W3C home > Mailing lists > Public > uri@w3.org > April 2003

RE: We don't need no stinkin' identity

From: pat hayes <phayes@ai.uwf.edu>
Date: Wed, 30 Apr 2003 10:26:59 -0500
Message-Id: <p05210601bad2f1bcd8b4@[10.0.100.12]>
To: Michael Mealling <michael@neonym.net>
Cc: Larry Masinter <LMM@acm.org>, uri@w3.org

(for some reason this didn't get sent a few days ago, sorry.)

>On Sat, 2003-04-26 at 19:25, pat hayes wrote:
>>  >First,
>>  >>  note that the _semantics_ of a URI are not defined in
>>  >>  this specification. Each URI scheme itself defines the
>>  >>  relationship between URIs of that scheme and the resources
>>  >>  they identify. In all known cases, that definition isn't
>>  >>  enough to allow the URIs of that scheme to be used, by
>>  >>  themselves, as unambiguous identifiers when trying to
>>  >  > make logical assertions.
>>
>>  Again, the problem is not that the logic requires unambiguity; quite
>>  the reverse, in fact: it is that imposing unambiguity as a defining
>>  characteristic of URIs is logically incoherent.
>>
>>  Suggested modification, without the logic-bashing:
>>
>>  >  What is a Resource? Can a URI be used to identify a Concept?
>>
>>  >  This specification does not define the word 'resource' carefully,
>>  >  nor does it define how a URI can be used to 'Identify' a
>>  >  'Resource'. The _semantics_ of a URI are not defined in
>>  >  this specification.  Each URI scheme itself defines the
>>  >  relationship between URIs of that scheme and the resources
>>  >  they identify or refer to.
>>
>>  The reason for the last three words is that it doesn't make sense to
>>  say that a URI identifies a *single* thing, in almost any logical
>>  language. (In fact, it doesn't make much sense in any language, IMO,
>>  but leave that aside for now.)
>
>Pat,
>   I think that's one of the places that we're falling down on our
>explanation. You assert that it doesn't make sense "in almost any
>logical language". And that's the point, URIs aren't _in_ a logical
>language, they exist outside all of them. URIs may be used _by_ a
>language but at that point it is incumbent on the language/system that
>uses them to define what it means concerning the things that URIs are
>allowed to identify.

But they ARE used in a logical language. I don't mean that they are 
*restricted* to such use, but they are used there (as well as in many 
other ways, of course). And I am not asking that this useage be a 
*restriction* on URIs or on what or how they mean what they mean. On 
the contrary, I'm asking y'all to avoid restricting their meanings 
too much. I want you to liberalize URI meanings, to set them free. 
The problem isn't that logic needs a tighter or more special notion. 
The notion that logic needs is about as bland and meaningless as you 
can imagine: anything can be denoted, and a name (URI reference) can 
denote anything, a priori; but if not, then the semantics have to 
specify the rules for saying what things cannot be denoted.  Which is 
why I was asking y'all to tell us what the rules are.  What you are 
saying is that ALL URIs denote something UNIQUELY, which is an 
incredibly strong assertion, as I said in  my message to Graham ; and 
you aren't telling us what the rules are, which isn't acceptable. 
("Social conventions" isn't a rule.) This really isn't anything to do 
with RDF, OWL, etc, by the way: that is an incredibly strong 
assertion to make about English, or German, or vector notation, or 
blueprints or maps.

You seem to think that everything was OK until these logics came 
along, and now we are rocking the boat.  I know the Web itself was 
OK, but what y'all were saying about it wasn't OK, and the boat would 
have been rocked in any case. The only point that makes it urgent for 
us is that on the SW, these matters suddenly become technical instead 
of meta-theoretical, and they have potentially dangerous consequences 
(things like friendly fire in combat, bombing the wrong embassy.) And 
logical semantics - model theory - in fact places about the weakest 
possible constraints on 'meanings' of expressions that it is possible 
for a theory to impose. It is like the minimal vanilla theory that 
all other semantic theories build on, or at least respect, including 
NL semantic theories, semantics of diagrammatic representations, of 
mathematical notations, etc etc. . So if you are saying something 
that violates *that*, then either you are rebuilding the entire 
foundations of semantic theory for all the world's known languages 
and notations - in which case I'd like to be able to read a few of 
the books or refereed articles that y'all ought to be writing to 
announce this intellectual revolution  - or else y'all have made a 
mistake somewhere. Not in the engineering of the Web, I hasten to 
add, but in the semantic/semiotic analysis of it that you have 
apparently assumed to be true.

(BTW, and only an aside, I think this doctrine on uniqueness of URI 
binding may arise  from a fear that if it were false, then the sky 
would fall.  It is false, and the sky hasn't fallen. So the right 
thing to do, seems to me, is to try to figure out WHY the sky doesn't 
fall even though URIs, like English words, are potentially ambiguous 
in their meanings. We have some glimmers of insight into how 
natural-language communication manages this trick.)

Now, I concede that this claim is likely to be more controversial and 
that y'all might not want to get into another discussion/debate over 
this; but wouldn't it be safer, as well as quite enough for your WG 
charter, to just say a tiny bit less about the issue? Three words 
less would be a good start.

>   As Graham has pointed out before, the problem here is that RDF, SW,
>OWL, etc have erroneously attempted to inherit RFC 2396's concept of
>Resources unmodified, and that was a severe mistake.

Well, I don't know if it was a mistake or not, because RFC 2396 
doesn't say what it means by 'resource' well enough to be able to 
tell. Which is another point than the above, and one which has 
nothing whatever to do with RDF or logical semantics, by the way, but 
is just a request for clarification.

But in any case, if you have a notion which is *incompatible* with 
model theory, then you are in very deep water already, never mind 
about RDF and OWL. This is like someone claiming that they are dong a 
new kind of physics which doesn't need no stinkin' mathematics. 
Maybe, but I beg leave to doubt until I see it proved.

>IMHO, it is
>probably worth inserting language into the new document suggesting that
>it is extremely dangerous to simply inherit this definition of a
>Resource without some sort of system specific profile of it. Systems can
>do it but they have to be very clear that they are inherit it, not the
>other way around...

But that is the way round that hurts. We aren't trying to impose any 
restriction on you, we are asking you not to impose on us.

>   To simply inherit this documents concept of a Resource would be the
>network equivalent of simply specifying an application layer (8)
>protocol directly on top of IP, ignoring its admonitions to use things
>like TCP/UDP/ICMP, etc that constrain IP to something that is usable by
>higher layers.

I see what you mean, but the issue here is that these notions of 
'bound to' and 'indicates' have got to do with more than network 
protocols, now. Along the way (somewhere close to where URLs morphed 
into URIs), they got inextricably mixed up with issues of 
representation, meaning and denotation.  And now you have to take 
account of more than just protocols: you are doing semiotics now, 
whether you want to admit it or not.

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@ai.uwf.edu	          http://www.coginst.uwf.edu/~phayes
s.pam@ai.uwf.edu   for spam
Received on Wednesday, 30 April 2003 11:27:06 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 13 January 2011 12:15:31 GMT