W3C home > Mailing lists > Public > www-rdf-logic@w3.org > February 2001

Global identifiers (was: universal languages)

From: Graham Klyne <GK@ninebynine.org>
Date: Thu, 08 Feb 2001 13:31:43 +0000
Message-Id: <5.0.2.1.2.20010208132238.041a6640@joy.songbird.com>
To: Stefan Decker <stefan@db.stanford.edu>
Cc: <www-rdf-logic@w3.org>
At 10:32 AM 2/6/01 -0800, Stefan Decker wrote:
>Probably you are right. But let us first establish the necessity of having 
>global IDs.
>Then we can think about how they can be engineered and
>what useful things we can do with them.

I suspect that global IDs are not so much "necessary" as "useful for 
performing certain feats".

One could imagine a web in which all relationships were expressed in terms 
of non-unique names.  One takes into account the place in which a reference 
occurs (its context) to decide that to which it refers. I fact, I think 
this is pretty much what you have when you walk into a library.

But, such an implementation would not scale so easily to a world-wide 
network of instantly accessible information.  (Note I don't say "not 
possible", just "not so easy".)

So I think the bottom line is that URIs are good because they give us a 
means to implement some aspects of our ideas, not because they (or any 
other global ID) is fundamental to the meaning of life, the universe and 
everything.  By weaving them into our particular view of life, etc., then 
we create possibilities for more easily implementing aspects of that view.

So it appears that I take a contrary view to yours:  global IDs (URIs) are 
adopted for their present engineered status and what we can do with them, 
not because they're necessary.

#g

------------
Graham Klyne
GK@NineByNine.org
Received on Thursday, 8 February 2001 09:22:09 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:52:38 GMT