Re: New draft of section 5.5

Can you check what I wrote at /latest/ to see if it matches what Ed is
saying? I think it does.

The answer doesn't have to be a partition of the statements, although
that would be one way to do it. My approach is closer to Ed's I think
- it's a classification of subject and object positions of properties,
i.e. a partition of properties into four categories. It doesn't even
strictly require an IR/NIR type distinction. The domain analysis and
model theory is a detail, and an important one if the idea is to be
pursued (e.g. I would think you'd want to prove it sound with respect
to RDFS). But the idea can (and should) be presented without it, just
by reference to requirements, which is what I think I've done.

Jonathan

On Tue, Apr 5, 2011 at 1:30 PM, Nathan <nathan@webr3.org> wrote:
> David Booth wrote:
>>
>> On Tue, 2011-04-05 at 17:06 +0100, Nathan wrote:
>>>
>>> David Booth wrote:
>>>>
>>>> This may get confusing having parallel versions of section 5.5 going
>>>> back and forth, but maybe it will help us converge.
>>>>
>>>> Anyway, here are comments on your latest version of sec 5.5
>>>> http://www.w3.org/2001/tag/awwsw/issue57/latest/#chimera
>>>
>>> Things are getting confused here, the use case doesn't capture Ed's view,
>>> and it's precisely the inverse of what David is discussing.
>>>
>>> Chimera is when the same graph uses a single name to refer to two
>>> different things, note Ed's terminology "and if I *also*"
>>>
>>> He's saying that all these statements would be in one graph:
>>>
>>> <http://en.wikipedia.org/wiki/William_Shakespeare> a foaf:Person ;
>>>   foaf:name "William Shakespeare" ;
>>>   dcterms:modified "2010-06-28T17:02:41-04:00"^^xsd:dateTime ;
>>>   cc:license <http://creativecommons.org/licenses/by-sa/3.0/> .
>>>
>>> Not in two graphs, or made by two different people.
>>>
>>> Thus David, sorry to say, but what you propose in your own section 5.5
>>> doesn't cover the case Ed is talking about (well it does, btu it doesn't say
>>> what you want, because the conclusions you come to would need to be applied
>>> to the above graph to either create two graphs or remove half the
>>> statements, *prior* to publishing, which == not asserting the above graph ;)
>>
>> Right, I see.  So it sounds like Ed is talking about the case where the
>> assertions are *already* co-mingled, and he wants to partition the graph
>> into two graphs, such that
>> <http://en.wikipedia.org/wiki/William_Shakespeare> refers to a
>> foaf:Person in one and an IR in the other.
>
> Almost, he doesn't want to partition the graph.
>
> He wants you to be able to partition the graph in to two if you care enough
> to do it. (by using the predicate infers the universe of x for that
> statement approach).
>

Received on Tuesday, 5 April 2011 18:05:32 UTC