RE: Metonyms, # or /, mimetypes, named graphs

[...]
> 
> What is a Roy exactly, a document?  A missing document?

A Roy is the abstraction that relates the different representations of a
document.  Its what an implementor of a web server might think of as a
resource. I tried to be clear in my original message [1] but clearly failed:

[[
So get maps URL x MimeType x Time x ... Other html headers to SEQUENCE of
OCTET x MimeType x ReturnCode.  (I'll have the details wrong, but I hope
this is close enough).  So lets say that mapping is composed of:

GET1: URL -> ROY
GET2: ROY x MimeType x Time x .. -> SEQUENCE of OCTET x MimeType x
ReturnCode
]]

I've split GET into to parts; GET1 maps to a ROY and GET2 maps to a
represententation.

> 
>  <snip>
> 
> > Can xsd:int denote a datatype and xsd:int identify on the 
> web a ... Uhmmm
> > what?
> 
> > Answer A) GET1(url) need not equal Denote(url). ...
> > 
> > Answer B) GET1(url/) must equal Denote(url/) but GET1(url#) 
> need not equal
> > Denote(url#).  ...
> > 
> > Answer C) GET1(URL) must equal Denote(URL).  ...
> 
> Answer A is clearly correct.  In fact, it would be very unusual for
> GET1(url) to equal the denotation of url.  For instance, the url for
> my web page might be taken to denote my web page.  But GET returns
> the characters that happen to be on my web page today.

Yes.  I was trying to distinguish between GET and GET1.  Does that make it
less clear?  It seems to be less clear to others.  And I did try to take
time into account in the arguments.

> 
> Of course, there is no precise characterization of what uri's denote,
> any more than there is for other sorts of names.  While some have a
> clear intended denotation, it can't be the case that they all do.  In
> other words, the semantic web has many models.  When I said that a
> certain url denotes my web page, I wasn't appealing to any formal,
> theoretical reason why it should denote that.  At best I was asking
> people to use it as though it meant that.
> 
> -- 
>                                              -- Drew McDermott
>                                                 Yale 
> University CS Dept.
> 

[1] http://lists.w3.org/Archives/Public/www-rdf-logic/2004Jun/0020.html

Received on Friday, 18 June 2004 03:11:31 UTC