Ian Dickson on Tue, 30 Dec 2003 00:54:33 +0100 (CET)


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: <nettime> the semantic web for beginners


In message <200312271829.hBRITSC31597@bbs.thing.net>, t byfield 

<tbyfield@panix.com> writes
>or: M*ke M*ney F*st as a part-time ontological envelope-stuffer!
>
>even boiled down to plain-vanilla ascii, this is *the* best discussion 
>i've seen about the chimerical 'semantic web.' i'm sending this version 
>to nettime as a sort of nostalgic doff of the hat to 'collaborative 
>text filtering' (which doesn't seem to be doing very well these days, 
>at least not in this neighborhood), but don't miss the page itself:
>
>     < http://poorbuthappy.com/ease/semantic/ >


I love all this Semantic Web talk.

Mainly because I haven't seen so much of a gap between theory and 
reality since the 5 Year Plans of Stalin's era.

I am involved in the Semantic Web, at a 100% commercial level, which 
gives me an interesting perspective. (I need to build something that 
works. Theory is interesting, but it doesn't feed the family).

For those interested here are some observations:-

1) Most people involved in the SW have completely failed to define what 
the SW is.

2) Given this lack of definition, it is not surprising that progress is 
slow.

3) Irony time - a Semantic Web that is undefined?

4) There are two core attitudes re SW issues.

a) That the SW will be a global all enveloping thing. Call this SW 
Major.

b) That is will be local in nature, helping domains of 
interest/expertise communicate more effectively. Call this SW Minor. 
(Some would call it SW Trivial and say that it's little more than XML ).

5) Many people think that SW Minor will evolve into SW Major.

Which camp you join depends on whether you think that the boundaries 
between the SW Minors will be discontinuities or blended.

6) Techies are in charge.

Which is why precious little actual progress is being made. The problems 
are not IT technical, they are linguistic and psychological. The tech 
issues arise from the L&P ones, not the other way around.

These L&P issues are the SW eqv of the elephant in the corner of the 
room - everyone tip toes around it and pretends it isn't there.

In essence I maintain that for the SW to work there needs to be a 
dictionary of meanings that are agreed by all participants in the SW, 
and that which also fits the mental constructs of the people who will be 
using the SW.

Building such a dictionary is hard work and tedious, (I should know, I'm 
writing one). It's not clever code. And for even modest SW Minor 
implementations (such as ours) it can amount to hundreds of thousands of 
entries, even millions. Because of this it is also essentially 
impossible to build it by committee.

Also any true SW will reflect the semantics that make it up.

This can be seen today by doing web searches, and the multiplicity of 
conflicting results that appear.

Try answering the question below. Think about what sort of answer you 
would expect from the SW, if it existed.

"Where can I find Hotblack Desiato"?

7) DreamTime

I CAN envision a world in which computing power has got to the level 
that enables it to do real time global textual analysis on the entire 
caucus of texts, and thus in effect boot strap an SW Major.

But if you give any thought to that at all you'd see that way before 
such a machine delivered the SW, it would first deliver flawless 
translations between languages.

Given the state of automated translation I'd say that a technical 
solution grounded in computing power alone is a  few years away:-)

Happy 2004.

PS, Thought for 2004 - Does the globalisation of the anti globalisation 
movement signify that that the anti-globalisation movement has failed?
-- 
ian dickson                                  www.commkit.com
phone +44 (0) 1452 862637                    fax +44 (0) 1452 862670
PO Box 240, Gloucester, GL3 4YE, England

           "for building communities that work"

#  distributed via <nettime>: no commercial use without permission
#  <nettime> is a moderated mailing list for net criticism,
#  collaborative text filtering and cultural politics of the nets
#  more info: majordomo@bbs.thing.net and "info nettime-l" in the msg body
#  archive: http://www.nettime.org contact: nettime@bbs.thing.net