W3C home > Mailing lists > Public > www-archive@w3.org > April 2004

Re: parentheses vs. braces in TriQ

From: Chris Bizer <chris@bizer.de>
Date: Fri, 2 Apr 2004 11:44:57 +0200
Message-ID: <00c301c41897$2c1e7b60$1f12fea9@named4gc1asnuj>
To: "Jeremy Carroll" <jjc@hplb.hpl.hp.com>
Cc: "Patrick Stickler" <patrick.stickler@nokia.com>, <www-archive@w3.org>, "Pat Hayes" <phayes@ihmc.us>

OK, braces in TriG.

What do you guys think about braces in TriQL?

SELECT ?a
WHERE ?b (?a <km:Skill> <km:Programming> .
          ?a <rdf:type> <km:Person> )
      (?b <swp:assertedBy> ?c .
       ?c <swp:authority> ?d )
      (?d <km:affiliation> ?e)
      (?e <rdf:type> <km:Project> .
       ?e <km:Topic> <km:Programming> )
AND COUNT(?e) > 2

or

SELECT ?a
WHERE ?b {?a <km:Skill> <km:Programming> .
          ?a <rdf:type> <km:Person> }
      {?b <swp:assertedBy> ?c .
       ?c <swp:authority> ?d }
      {?d <km:affiliation> ?e}
      {?e <rdf:type> <km:Project> .
       ?e <km:Topic> <km:Programming> }
AND COUNT(?e) > 2

I tend to stay with parentheses because it makes typing queries easier, RDQL
also uses parentheses and a query lanuage is related more to the abstract
syntax than to one concrete syntax.  Also weak arguments :-)

Any optinions?

Jeremy, do you have a up-to-date version of our paper, which you could send
me?

Chris


----- Original Message -----
From: "Jeremy Carroll" <jjc@hplb.hpl.hp.com>
To: "Chris Bizer" <chris@bizer.de>
Cc: "Patrick Stickler" <patrick.stickler@nokia.com>; <www-archive@w3.org>;
"Pat Hayes" <phayes@ihmc.us>
Sent: Friday, April 02, 2004 12:18 PM
Subject: Re: parentheses vs. braces in TriQ


> Chris Bizer wrote:
>
> > Hi Patrick,
> >
> >
> >>>>Section 4: use of parentheses
> >>>>Perhaps we should use crather than parentheses
> >>>>to be more compatible with N3/Turtle, which use parentheses
> >>>>for collections.
> >>>
> >>>I am easy - parentheses as opposed to braces was Chris's choice - I'll
> >>>let him argue the case (if he wishes to).
> >>
> >>I'm thinking about what will create the least path of resistance
> >>if folks start adopting this -- and compatability with N3/Turtle
> >>seemed to me to be a big win.
> >>
> >
> >
> > Hmm, yes, I see arguments for both options:
> >
> >
> >
> > pro parentheses:
> >
> > - we started with them and already used them in the SWIG paper
> >
> > - a named graph is not a N3 formula, which is underlined by using
> > parentheses
> >
> > - TriG is based more on N-Triples than N3.
> >
> > - We didn't decide on a list syntax for TriG yet, or whether we want to
> > include lists at all.
> >
> > - using braces would also imply braces in TriQL which would move it
further
> > away from RDQL
> >
> > - parentheses look "nicer" and are easier to reach on German and English
> > keyboards.
> >
> Much easier on italian keyboards which do not have braces.
>
>
> >
> >
> > I think that all these arguments are not very strong. So if you have the
> > strong feeling that we should change to braces it is OK with me.
> >
> >
> >
> > Chris
> >
> however, none of the arguments is strong
>
> Jeremy
>
>
>
Received on Friday, 2 April 2004 05:44:24 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 7 November 2012 14:17:42 GMT