W3C home > Mailing lists > Public > public-owl-wg@w3.org > March 2008

Re: General discussion for TC Wednesday

From: Peter F. Patel-Schneider <pfps@research.bell-labs.com>
Date: Tue, 11 Mar 2008 10:13:46 -0400 (EDT)
Message-Id: <20080311.101346.34297941.pfps@research.bell-labs.com>
To: hendler@cs.rpi.edu
Cc: alanruttenberg@gmail.com, public-owl-wg@w3.org

From: Jim Hendler <hendler@cs.rpi.edu>
Subject: Re: General discussion for TC Wednesday
Date: Tue, 11 Mar 2008 03:56:44 -0400

> Alan-
>   As I cannot attend this week's meeting (I'm on travel, out of the US
> and unable to phone in), I'd like to give my opinion re:  the question
> re: Manchester syntax (and other serializations) in the primer .  I
> think we need to be careful on our choice of which syntactic
> realizations to include.  There are at least another 4-5 RDF
> serializations floating around out there that I know.  

> Some, like N3
> and Ntriples, have some W3C legitimacy, some, are used in popular
> tools (like the SWOOP notation), some are used in some other research
> papers (I noticed 2-3 in various presentations at ISWC last year).

I wonder what the defining document for N3 is?

The W3C team submission on N3
	http://www.w3.org/TeamSubmission/n3/
is probably the most "official", but it doesn't seem to have a
translation from N3 to RDF graphs. 

Turtle, with "official" document the W3 team submssion on Turtle 
	document http://www.w3.org/TeamSubmission/turtle/ 
is supposed to be another serialization but, again, I don't see a
translation from Turtle to RDF graphs.

Ntriples, at http://www.w3.org/TR/rdf-testcases/#ntriples, has the
closest to a mapping to RDF graphs.  The missing parts of the mapping
from Ntriples to RDF graphs are probably sufficiently obvious so as not
to require any further explication.

> Manchester has a somehat more mature realization than many of these,
> but it is still defined in a document with only a draft syntax and no
> publication status (i.e. copyright etc) [1].  In a previous email
> thread it was pointed out that some of the OWL tools handle it, but
> then most also handle N3 (which is more widely used) and SWOOP, for
> example, has it's own which (and SWOOP is still highly used, despite
> not being supported at the moment). 

I note that there currently is no document that defines the mapping from
the Manchester syntax to OWL 1.1.   This is expected to change in a
matter of days.  :-)

>   My proposal would be that we need some specific criteria for what is
> and is not used in the document. Once we have agreed to principles, we
> can agree to which serializations to include -- I believe it is
> important that the document include some discussion of that criterion as
> well, so that we cannot be accused of arbitrarily choosing without
> cause.
>    -Jim H.
> 
> [1] http://www.w3.org/2007/OWL/wiki/ManchesterSyntax

peter
Received on Tuesday, 11 March 2008 14:16:48 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 11 March 2008 14:16:50 GMT