Azamat, Pavithra, and Paola, (01)
Before commenting on your notes, I'd like to emphasize that I
strongly support the goals of the Semantic Web. My major complaint
is that the original goals of supporting semantics have been lost
in a mass of syntactic detail. In fact, the single most misleading
aspect of the Semantic Web is the word 'web'. It places the focus
on technology rather than semantics. That has led some people to
make some horribly misguided assumptions: (02)
1. It is possible or even desirable to distinguish different
kinds of semantics for different kinds of applications. (03)
2. The WWW was virgin territory for which new standards and
technology could be defined without integration with and
a migration path from a half century of legacy systems. (04)
3. All previous work in semantics for databases, software
specification, artificial intelligence, and natural
language processing was irrelevant and could be ignored. (05)
4. Character sets are the foundation for syntax, and
syntax is the foundation for semantics. (06)
5. Unique identifiers, by themselves, will magically solve
the thorniest problems of semantics. (07)
6. Having unique identifiers that point to unique definitions
will magically cause billions of people to pay attention
to those definitions and use them consistently. (08)
Just a glance at history shows that none of these assumptions
have any validity whatever. No technology from stone carving
to supercomputers has ever made any fundamental change in the
ways that people think, talk, and behave. Technology might
speed up and facilitate many kinds of interactions, but it
never changes human nature. (09)
AA> What it is really missing here: standard ontology and
> semantics, this we are steadily repeating for a long while now. (010)
The most important developments were standardized terminologies
for every branch of science, engineering, medicine, law, business,
and the arts. Those developments began a century before the SW. (011)
There was a lot of work on technology and standards for knowledge
bases, databases, and software specification since the 1960s.
But the SW developers ignored that. In fact, RDFS and OWL are
a major step *backwards*. If you look at publications on knowledge
representation, there has been a noticeable *decline* in quality
since the 1990s -- largely because many of the new publications
focus on syntax rather than semantics. (012)
AA> It is more politics than science "...This Institute will help
> place the UK at the cutting edge of research on the Semantic Web
> and other emerging web and internet technologies." G. Brown. (013)
I agree. But the UK made drastic cuts in their universities
recently, which caused many good professors and young researchers
to leave. I don't know the exact amounts of the funding, but
I suspect that they cut much more than they are adding. (014)
AA> First, it makes a big difference to invent the web of data or
> the web of knowledge. (015)
Those are just buzz words. What really matters is what kinds
of logics, ontologies, methodologies, and supporting tools become
available. (016)
AA> Second, it is in the nature of humans to create something new
> destroying (replacing) the old things and structures. Sombart
> and Shumpeter formulated this habit as a critical socal/economic
> phenomenon, dubbed as "creative destruction" or rather destructive
> creation. (017)
Sometimes the new constructions are improvements, and sometimes
they're not. In the US during the 1950s, there was a strong push
for "urban renewal" and "slum clearance". That often meant
bulldozing thriving neighborhoods of working class people, and
putting them into "projects", which were huge buildings surrounded
by a little fringe of grass. Most of those projects were disasters. (018)
If you look at the most successful innovations of the past decade
-- technologies such as AJAX, Flash, and PHP or companies such as
Youtube, Facebook, Twitter, etc. -- all of them were started by small
groups of creative individuals, none were based on SW technology,
and none of them came out of government-funded "institutes". (019)
AA> Lets hope that the Web Science is a next generation metascience,
> and Sir Timothy Berners Lee is going to give the world another
> decisive innovation. (020)
Sir Tim was one of those creative individuals like the ones who
founded Google and other start ups. He deserves a great deal
of credit for that. (021)
But remember the "second system syndrome" as stated by Fred Brooks:
"Beware the man who is designing his second system." Instead of the
original small, clean little system, he succumbs to the temptation
to put in all the features that he didn't have the time or funding
to include in the first system. Check Google for "second system". (022)
PK> I do believe that semantic web has been in operation in some form
> or an another already! I do not agree that it has not taken off... (023)
Yes, there are some applications. But the rate of adoption and
development of the SW technology has been painfully slow compared
to the speed of the original WWW. Other technologies like AJAX
and PHP grew much, much faster than RDF & OWL, even though
nobody was promoting them. (024)
As I said before, commercial web sites, large and small, were built
around a relational DB. If the SW had been better integrated with
that technology, it would have been adopted very rapidly. (025)
PK> Google already provides some aspect of it... (026)
No. Google does not use RDF and OWL because they are too
slow and bloated for their large volumes of data. (027)
PK> All collaboration and social networking web tools, Youtube,
> rss feeds, GIS and GPS and recovery.org... etc all representation
> of Semantic web. (028)
No. They don't use SW technology. (029)
PK> Help during Haiti earthquake was one of the fine example of
> what Semantic web did for humanity. (030)
They used the WWW and cell phones. The SW was not involved. (031)
PDM> I think there is widespread agreement that more contributions
> from information scientists would be advisable, (032)
I agree. But the most important first step is to drop the word 'web'
and replace it with 'systems'. (033)
What we must do is to integrate *all* systems on a semantic basis.
That includes every legacy system, and every future system that
anybody might develop using any kind of tools or platforms. (034)
John (035)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (036)
|