W3C home > Mailing lists > Public > www-rdf-rules@w3.org > September 2001

Re: What is an RDF Query?

From: Pat Hayes <phayes@ai.uwf.edu>
Date: Wed, 12 Sep 2001 16:31:18 -0500
Message-Id: <p05101004b7c57932e4eb@[205.160.76.192]>
To: Sandro Hawke <sandro@w3.org>
Cc: www-rdf-rules@w3.org

>  > OK, never mind the terminology. The more substantive point is the
>>  idea that the meaning of logic is given in English. This curious
>>  misapprehension seems to be widespread. Since about 1945, the
>>  meanings of formal logics have been given in mathematical terms
>>  (usually set theory), not English. That is why it is possible to
>>  manipulate logic using machines. (If the computers had to first
>>  understand English, we wouldn't even be able to send email to one
>>  another.)
>
>This idea of computers "understanding" is probably taking us the wrong
>direction.  I think we'd agree that computers are simply physical
>objects which obey the laws of physics and are constructed so that
>those laws cause them to do useful things.

Yes; but then I think the same applies to human beings (omitting the 
theistic implications of 'constructed', of course). (I wonder if you 
know what a mare's nest of philosophical controversy you just 
entered, my friend..... :-)

>So the question in how did
>their builders (in terms of hardware and software) come to understand
>what would be useful (eg sending e-mail).  And that probably brings us
>back to the rdf-logic thread about people/children learn the meanings
>of words.  When I say "in English" I'm trying to say "however people
>communicate meaning to each other."  The fact that mathematical terms
>have been very useful for clear human communication is sort of beside
>my point, which is that I'm not asking computers to do any sort of
>understanding.

Ah, but I *am* asking them to do some understanding.  And I think you 
are, too. If we all want to take the declared aim of the semantic web 
seriously, then we all are. The whole idea, if you recall, is to 
provide markup not for humans to read, but for *machines* to be able 
to draw conclusions from, all by their mechanical selves, with the 
humans out of the (runtime) loop. And drawing conclusions from 
something is a kind of understanding of it.

>  > >How can we possibly standardize this?
>>
>>  ?? What's the problem? The usual way: form a committee, have it
>>  collect views, argue a lot, come to a decision, etc., then publish a
>>  standard. In fact its already been done several times (KIF 3.0, CGs
>>  (an ISO standard), IEEE-KIF (still going on, but..), etc.)
>
>"The great thing about standards is that there are always new ones to
>chose from."   Yeah.

Sure, and we probably need to invent a new one for Web use (paying 
attention to URI/L/Es and XML lexicalisation, etc.). But these all 
have a lot in common, and the syntactic variations are more like 
changing hubcaps than re-inventing the wheel, as Peter correctly 
emphasises.  Its a lot easier to write new parsers than new reasoning 
engines.

I'll drop the sociology-of-W3C line, its only going to annoy people 
needlessly. I just get a little impatient at a kind of 'let's invent 
something NEW today!' culture that I seem to often run into when 
dealing with folk with w3c.org in their email.  (Extrapolating 
backwards, maybe it has something to do with listening to "another 
brick in the wall" while in middle school. )

But you are right, I do tend to confuse staff with membership, and I 
should be more careful.

Pat

>Yeah.    Still, it beats paying too much attention to the news.

Amen.

-- 
---------------------------------------------------------------------
IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
phayes@ai.uwf.edu 
http://www.coginst.uwf.edu/~phayes
Received on Wednesday, 12 September 2001 17:31:27 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:53:09 GMT