Re: BNF expression of RDF Concepts (ISSUE-176)

On Dec 5, 2013 1:35 PM, "Dan Brickley" <danbri@google.com> wrote:
>
> On 5 December 2013 11:48, Markus Lanthaler <markus.lanthaler@gmx.net>
wrote:
> > On Thursday, December 05, 2013 12:25 PM, Richard Light wrote:
> >> > On 05/12/2013 10:28, Eric Prud'hommeaux wrote:
> > [...]
> >> This would have to be relatively abstract, in that the Concepts
> >> recommendation isn't specifying a serialization format, like Turtle.
> >
> > Exactly. I fear that including BNF into Concepts would thus be very
> > confusing. We tried hard to separate the abstract syntax from concrete
> > syntaxes in RDF 1.1.

There is a large community that counts on type definitions like this.
Almost every math-y paper on RDF includes a (largely redundant) abstract
type definition for RDF. Perhaps using set notation would better serve this
community and avoid any implications of a concrete syntax. (I note also
that the only syntax implications in the strawman abstract syntax come from
ordering choices like (s p o) and (lexical datatype langtag), which reflect
the prose definitions.

> >> > Given the timing, it may not be possible to include this at all, or
to
> >> > include this in a normative section. Will you accept either of those
> >> > outcomes?
> >>
> >> Yes.  I realise that I have come to this discussion at a point where
> >> you are about to finalise this document. Also, I am interested to
> >> hear whether there is wider support for this idea from within the
> >> developer community, but do not take it for granted that such support
> >> exists.
> >
> > I personally am against this for the reason stated above. BNF is, IMO,
of
> > very limited use if it isn't describing a data format but a data
*model*.
> > Could you please elaborate a bit on why you think
> >> it would introduce standard naming conventions, and structures,
> >> which could be followed in whichever programming language was being
> >> used for development.
> >
> > If RDF triples are stored in a relational database for example, do you
think
> > developers would benefit from the BNF? What if it is stored as JSON-LD
in a
> > database such as MongoDB or perhaps ElasticSearch?

I would expect developers in both of those communities to benefit from a
terse, comprehensive enumeration of the elements of RDF. It makes it a lot
easier to ask "what does my SQL table have to capture?"

> > In any case, I've raised ISSUE-176 [1] to track this. We will get back
to
> > you with an official shortly.
>
> For a concrete syntax,
> http://www.w3.org/TR/rdf-syntax-grammar/#section-Infoset-Grammar makes
> a lot of sense. For the abstract data model, less so...
>
> Dan
>
> >
> > Cheers,
> > Markus
> >
> >
> > [1] https://www.w3.org/2011/rdf-wg/track/issues/176
> >
> >
> > --
> > Markus Lanthaler
> > @markuslanthaler
> >
> >