W3C home > Mailing lists > Public > semantic-web@w3.org > July 2010

Re: Show me the money - (was Subjects as Literals)

From: Pat Hayes <phayes@ihmc.us>
Date: Thu, 1 Jul 2010 23:05:52 -0500
Cc: Henry Story <henry.story@gmail.com>, Jeremy Carroll <jeremy@topquadrant.com>, Semantic Web <semantic-web@w3.org>
Message-Id: <D5B86564-DF6B-4FED-969D-3D5A09693D79@ihmc.us>
To: Reto Bachmann-Gmuer <reto.bachmann@trialox.org>

On Jul 1, 2010, at 11:16 AM, Reto Bachmann-Gmuer wrote:

> On Thu, Jul 1, 2010 at 5:46 PM, Henry Story <henry.story@gmail.com>  
> wrote:
>>> I have loads and loads of code, both open source and commercial  
>>> that assumes throughout that a node in a subject position is not a  
>>> literal, and a node in a predicate position is a URI node.
>>
>> but is that really correct? Because bnodes can be names for  
>> literals, and so you really do have
>> literals in subject positions.... No?
> BNodes and UriRefs can be used in place of and be the same resource as
> a literal. The abstract syntax forces literal in object position, but
> with semantic extensions (owl:sameAs) you can express the same as you
> could having Literals in subject position.
>
>>
>>
>>> Of course, the "correct" thing to do is to allow all three node  
>>> types in all three positions.
>
> While allowing Literal as subjects would cause cost of adapting
> existing code, allowing bnode as predicates i think would make many
> algorithms computationally more expensive.

That is true, it might indeed.

> I don't see how a literal
> could be a property (we could syntactical allow it, but wouldn't every
> use of this feature be a contradiction?)

Well, actually, no. I agree this *seems* unintuitive, but in another  
context we found a solid use for it in Common Logic applications. In  
the IKRIS project, funded by IARPA, we had to make a collection of  
disparate logic-based systems interoperate. This required, among other  
things, a syntactic device to keep track of how names changed their  
meanings between different contexts of use. The elegant (and  
effective) way to do this was to treat a character string as being a  
function from contexts to referents. For details, see the discussion  
of 'captured names' (slide 17++)  in  http://www.slideshare.net/PatHayes/ikl-survey 
  .  For a more general polemic in this whole topic of what 'makes  
sense', see http://www.slideshare.net/PatHayes/translating-into-common-logic-459009 
, slides 13 to the end.

Pat



>
> Cheers,
> reto
>
>

------------------------------------------------------------
IHMC                                     (850)434 8903 or (650)494 3973
40 South Alcaniz St.           (850)202 4416   office
Pensacola                            (850)202 4440   fax
FL 32502                              (850)291 0667   mobile
phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes
Received on Friday, 2 July 2010 04:06:56 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 07:42:20 UTC