Re: datatypes and MT

>>>Are you suggesting that
>>>
>>>   <rdf:Description>
>>>     <eg:prop>foo</eg:prop>
>>>   </rdf:Description>
>>>
>>>is really shorthand for
>>>
>>>   <rdf:Description>
>>>     <eg:prop rdfs:str="foo"/>
>>>   </rdf:Description>
>>>
>>
>> yes
>
>
>Now I understand, I think.  Of course this is not what M&S says, right?

right

>So you are proposing that we redefine what triples are generated for:
>
>    <rdf:Description>
>      <eg:prop>foo</eg:prop>
>    </rdf:Description>
>
>i.e. they should be:
>
>   _:anon <eg:prop>  _:lit .
>   _:lit  <rdfs:str> "foo" .
>
>I guess that's a test case.  Have I got that right?

right

>If you are serious about
>this, can I suggest you post the suggestion, along with an explanation of the
>advantages and disadvantages to the WG.  But please note that its a change to
>M&S and will break every implementation out there, so it'll need a strong case.

Well, as I said yesterday, "I came to realize that..."
so that's pretty recent...
and you are right to ask for a strong case
however I will need some time for that
(and a fresh mind, because now I'm feeling
weak, even a bit sick I think) and I also
have to leave in a couple of hours to
Nice for the AC meeting, but I will think
and work further as much as possible
and come back to you and the WG
The main point is that the object in
  :s eg:shoeSize "10".
is *not* a string object but some-thing that
can be further characterized as having a
rdfs:str property with string value "10"

--
Jos

Received on Sunday, 4 November 2001 06:52:45 UTC