W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > April to June 2004

Human-friendly Syntax (was: Agenda: RDF Data Access WG telcon 25M ay)

From: Seaborne, Andy <andy.seaborne@hp.com>
Date: Tue, 25 May 2004 11:31:16 +0100
Message-ID: <E864E95CB35C1C46B72FEA0626A2E808031A9883@0-mail-br1.hpl.hp.com>
To: "Thompson, Bryan B." <BRYAN.B.THOMPSON@saic.com>, "''RDF Data Access Working Group' '" <public-rdf-dawg@w3.org>

Bryan,

For me, the "Human-friendly Syntax" refers to the query lanaguage.  It makes
no statement about the results format.

We have had discussion about RDF syntaxes for a query lanaguage.  I think it
is important that the WG produce a syntax for the query language that is
oriented towards clear use by application writers.  Maybe some RDF syntax
would be clear, maybe the solution is N3, including formulae, which is
beyond RDF, but the goal is that application writers, maybe power end users,
can write queries in a clear form.  Think SQL - the friendliness is
relative.

>  Would you support
> the requirement "human frendly syntax with human friendly results?"
 
No - I would not support that as it mixes two orthogonal issues, Ql syntax
and results syntax.  We could have a result format that is human-friendly,
we could choose not to.  Personally, I don't see a human friendly syntax as
being needed - see my reply to your message "Building a bridge from RDF to
the web?" - there isn't just one such format and it is in the domainof other
systems (e.g. XSLT, XQuery).

From looking at the requirements we have so far, the result formats I see as
necessary are variable bindings (a streaming format and possibly RDF format)
and an RDF extracted subgraph. Ther may be othgers (e.g. squnce of matching
subgraphs).   There isn't a single format that can meet all the
requirements.

> What message would
> you expect people to take away from this requirement?

That the expected user of the query language is a person, an application
writer.  It is not exclusively machine generated queries.

Note that having an human-friendly syntax does not automatically preclude
other synatxes.  An RDF one would be good but proper handling of variables
while retaining the full RDF semantics is going to be hard.

> So, something like DIG
> which makes it very easy to ask certain kinds of questions?  

You might be interested in this tech report by Ian Dickinson on implementing
to the DIG 1.1 specification:
http://www.hpl.hp.com/techreports/2004/HPL-2004-85.html

	Andy

-------- Original Message --------
> From: Thompson, Bryan B. <>
> Date: 24 May 2004 21:56
> 
> Andy,
> 
> I would like to understand what this would entail.  Would you support
> the requirement "human frendly syntax with human friendly results?"
> If not, why have human friendly syntax without friendly results?  Or
> do we mean "'expert RDF user' friendly syntax?"  So, something like DIG
> which makes it very easy to ask certain kinds of questions?  Or do you
> consider something like RDQL to be "human friendly?"  What message would
> you expect people to take away from this requirement?
> 
> -bryan
> 
> -----Original Message-----
> From: public-rdf-dawg-request@w3.org
> To: 'Dan Connolly'; 'RDF Data Access Working Group'
> Sent: 5/24/2004 3:18 PM
> Subject: RE: Agenda: RDF Data Access WG telcon 25May
> 
> 
> > 4. Requirements
> > 
> > Have any proposals achieved consensus recently?
> > 
> > Refine requirements by evaluating designs?
> 
> If there is one item we could agree before publishing, I would suggest:
> 
> "4.1 Human-friendly Syntax"
> 
> As it makes it clear to the audience we intend to produce this as a
> concrete
> output.  At the F2F it had strong support, possibly because of this, it
> has
> had much discussion since.  It would not exclude other syntaxes.
> 
> 	Andy
Received on Tuesday, 25 May 2004 06:32:04 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:19 GMT