Re: resources and URIs

>pat hayes wrote:
>>
>>  >pat hayes wrote:
>>  >
>>  >>Well, I think that to deny it would be a dangerous step. although
>>  >>foolish might be better. Look, to deny this would be to say that
>>  >>things don't exist until they are given a URI... ah, but now I read
>>  >>on ....
>>  >
>>  >They don't exist on the Web till they have a URI.  What's
>>  >controversial about that?
>>
>>  Well, maybe nothing.  What does 'exist on the Web' mean? What is the
>>  difference between "exists" and "exists on the Web" ? Are these
>>  related in any way? I guess I find it odd to talk of something being
>>  brought into (any kind of) existence by something that happens on the
>>  Web.
>>
>>  I take existence to mean actual existence, right? Being something in
>>  the actual world, as in "Sherlock Holmes does not exist".
>>
>
>"Sherlock Holmes" 'exists on the Web' http://www.sherlockian.net/ without
>existing in the physical world.

Well, take a look at that web page. It has a picture of two actors 
and a bunch of URLs. In fact I spent some time cruising around this 
site looking to see if I could find a fragID URI which might even 
slightly plausibly be understood as 'referring to' Sherlock Holmes. I 
couldn't find one anywhere, not even an HTML anchor on his name. 
Which makes perfect sense, because there is no *need* to use a URI to 
refer to Sherlock Holmes, in fact: he's famous enough already for any 
English-speaking reader to be able to understand the plaintext 
version of his name. So I will make a bold claim: in the terms of the 
TAG architecture and RFC 2396, Sherlock Holmes is NOT on the Web. 
There is of course a hell  of a lot *about* him on the Web, but (I 
will claim) he himself is not on it. Nor is there any particular 
reason why he should be, mind you. It would be easy to put him there, 
I guess, but nothing would be the better for it.

BUt to get back to the main point, it seems to me that this notion of 
'existing on the Web' is empty, meaningless. Since it does not 
connote existence, it just means 'being indicated by a URI'; and 
since URI's aren't really required by the actual architecture to 
indicate anything, it is vacuous. One could say everything that needs 
to be said just by talking about URIs, and never mentioning these 
virtual things that they are supposed to refer to, which need not 
exist at all and which are brought into a ghostly kind of Web 
"existence" merely by someone inventing a URI. They serve no purpose, 
and the idea that they must be 'there' and must be unique only get in 
the way of a clear discussion.

>If you would prefer to say that a 'concept'
>does not exist, then we can alternatively say that a resource _is not part
>of_ the Web until it is given a URI. Alternatively: A "thing" is not a
>member of the class of "WebResources" unless the thing is identified by a
>URI.

None of that makes any sense to me, Im afraid. It sounds like the 
Web, in your mind, is something completely detached from the actual 
world, a kind of mass hallucination or a world wide videogame.  The 
WebResources it talks, er, about have no connection to anything real.

Really??

>
>The world we are talking about here is defined by URI coordinates. In the
>same fashion we might define criteria for a thing to exist in the physical
>world e.g. as occupying a certain volume/location in x,y,z,t space etc. I
>think we can make a strong analogy between existence on the Web and
>existence in the physical world as long as we get our coordinates right.

I don't find the analogy compelling or useful.

>
>This is all wordsmithing, but in the context of defining architectural
>principles, it is important to get the base definitions correct.

Well, amen to that.

Pat


-- 
---------------------------------------------------------------------
IHMC	(850)434 8903 or (650)494 3973   home
40 South Alcaniz St.	(850)202 4416   office
Pensacola			(850)202 4440   fax
FL 32501			(850)291 0667    cell
phayes@ihmc.us       http://www.ihmc.us/users/phayes

Received on Monday, 21 July 2003 20:09:21 UTC