W3C home > Mailing lists > Public > www-rdf-interest@w3.org > December 2003

RE: Trust, Context, Justification and Quintuples

From: Graham Klyne <gk@ninebynine.org>
Date: Sun, 21 Dec 2003 19:13:53 +0000
Message-Id: <5.1.0.14.2.20031221184224.00ba5ab0@127.0.0.1>
To: "Jos De_Roo" <jos.deroo@agfa.com>
Cc: www-rdf-interest@w3.org

At 13:34 21/12/03 +0100, Jos De_Roo wrote:
> > BTW, in N3, what do you think this means?:
> >
> > :id :- { :a  :b  :c  .
> >           :a1 :b1 :c1 .
> >           :a2 :b2 :c2 . }
> >
> > :id :- { :a3 :b3 :c3 . }
> >
> > (all in a single document.)
>
>I really don't know and have seen that ":-" only once
>before (in one of your mails) but can't find it in
>http://www.w3.org/DesignIssues/Notation3
>nor in
>http://www.w3.org/2000/10/swap/grammar/n3-report

Oh, I can't see it there either, but I'm sure it *was* there.  Notation3 
seems to be something of a moving target ;-)  The form :- does appear, for 
example, in:
   http://dev.w3.org/cvsweb/2001/blindfold/sample/n3.bnf?rev=1.4&content-type=text/x-cvsweb-markup

as:
[[
property_list ::= void   # to allow [...].
                 | verb space+ object_list
                 | verb space+ object_list space+ ";" space+ property_list
                 | ":-" anonnode  #to allow two anonymous forms to be given 
eg [ a :Truth; :- { :sky :color :blue } ] )
                 | ":-" anonnode ";" property_list
                 ;
]]

(which is linked from http://www.w3.org/DesignIssues/Notation3) as:
"n3.bnf  from the blindfold tools")

This happens to be a feature I use quite a lot, in the form:
   name :- node-expression
where I have understood it to mean that the name (or expression) to the 
left of :- is a referent for the think on the right.  Or, in the example above:
   [ a :Truth; :- { :sky :color :blue } ]
I think would mean the same as:
   { :sky :color :blue } a :Truth

I must confess that I have never seen this spelled out in detail, but I 
have had discussions with some people that tend to confirm this view.

I use it particularly in this form:

    ex:someName :- (item1 item2 item3)

to indicate that the first "cons" in the list is identified as ex:someName.

> >> I've never felt the need for more than {triples} names;
> >> those names are written on documents which have URI's
> >> and those URI's are the pivotal points.
> >
> > Er, I'm not following you here
>
>We *call* some-set-of-triples in 2 different ways
>1/ either as
>    { some-set-of-triples-in-notation3 }
>2/ or as
>    <uri-of-some-set-of-triples>.log:semantics

[I assume that is <uri-of-some-set-of-triples>!log:semantics per the 
current (2003-12-21) version of http://www.w3.org/DesignIssues/Notation3]

Ah, I think I see.  The difference here would seem to be in the use of 
log:semantics.  My take on what we *call* some set-of-triples would be:

1/ either as
    { some-set-of-triples-in-notation3 }
2/ or as
    <uri-of-some-set-of-triples>

I.e. that the URI and the {...} expression are interchangeable.

To my mind, this:
    { some-set-of-triples-in-notation3 }!log:semantics
would be equally valid.  Indeed:

   :someName :- [{ some-set-of-triples-in-notation3 }!log:semantics]

would be the same statement as:

   :someName log:semantics { some-set-of-triples-in-notation3 }

by my understanding of :- and !

>All that is written in particular documents
>which are URI identified and so we can
>explicitly and precisely load that in engines.
>There is a means to show what was loaded
>and thus trusted.

Yes... I think that name :- { ... } allows one to do the same thing within 
a single Notation3 "document".

#g


------------
Graham Klyne
For email:
http://www.ninebynine.org/#Contact
Received on Sunday, 21 December 2003 14:23:28 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:52:03 GMT