Re: RDF semantics, access control description and timeranges

At 11:11 AM 12/15/02 -0500, tim finin wrote:
> >   ex:Host ex:access                         (1)
> >      [ a ex:AccessPermission ;
> >        ex:allowProtocol ex:HTTP ;
> >        ex:allowProtocol ex:FTP ;
> >        ex:allowProtocol ex:SSH ] .

[...]

>I'd say the *and* reading is right for (1).  The host has an
>accessPermission _G0042.  _G0042 has an allowProtocol http *and*
>_G0042 has an allowProtocol ftp *and* _G0042 has an allowProtocol ssh.

On further reflection, you've put your finger on what I was getting 
confused about.

As you say the *and* reading works fine when talking about permissions.

The step I'd taken, and was getting confused by, was transferring this 
reading to the conditions for granting access, which really is a query 
against the available permissions.  And being a query, it's satisfied by a 
disjunction of (any one of) the permissions that might be conjoined in the 
permission database.

Hmph!  I thought I knew this stuff, but still didn't avoid the pitfall :-(

#g
--
(Who is now reworking his RDF access control rules experiment)


-------------------
Graham Klyne
<GK@NineByNine.org>

Received on Monday, 16 December 2002 12:52:58 UTC