Re: Problems with MUO units ontology

Dear Tim,

thank you for your comments. During the last months, Diego and I have
been unable to spend time on MUO, although we have internally discussing
the comments you made at the Mandelieu Plenary Meeting.

We understand your points and we acknowledge the issues. During the
following days, we will try to come up with a new proposal that
integrates our previous work and motivation, and the Interpretation
Properties Pattern.

We will contact you as soon as we find a proposal that combines the
benefits from both points of view.

Best,

p.

-- 
Luis Polo Paredes
R&D Department  
Fundación CTIC
E-mail: luis.polo@fundacionctic.org
Tfno:+34 984 29 12 12
Parque Científico Tecnológico Gijón-Asturias-Spain
www.fundacionctic.org


El sáb, 04-07-2009 a las 11:03 -0400, Tim Berners-Lee escribió:
> Diego and Luis,
> 
> I am concerned that the MUO ontology you are proposing people adopt  
> has serious issues.  I was reading
> 
> http://forge.morfeo-project.org/wiki_en/index.php/Units_of_measurement_ontology#Alternatives_to_represent_measurable_quantities_in_RDF
> 
> 
> You refer to my Design Issues note on Interpretation properties but  
> you don't seem to have taken it into account.
> 
> You say, (I add the <> brackets you forgot from the N3 in the article)
> 
> 
> :Spain :area [
>             muo:numericalValue "504782"^^xsd:double ;
>             muo:measuredIn :Sq_Km
>             ] ;
>         :area [
>             muo:numericalValue "194897"^^xsd:double ;
>             muo:measuredIn :Sq_Miles
>             ] .
> 
> But I would expect Spain to have, in principle, just one area. I would  
> expect :area to be a functional property.   If Spain has one area,  
> then we get the confusion:
> 
> 
> :Spain :area [
>             muo:numericalValue "504782"^^xsd:double ;
>             muo:measuredIn :Sq_Km;
>             muo:numericalValue "194897"^^xsd:double ;
>             muo:measuredIn <Sq_Miles>
>             ] .
> 
> which is completely muddled an ambiguous of course.
> 
> I suppose you can avoid this using the extra arc as in
> 
> :Spain :area [
>      muo:expressableAs [
>             muo:numericalValue "504782"^^xsd:double ;
>             muo:measuredIn :Sq_Km
>             ] ;
>     muo:expressableAs [
>             muo:numericalValue "194897"^^xsd:double ;
>             muo:measuredIn #Sq_Miles
>             ] .
> 
> but that is cumbersome.
> 
> By contrast, the Interpretation Properties pattern, which you do not  
> review in your article at all, makes it much simpler.
> 
> 
> :Spain   :area  [ :Sq_Km  "504782"^^xsd:double].
> 
> 
> By the way, in N3 you can write the above as
> 
> :Spain   :area  [ :Sq_Km  504782e0].
> 
> or even
> 
> :Spain   :area  504782e0^:Sq_Km.
> 
> This does not have the problem above, as there is no issue with
> 
> :Spain   :area  504782e0^:Sq_Km ,  194897e0^:Sq_Miles .
> 
> that is
> 
>      :Spain     :area  [ :Sq_Km 504782.0e+00 ];
>                     :area [ :Sq_Miles 194897.0e+00 ] .
> 
> as the same area can be expressed using two different units.
> Could you convert the MUO work to use interpretation properties?
> 
> Tim BL
> 
> 
>

Received on Monday, 6 July 2009 10:54:15 UTC