W3C home > Mailing lists > Public > public-sweo-ig@w3.org > November 2007

RE: SWEO - Thoughts, Not Slides

From: Wilson, MD \(Michael\) <m.d.wilson@rl.ac.uk>
Date: Tue, 20 Nov 2007 22:04:11 -0000
Message-ID: <677CE4DD24B12C4B9FA138534E29FB1D02A38551@exchange11.fed.cclrc.ac.uk>
To: "David Provost" <dprovost@metatomix.com>, <public-sweo-ig@w3.org>
David,

 

Also thoughts -

 

A key argument for the CIO audience is represented in the SW layered
pyramid (whatever layers) in contrast to previous AI/expert systems
solutions. 

 

Each step provides a solution to a specific problem, and they are the
progressive problems of organizations as they grow in size and maturity
of IT/IS systems. OMG have tried similar arguments for their model based
design & technologies. SW uses the Web as the starting point (the
outward face of an organization) rather than software development (an
internal, organogramatically minor part of most organizations).

 

SW technology builds in stages where the skills of staff and the cost of
tooling is not wasted at each step, but each step adds to the skills and
tools that the staff already have, and add new functionality at the same
time. Each step is accessible to staff who understand the previous step,
and they can grow there skills while building on their understanding of
the complexity and risks involved.

 

The cost arguments from this are that relatively small investments move
up each step yielding identifiable benefits, rather than the glory
arguments that a single vast investment will yield everything. The risks
of investment can be managed progressively, and the Return on Investment
(ROI) on each stage identified and used to justify the investment in the
next step. 

 

So the SW is designed to link risk management, investment strategy, ROI
and maturity of IT/IS into a single progression.

 

Prof Michael Wilson

Manager, W3C Office in the UK & Ireland

STFC Rutherford Appleton Laboratory, UK

http://www.e-science.stfc.ac.uk/organisation/staff/michael_wilson/

 

________________________________

From: public-sweo-ig-request@w3.org
[mailto:public-sweo-ig-request@w3.org] On Behalf Of David Provost
Sent: 20 November 2007 21:46
To: public-sweo-ig@w3.org
Subject: SWEO - Thoughts, Not Slides

 

All -

I've been thinking and working on my presentation ideas and seem to be
ending up with questions and thoughts, not slides. My thoughts are still
gelling, but off the top of my head (all assuming a business/CIO/CTO
audience):

1. Businesses buy solutions to problems. They only buy technologies if
they plan to use them to either solve problems or build solutions for
same. (Companies don't buy Visual Basic or C, they buy MS Office.)

2. There is no reward for buying the latest technology for its own sake.
There is a very tangible reward for buying things that are better,
faster, cheaper, more flexible/adaptable, etc. As a matter of fact, some
might argue toward buying older, more proven technologies that are more
stable.

3. Since 2003 I've seen presentations from engineers and from business
people (who are painfully lacking in background) that focus more on
>how< a problem is solved rather than how the solution is >so much
better< than what's presently available - which leads to...

4. Economists use the term "substitute." An example is "My old car is a
fine substitute for a new car." Separately, Gallium Arsenide is an
excellent basis for microchips. Silicon hasn't been widely displaced
because it continues to get better and cheaper (Wikipedia does a great
job of summarizing this.) I believe there are substitutes for a lot of
what Semantic technology can do and that it's fair to say these
substitutes may or may not be better in closed- vs. open-world
environments. (The idea of data warehouses sounds great, but in
practice, their creation, maintenance, and usefulness can leave a lot to
be desired.)

This leads to a central point in my thinking: When Web adoption became
mainstream, many corporations were caught severely lacking. (I believe
the decision to adopt the Web was simpler - you needed a computer, a
modem, a browser, a few other basic ingredients and you were in
business.) Aside from selling specific solutions to specific problems,
how can corporations be convinced to broadly invest in a non-proprietary
technology now, so that they aren't left behind again? Or is the correct
approach to highlight solutions first, and mention that Semantic
technology powers them second? Perhaps a starting point is to pose the
question "What are the most mature aspects of the Semantic Web and what
problems are these aspects best suited to solving?"

I've attached a presentation that portrays corporate IT portfolios as an
investment portfolio, which takes the form of a pyramid. The most risky
investments are at the top, which means that only a small amount of
money is being put at risk. Thoughts on the usefulness of this
perspective might be helpful.

5. All this leads me to identify those things that SemWeb technology
relies on as its main claims: Data integration, application
interoperation, reasoning, and the fact that it's Web based. My approach
to SWEO would be to ask "Why is a SemWeb approach to these things better
than existing substitutes?" I'm convinced these points can be broken
down in convincing fashion, it's just that I think I'm realizing I need
to discuss these issues/questions, not assert a position.

I'm looking forward to discussing these points on tomorrow's SWEO
conference call.

David


David Provost
Product Manager, Platform
Metatomix, Inc.
3 Allied Drive, Ste 210
Dedham, MA 02026
*781.254.2769
dprovost@metatomix.com


This email and any attachments thereto may contain private,
confidential, and privileged material for the sole use of the intended
recipient named in the original email to which this message was
addressed. Any review, copying, or distribution of this email (or any
attachments thereto) by others is strictly prohibited. If you are not
the intended recipient, please return this email to the sender
immediately and permanently delete the original and any copies of this
email and any attachments thereto.
Received on Tuesday, 20 November 2007 22:05:13 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:28:58 UTC