Re: N3 and N-Triples (was: RDF in HTML: Approaches)

From: "Michael Kifer" <kifer@cs.sunysb.edu>

> "Seth Russell" <seth@robustai.net> writes:
> >
> > From: "Michael Kifer" <kifer@cs.sunysb.edu>
> >
> > > >>>>> "SR" == "Seth Russell" <of Mon, 03 Jun 2002 10:35:26 PDT>
writes:
> > >     MK> NTriples can be naturally encoded in XML and exchanged.
> > >     SR> Is that actually true?   How?
> > > <triple><subject ...>subj</subject><property>...</property> <object>
...
> > </object> </triple>
> > >
> > > An attribute in subject/object can tell whether the content is a
literal,
> > > uriref, or a blank node.
> > > Cross-referencing blank nodes presents some small problem, which can
be
> > > fixed in a number of ways.
> > > But, as I said before, a better syntax would be to use F-logic, which
is
> > > not that far from the triples, but is cleaner. (Reification and blank
> > nodes
> > > are easier to represent.)
> >
> > Ok, now I know what you meant ... but incidentally that is not N-Triples
> > .... for the record, below are some references to N-Triples, which to my
> > knowledg cannot be 'naturally encoded in XML' unless you do something
like
> > ParseType='literal' but then just about anything could be 'naturally
encoded
> > in XML'.
>
> This is too subtle for me. Do you mean that you actually want to write
> stuff like <http://purl.org/dc/elements/1.1/source> inside the XML
> representation of N-triples?

No, but it's not what I want that counts for anything here.

> I meant that there is a reasonably simple XML language that is isomorphic
> to N-triples. This is all you need. Write in a simple non-XML surface
> language and have machine translate it into XML. XML is not for people --
> it is for machines. All you need is a 1-1 mapping between a
> human-friendly surface language and machine's.

Well I think that a lot of people embed RDF in XML using the
ParseType='literal' technique because it comes out the most compatable ..
but I may be wrong .... I'm not an expert on the w3c standards.  A processor
of a KRep language doen't really need to inherit complexity and limitations
from XML ... in other words: what is gained by trying to use the same parser
for both languages?   So if the carrier languague must be XML, then it is
simple, imho, to have the XML parser pass off  to the KRep parser with
ParseType='literal'.   Otherwise me thinks you will have two languages
fighting with each other.  But I have no particular problem with the way you
emedded triples above ... my only point was that they were not what the w3c
has been calling N-Triples .. it was a bit of a nit ... sorry.

Seth Russell

Received on Monday, 3 June 2002 22:26:23 UTC