RE: What the #@&*$(!@ is going on here?! (was RE: N-Triples for proposed xsi:type [was Re: xsi:type test case ]

> -----Original Message-----
> From: ext Frank Manola [mailto:fmanola@mitre.org]
> Sent: 07 August, 2002 16:09
> To: Stickler Patrick (NRC/Tampere)
> Cc: w3c-rdfcore-wg@w3.org
> Subject: Re: What the #@&*$(!@ is going on here?! (was RE: 
> N-Triples for
> proposed xsi:type [was Re: xsi:type test case ]
> 
> 
> 
> 
> 
> Patrick.Stickler@nokia.com wrote:
> 
> > 
> 
> snip
> 
>  
> > The WG reached IMO full concensus on a local/explicit datatyping
> > idiom, as well as its associated semantics. Why then are we now
> > considering another local idiom -- especially one which requires
> > changes to the graph syntax and essentially duplicates the 
> > globally unambiguous and tidy nature of URIs?!
> 
> >
> 
> 
>  
> 
> I can only speak for myself, but one of the things going on 
> here from my 
> perspective is that it's not 100% clear to me what we've 
> agreed on now. 
>   You've made the above point about our having reached full 
> concensus in 
> several recent messages.  I don't necessarily disagree with this, but 
> these lengthy back-and-forths about various pieces of this 
> "concensus" 
> are not clarifying things for me.  Is there a *single* description of 
> what we've supposedly agreed on that you could point me to (*not* the 
> statement in the minutes of the F2F).  I'm talking about 
> something like 
> a single description of the whole datatyping idiom you've 
> referred to; 
> e.g., if it was one of the options in the message Brian sent to 
> rdf-comment, you can point to that);  and I'm not talking about a 
> lengthy justification of that idiom either, just a 
> description of what 
> it is.
> 
> --Frank

The local idiom I'm referring to is the datatype property
idiom, which is outlined in detail in the stake-in-the-ground
WD as well as included in Brian's inquiry to rdf-interest.

C.f.

http://lists.w3.org/Archives/Public/www-rdf-interest/2002Jul/0045.html
http://www.coginst.uwf.edu/users/phayes/simpledatatype23-02-2002.html (section 3)

Namely,

   Jenny ex:age _:x .
   _:x xsd:integer "10" .

where the semantics are understood that

   _:x rdf:type xsd:integer .

I.e. Jenny's age is the integer ten, not the string "10".

The XML serialization for the above triples would be

   <rdf:Description rdf:ID="Jenny">
      <ex:age xsd:integer="10"/>
   </rdf:Description>

From all my recollections, which admitedly may be imperfect,
it has been quite some time since any of the XML or graph
syntax, or the semantics of the above local idiom was an
issue of debate.

For quite many weeks, the only real issue of debate remaining
has had to do with the semantics of the global/implicit
idiom. I.e., given 

   Jenny ex:age "10" .

does "10" denote some (possibly underdefined) value, or simply
the literal string "10".

Now, some folks are suddenly suggesting instead that we

1. write the XML as

  <rdf:Description rdf:ID="Jenny">
      <ex:age xsi:type="xsd:integer">10</ex:age>
   </rdf:Description>

2. introduce a new kind of graph node, a typed literal
   and map the above to

   Jenny ex:age xsd:integer(10) .

The WG agreed that we would go with the stake-in-the-ground 
datatyping proposal unless there was clear technical or
implementational reasons not to. The re-opening of the 
issue of tidy versus untidy was motivated by such reasons,
but no such shortcomings have, to my knowledge, been
presented which would justify any deviation from the 
local idiom and semantics as defined in the stake-in-the-
ground specification.

This is why I am wondering why we are even spending time
on it, when it has nothing whatsoever to do with the
single remaining issue, namely the tidyness or untidyness
of literals.

Patrick

Received on Wednesday, 7 August 2002 09:27:07 UTC