RE: [xml-dev] Semantic web?

 ...

 [Andrew said]
Perhaps, in a broader context, an even more important aspect is the notion
of eXtreme Monitoring Language!

The article from TBL et al in Scientific American started with an example
where medical data which, in Europe at least, would be seen as confidential
information was passed around with what at least some would view as gay
abandon.

If machine processing of semantics is implemented we, as individuals, are
highly likely to lose control of the privacy of our personal information if
we cannot know or influence directly which parts of personal information
(and its meaning) is accessible to "Big Brother".

[Danny says]
The bad news is that this is the state of affairs we see now - our personal
information is to a large extent out of our control. There's a slight
difference in the approach to such information in the cultures on either
side of the Atlantic (and Pacific), on how far we trust private and/or
public organisations with the data, but with the current "war on terror",
whatever restraint there was has gone as far as governments are concerned.
The existing web is crawling with spam merchants, all too ready to lift and
abuse personal information.
Some of the current approaches to managing sensitive personal information on
the net (e.g. MS Passport & Palladium) are also worrying, in that they are
also driven by purely commercial interests. You mention the medical data
issue - the kind of outsourcing of data processing that goes on in the uk
National Health Service, where (often questionable) financial and political
concerns are primary isn't exactly confidence-building either.

The good news is that consideration of things such as security policies are
effectively written into the "semweb constitution", or to be more precise
woven into the principles of the W3C's Semantic Web initiative, and the
premises on which their working groups are operating. I'm not entirely
optimistic, but I certainly feel the Semantic Web, through improving
communication structures and empowering the end user in general at least
offers the opportunity for greater personal control of confidential
information. It's got to be one of the most aesthetically displeasing words,
but perhaps 'democratization' would sum this up.

 [Andrew said]
Another aspect of the Sem-Web which deserves more attention is the likely
distortion of focus on to meanings which are culturally narrow and which are
more readily definable. The history of science is littered with "realities"
which have been determined more by what we can measure than what is
necessarily important.

[Danny says]
Indeed. Though again localization is in the "semweb constitution", this is
relatively unexplored territory as far as the web is concerned, and I
personally think this could well be a problem for civilization (grandiose,
I?) or at least a  huge can of worms when it comes to implementation.

 [Andrew said]
I think there is a real possiblity that the Sem-Web will focus more on what
we can *define* rather than what is important. Many of the most important
aspects of life are intriguingly intangible and are likely to be resistant
to a culturally narrow standardization. Therefore there is a prospect of a
distortion of meaning (and therefore potentially of thinking) perhaps
unparalleled in the history of Mankind.

[Danny says]
Heh, well yes, I couldn't agree more. Though here I would once again point
to the current situation, where incredible cultural distortion is already
happening :  prior to a couple of weeks ago, when did we last see Bali
represented on television as anything other than a playground for
Westerners? The current web already offers quite a bit more access to
information than mass media spoon-feeding, and again I'm cautiously
optimistic that the Semantic Web will improve matters here too.

[Andrew said]
There are issues involved which we might do well to consider which require
us to lift our eyes (or nose) from the silicon grindstone. :)


Alas, it's back to the carborundum ;-)
Cheers,
Danny.

Received on Wednesday, 23 October 2002 13:11:06 UTC