Re: An intuition pump

* Jim Hendler <hendler@cs.umd.edu> [2003-09-23 08:48-0400]
> 
> I remember that when I first learned that I would someday receive a 
> "Doctor of philosophy" degree (instead of Dr. of Science) I was quite 
> disappointed.  However, I then learned that pragmatism was a 
> legitimate branch of philosophy and having read Dewey and a few other 
> folks like that [*] I decided I could live with ...
> 
> SUMMARY:
> I bring this up because in my simple, and somewhat pragmatic world 
> view, the current discussion makes me uncomfortable - it borders on 
> "aesthetic" rather than practical -- what I'd really like to see is 
> if we could expose a couple of "test cases" where it makes a 
> difference.
> 
> DETAILS:
> 
> Let me try this:
>  I see URIs as a nearly infinitely large addressable memory - we have 
> a naming convention to avoid simply numbering the spaces -- i.e.
>  http://www.sally/RDFDocument#person is just a short form for
> "Web memory location 33290948209842398"
> we then have bits that live inside these memory locations.
> 
> Stating the obvious, it seems to me the goal of this group is to say 
> something about the ownership of these memory addresses and the 
> relation of that to what is contained in the locations.
> 
> So Mary's person and Sally's person are different boxes -- the fact 
> that Mary says in her box that Sally is wrong in hers seems to me to 
> be a difficult issue in the aesthetic world of words and meanings, 
> but not in the pragmatic world of boxes and their contents -- as long 
> as my machine can tell me whether I'm looking in Mary's memory 
> location or Sally's, then I can tell who is doing the asserting.  If 
> Mary asserts something about Sally who asserts something about Bijan 
> who asserts something about Pat, etc -- I don't have a problem.
> 
> Seems to me the issue of what is in the boxes is different -- in my 
> naive way I read the TAG issue as "Is the owner of a box responsible 
> for what it says in that box"   (and what if that points at another 
> box)
> 
> I guess in my stupidly naive way, I don't see why this is drastically 
> different than if on my Web Page I claim (in words) Bijan is my 
> employee and on his Web Page he claims (in words) that he is my 
> underpaid slave.    We each are able to state our intended meanings, 
> someone pointing to my page can tell the difference from if they are 
> pointing to Bijan's page, and those referring to things on the pages 
> have to be careful they point at the right place.  I can steal 
> content from Bijan's page, I can cut and paste from Bijan's page, the 
> one thing I cannot do is "coopt" Bijan's page in any way that will 
> fool your software.
> 
> (I admit that I could fire Bijan, keep the URI he has, and change 
> what it says - but that seems to me to be an extreme case and one way 
> beyond the call of our software to handle. 

Bijan should PGP sign his HTML and RDF pages, in that case...

eg http://rdfweb.org/2003/09/mirrors/jetblue/jetblue.rdf
http://rdfweb.org/2003/09/mirrors/jetblue/jetblue.rdf.asc

Dan

Received on Tuesday, 23 September 2003 09:02:16 UTC