[Top] [All Lists]

Re: [ontolog-forum] Accommodating legacy software

To: Gian Piero Zarri <zarri@xxxxxxx>
Cc: "[ontolog-forum]" <ontolog-forum@xxxxxxxxxxxxxxxx>, Gian Piero Zarri <gian_piero.zarri@xxxxxxxxxxxxxxxxx>
From: William Frank <williamf.frank@xxxxxxxxx>
Date: Thu, 6 Sep 2012 11:16:52 -0400
Message-id: <CALuUwtBVxg7JsjgwkGz3aTfsRUmOKNCK_LuTUVjUKL9xt=r=Pw@xxxxxxxxxxxxxx>
On Thu, Sep 6, 2012 at 10:30 AM, Gian Piero Zarri <zarri@xxxxxxx> wrote:
On 06/09/2012 15:28, William Frank wrote:

But more generally, an n-ary relation instance is always expressable as n independent role assertions.

For example:

George gave the book to Mary.

There is an instance g of the giving action G. and

In g, George played the role of giver, the book played the role of given, and Mary played the role of reciever.

Yes, and this is simply expressed in NKRL - an n-ary representation language, see http://www.springer.com/computer/ai/book/978-1-84800-077-3 - as a "predicative occurrence" (instance of a standard NKRL template) like:

date-1: 2012-09-06-16:30

BENF = BEN(e)F(iciary) role.

Why always reinvent the wheel?

Indeed, my sentiments exactly.  I am pretty sure the fact I express: every n-nary relationship is expressable as n role instances with respect to a an instance of a relation, has been well-known for more than 50 years. I learned it from Lakoff's dissertation, (1963).   It is also a core part of in many fact-oriented metamodels. 

Predicative occurence is a beatiful name.  I am glad that this is expressable in NKRL.   But the idea is simply *a fact* about languages and _expression_, which can't be the sole property of a given formal language.  The reverse is the case, NKRL is a good language because it uses this important fact.  If a formal language CAN'T do this, I would not want to be forced to use it.  This is why I prefer simple examples in something that looks a bit like English, to talk about meaning, not to in any way disparage the importance of formal languages.


Gian Piero ZARRI

Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/  
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/  
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/ 
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J    (01)

<Prev in Thread] Current Thread [Next in Thread>