an old fashioned idea?Is it bubble, unworthy or an interesting research area - Post your c omments

From: Charles McCathieNevile <charles@w3.org>
Date: Mon, 30 Aug 2004 12:01:29 -0400 (EDT)
To: Adrian Walker <lewis_arise@bumerang.ro>
Cc: semanticweb@yahoogroups.com
Message-ID: <Pine.LNX.4.55.0408301131560.11560@homer.w3.org>

[moved the spam-style cross-posts to Bcc. If anyone is interested outside
semweb@yahoogroups and doesnt want to follow up there, feel free to discuss
wherever. But please don't dump massive threads onto multiple lists. It is
easy enough to put a reference to a nicely-threaded archive (such as
www-rdf-interest has) in any other list.]

On Mon, 30 Aug 2004, Adrian Walker wrote:

>However, I would like to add that it was not the "Semantic Web" as such
>which helped us in succeeding so, but rather particular tools with some
>"semantic awareness", which have been operationally used.

Yes, this seems an intelligent way to describe what happens.

It seems there is a gap in terminology (which is common). The "Semantic Web"
as I understand it is not some magic object that will appear on the annointed
date, it is a shorthand term for describing a Web that has more tools with
"semantic awareness", and more data that they can use. In discussion people
tend to seek shortcuts for long complicated descriptions - unfortunately
people don't always manage to convey the idea behind the shortcut, and others
make their own interpretation of what was said, and run off at the mouth (or
keyboard) on some path that is founded on misunderstanding. Which mostly just
serves to compound misunderstanding.

...
>-Epaminondas
>
>
>Peter Crowther wrote:
>
>>>From: Jean-Luc Delatre [mailto:jld@club-internet.fr]
>>
>>Well... as someone who was on WebOnt and who founded a company (and got VC
>>funding) based partly on ideas about the Semantic Web, I've got a few
>>comments on that.  Note that I left the company (and WebOnt) at the end of
>>2002 due to health problems.  These are a fairly mixed bag, I know; let's
>>see if they add some fuel to the fire.  Note that these are all personal
>>opinions; they may or may not be the views of W3C, any other member of
>>WebOnt, any member of any company I have ever worked for, or indeed any of
>>my cats.
>>
>>3) Far too much faith has been placed and is being placed on one
>>architecture slide that Tim Berners-Lee created, showing RDF as the
>>substrate for the entire Semantic Web.

Indeed. This slide was a quick attempt at explaining how the semantic web
might work, as part of a presentation. It's easy to see that it isn't taken
as "the vision", since Tim himself bases a lot of his work on CWM, a tool
that didn't use XML (nor Unicode, if I recall correctly) in its early
versions, and is still not based on XML (If you look at the slide, the real
bases are URIs, XML, Unicode, ...)

This is a classic example of people taking a little bit of the message and
confusing it for the whole. As Tim explains the slide it probably gives a
reasonable explanation of his understanding of the Semantic Web (at least to
him it seems reasonable). Others explain it (and for that matter use and see
it) in slightly different ways.

>>This slide should have been ceremonially burned long ago.

It's hard to burn something that exists only in an abstract representation
:-) On the other hand there are people in W3C who stopped using the slide
long ago, because their manner of explaining things was slightly different.
(If you think you can represent the required architecture to build a
semantic web semantic web in one static image, I'd love to be able to present
that. I have always found it just too difficult, and use a sequence of
stuff).

>>Unfortunately, it's probably too late to change now - at least within the
>>W3C-supported Semantic Web initiative.  RDF (any version) is far too
>>limited in its expressive power to be a useful substrate, and the idea of
>>building all the other layers on top of it is akin to trying to build a
>>communications framework on Morse code when you have dots but no dashes.

Oh? I would have thought that the modern digital voice/data network
infrastructure, built as it is on binary formats (dots and blanks, unlike the
ternary format of Morse which has dots, dashes and blanks) counted as a
reasonably useful communications framework.

RDF isn't the entire system - you have to build things like trust management
(as you noted - and I agree with you that one of the things standing in the
way of realising a "semantic web" is the fact that work in this area is
really just beginning). I believe that you can express this kind of thing in
RDF, but that to make it work you have to standardise how you interpret some
more stuff. If you look at the development from RDF syntax (in the 90's) to
RDF+OWL (this year) there is a pattern of doing just that. It's clear that we
need a querying system, and it seems clear that there is value in having a
standardised one. Several years experimentation in querying seems to have
provided a basis for rapid and effective standardisation, and I expect the
DAWG working group to produce a useful result much faster than the production
of the syntax. (Note that some light modification was required to the
5-year-old version, which became apparent with deployment experience).

We can in fact run before we learn to crawl. But we're likely to fall over a
lot more often. That's fine for experimentation, but not very useful for
people who are trying to deploy large-scale systems fundamental to their
work, who rely on the somewhat slower process of standardisation to give them
something with the obvious wrinkles ironed out through experience.

It seems to me (having built and used some "semantic tools" that some things
work pretty well already, and there seems no reason not to believe that they
will continue to scale nicely. It seems that there are other aspects that are
more complex. Trust management isn't that tricky in concept. But building
user interfaces so ordinary people can actually cope with it is something I
see as being at a very rudimentary stage. To reach the full potential of the
Semantic Web includes solving this problem, but there are a large number of
very useful things we can do before then...

-My personal opinion. Not endorsed by W3C (different folks there probably
agree and disagree in slightly different ways with different parts of what I
have said).

cheers

Chaals

--
Charles McCathieNevile  http://www.w3.org/People/Charles  tel: +61 409 134 136
SWAD-E http://www.w3.org/2001/sw/Europe         fax(france): +33 4 92 38 78 22
 Post:   21 Mitchell street, FOOTSCRAY Vic 3011, Australia    or
 W3C, 2004 Route des Lucioles, 06902 Sophia Antipolis Cedex, France
Received on Monday, 30 August 2004 16:01:29 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:14:57 UTC