Re: Web Semantics for Datasets

On 07/10/11 13:53, Ivan Herman wrote:
>
> On Oct 7, 2011, at 14:48 , Andy Seaborne wrote:
>>
>> On one points:
>>
>> I don't see why
>>
>> <http://example.org>   {<s>   <p>   <o>  . }
>>
>> should mean it is ONLY that triple rather than CONTAINS that
>> triple.  If the data publisher wants to say "and that's all" then
>> they should say so as an additional fact.  The converse of "it's
>> closed by default" is harder to see how to allow it to be open
>> sometimes.
>>
>
> Yes, I had a similar reaction reading Sandro's mail. But that is a
> only a syntax issue for a TriG like stuff, so...

I read this as describing the four column, so abstract syntax, not 
concrete syntax.

Example 1 says "the serialized graph will consist of the single 
triple,<s>  <p>  <o>."  I thought the point of Example 2 (aside from the 
statefulness of actually getting the graph) was that

<http://example.org>  {<s>  <p>  1. }
<HTTP://example.org>  {<s>  <p>  2. }

was impossible because the graphs are closed.

Sandro - did I misread it?

	Andy

Received on Friday, 7 October 2011 13:10:42 UTC