W3C home > Mailing lists > Public > www-rdf-interest@w3.org > July 2002

Re: Minus sign in ids Was: Alternative N3 Parsers for CWM

From: Tim Berners-Lee <timbl@w3.org>
Date: Tue, 16 Jul 2002 16:58:16 -0400
Cc: "Aaron Swartz" <me@aaronsw.com>, "Sean B. Palmer" <sean@mysterylights.com>, <www-archive+n3bugs@w3.org>, "Dan Connolly" <connolly@w3.org>, www-rdf-interest@w3.org
To: "Seth Russell" <seth@robustai.net>
Message-Id: <BFC7829A-98FE-11D6-9786-000393914268@w3.org>


On Monday, July 15, 2002, at 11:51 AM, Seth Russell wrote:

> From: "Aaron Swartz" <me@aaronsw.com>
>
>>
>> On Monday, July 15, 2002, at 10:03  AM, Tim Berners-Lee wrote:
>>> Hmmmm...   Are you  *sure* you don't want to be able to write "x-1"?
>>>
>>> { y > x+1} => x a foo:Interesting}.
>>>
>>> Note that currently   x-1  as an XML name becomes x__1 in n3 .
>>
>> I'd be willing to write  "x + 1" or "[x + 1]" or something instead so I
>> could use ical:WHEN-DATE without name mangling.
>
> Then how would you say that "x + 1" means the sum of x and one ?  We 
> still
> can't (dont't want to) use quoted literals as subjects in n3, can we 
> (do we
> want to)?
>

Let's be clear at this proposed syntax, clear for which I wanted to keep 
the way.
(which I wanted to keep the way clear for).

x + 1  would be short for   [ is math:sum of (x 1)]
The BNF would have to become a typical expression
  sum :: factor : sum * sum
and so on

I'd like to keep it open, anyway,

BTW, The other thing which I am must more eager to put in is

my:event.cal:start.cal:timepart.time:hours
as shorthand for
[ is time:hours of [ is cal:timepart of [ is cal:start of my:event]]]

The expression above could, incidentally, be written

(x 1).math:sum

using this "rdfpath"  :-)  feature.

Re quoted literals as subjects, I want to. I do. cwm accepts it.
It is more symmetrical, which is simpler, and good.
The lack of literal as subject is IMHO a bug in RDF 1.0

"0" cal:englishDayOfWeekName "Sunday".
"1" cal:englishDayOfWeekName "Monday".

(  Also, while a given system of logic may have axioms that
for example literals and predicates are distinct, I don't think that this
should be enforced by the N3 base syntax necessarily.
Certainly bnodes can be predictaes as in
:joe [ is daml3:transitiveClosue of :parent] :alan.  )

Tim


> {{y > xPlus1 } => {x a foo:Interesting} } (works for) me.
>
> Or alternatively maybe we *should* allow + and - to appear in names:
>
> {{y > x+1. x+1 sumOf  (x, 1)  } => {x a foo:Interesting} }
>
> Seth Russell
>
Received on Tuesday, 16 July 2002 16:58:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:54 GMT