W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > November 2002

Correction Re: n-triples for datatype values [was: Agenda for RDFCore WG Telecon 2002-10-18]

From: Tim Berners-Lee <timbl@w3.org>
Date: Fri, 1 Nov 2002 14:42:03 +0100
Cc: w3c-rdfcore-wg@w3.org
To: Tim Berners-Lee <timbl@w3.org>
Message-Id: <B43E97BA-ED9F-11D6-ABEB-000393914268@w3.org>

I'd just point out that if one wanted to make the n-triples design look 
compatible with n3's path syntax from the point of view of user 
understanding, that in the path syntax, traversal (the derivation of 
new bnodes by specifying a forward or reverse property) goes from left 
to right.

"10"^^xsd:integer

would be the way around, where ^^ could be informally read as something 
like "interpreted as".

I think as a syntax its more common by analogy with (eg) units to put 
the units at the end.

While ^^ is not a real traversal operator (operating at the syntax 
level in the parser and never being visible in the triples produced),  
it doens't  *have* to be compatible with path syntx, and of course 
n-triples nd n3 don't *have* to be subset and superset.  It's just a 
question of trying to minimize the messing with the user's brain, 
reducing learning etc.

I think the examples I had in my last message were the wrong way around.

Tim


PS: Language design is a question of balancing different metaphors to 
make it more natural. I have listed some of the choices in the path 
syntax in http://www.w3.org/DesignIssues/N3Alternatives
Received on Friday, 1 November 2002 08:41:44 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:53:55 EDT