RE: HM.applications-Translations

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

The point of single system is single representation. 
DAML+OIL is a system. *(any language is)*.  XML is 
a system.  It is a metalanguage.  RDF is such and 
so on.  That isn't an accusation.  It is a fact. 
But it is DARPA driving the train now, now TimBL. 
(which is ironic given they built the Internet 
and funded almost everything remarkable since.)

The unremarkable part is not because it won't work; 
as all the years of research in Prolog, expert systems 
etc. proved, in some cases and at some scales, it will 
work.  But just as those systems demonstrated, there 
are representation exchange issues (what the SW is 
solving as best as it can be given representations 
have to be adequate to task) and there are unknown 
scale problems.  

Then there are deontic logic issues:  propriety.
Given what can happen with non-authoritative 
or dated sources, the idea that the IRS collection agent 
and the local police agent can confer with my lawyer 
agent and conduct an audit at night then dispatch real 
human fetchers to my house in *internet time* is not 
something I like to think about.  You probably don't 
either.  But if you want to be an SW supporter, name 
out front and all that, you don't want to be the 
next Henry Blodgett either.


Len 
http://www.mp3.com/LenBullard

Ekam sat.h, Vipraah bahudhaa vadanti.
Daamyata. Datta. Dayadhvam.h


-----Original Message-----
From: Sean B. Palmer [mailto:sean@mysterylights.com]
Sent: Thursday, September 06, 2001 3:15 PM
To: Bullard, Claude L (Len); Mark Brownell; gurun@acc.umu.se;
humanmarkup-comment@lists.oasis-open.org
Subject: Re: HM.applications-Translations


[...]
> Again, the architecture of the Semantic Web is fairly
> unremarkable.

Actually, it's *very* unremarkable.

> All it really is is an attempt like more Berners-Lee projects
> to limit representation to a single system.

I don't think that's true, but it depends on how you define "systems".
Is/are the person(s) responsible for the "book" to be accused of trying to
limit representation to a single system? I don't think so: the SW is as
non-constraining as it can possibly be. I don't expect its scope ot grow as
large as HTTP, or the printed word, and I think that anyone who does is
being foolish, but I do expect it to be able to do simple queries, merging,
inferences, and other stuff that I can't be bothered to do. Too many people
use the phrase "when/if the Semantic Web works" - I've got news for them:
it's already working. I've only ever really had two uses for it (a big
database merge that would have taken ages manually or using any other
software, and a cool little server logs program that I wrote), but others
should find more uses. There's just this huge cloud of misunderstanding
which hangs over the topic and pisses down on us all, and it's getting
annoying.

> There are pluses and minuses to that just as the myths
> of naming, location and addressing he promoted have
> advantages and disadvantages. [...]

Yeah, some of that stuff is nuts. But I don't think that anyone in the
world truly understands the axioms behind URIs/naming/location fully. Too
many disagreements, too many problems. But it all sticks together somehow,
so who cares really?

--
Kindest Regards,
Sean B. Palmer
@prefix : <http://webns.net/roughterms/> .
:Sean :hasHomepage <http://purl.org/net/sbp/> .