W3C home > Mailing lists > Public > www-ws-arch@w3.org > May 2002

RE: Web services and the Semantic Web

From: Eric Newcomer <eric.newcomer@iona.com>
Date: Sun, 26 May 2002 13:04:33 -0400
To: "Mark Baker" <distobj@acm.org>
Cc: <www-ws-arch@w3.org>
Message-ID: <FFEDIILGIAFELAMIJNMNOEAFDPAA.eric.newcomer@iona.com>
The ontology problem I was referring to was the Shoe Standards.org
reference, by which the Web can understand what is a shoe, what is a shoe
ordering service, etc.  I agree the Web is very powerful, but I still do not
understand the compelling reason to change something proven to work in favor
of something that remains theoretical -- for all the tremendous promise and
vision involved in the Semantic Web Activity, and I am actually a proponent
of it -- I do not see its practical application to existing Web services

Because you can do the same thing with RDF does not mean it's a good reason
to change.


-----Original Message-----
From: Mark Baker [mailto:distobj@acm.org]
Sent: Sunday, May 26, 2002 10:29 AM
To: Eric Newcomer
Cc: www-ws-arch@w3.org
Subject: Re: Web services and the Semantic Web

On Sat, May 25, 2002 at 10:33:14PM -0400, Eric Newcomer wrote:
> Mark,
> Your article assumes the ontological problem is solved -- I am saying I
> don't assume that the Semantic Web is yet reality.  However, Web services
> are.

In that example, what problem is assumed to be solved?

I did nothing more than this;

"interface ShoeOrderAcceptor extends OrderAcceptor"

where OrderAcceptor is defined as;

interface OrderAcceptor
  (status-code,headers,body) GET ( headers );
  (status-code,headers,body) POST( headers, body );
  (status-code,headers,body) PUT ( headers, body );

This grants something that knows how to interact with OrderAcceptors,
the ability to interact with ShoeOrder Acceptors.  That each interface
is exactly the same makes this even easier to automate, especially
between trust boundaries over a network, since it can be trusted.

> With regard to mapping I mean mapping to an executable program.  I don't
> anything in the RDF examples in your article that indicates how RDF is
> mappable to programs.  You are just showing that an endpoint can be
> described using RDF.

I'm still not clear on what you mean by "mapping to an executable
program".  Doesn't that interface above, the one bound to HTTP, do

> I am not sure what you mean by my not understanding the Web.

Not you specifically, just the industry at large.

I assume that every member of this WG has substantial experience
with distributed systems.  But I've heard a lot of things, both here
and elsewhere that demonstrates to me that people have not applied
any experience they might have to understanding the Web.

FWIW, four years ago I was in a similar position.  I had several years
of experience with OMA/OORPC architectures, and MPI, and I had just
prototyped a system using a tuple space architecture.  Shortly after
that, I decided to find out what made the Web tick.  Needless to say,
I was quite surprised by what I learned; that this thing that served
up porn was more powerful than any other system I had used before, and
was already sufficient for solving most problems that Web services tries
to solve.

Mark Baker, CTO, Idokorro Mobile (formerly Planetfred)
Ottawa, Ontario, CANADA.               distobj@acm.org
http://www.markbaker.ca        http://www.idokorro.com
Received on Sunday, 26 May 2002 13:08:31 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:40:56 UTC