ontolog-forum
[Top] [All Lists]

Re: [ontolog-forum] Pragmatic slant to context (was Current Semantic Web

To: "[ontolog-forum]" <ontolog-forum@xxxxxxxxxxxxxxxx>
From: Duane Nickull <dnickull@xxxxxxxxx>
Date: Wed, 05 Sep 2007 12:12:19 -0700
Message-id: <C3044DA3.659A%dnickull@xxxxxxxxx>
Gary:

Yes – this example is different.  IN UN/CEFACT we identified 8 different contextual categories of qualifiers.  There are categories for Role, Process, Systems Capabilities, Geo-political, legislative and others.  I totally concur with John’s example and my example was orthogonal.

Another example that is more closely aligned with John’s statements occurred in work Adobe did for the Government of Canada Integrated Justice Initiative (IJI).  They defined a 353 element data dictionary however the representation terms were not bound to any data element (in true ISO/IEC 11179 2004 part 3 style) until the point at which they were subject to presentation and subsequent interpretation by systems and/or humans.  The work went very well but the end report asked more questions than it answered. It is a very complex problem to address.

Anyone interested in getting some information on this work can ping me privately and I can forward some material.

Duane


On 9/5/07 11:09 AM, "Gary Berg-Cross" <gary.berg-cross@xxxxxxxx> wrote:

Duane,

I had a thought looking at your contextual example of  data element named "FirstNameOfParty".  You note that an instance
is slightly different if it occurs within a hierarchy of //PurchaseOrder/BuyerParty vs.
 //PurchaseOrder/SellerParty even though the semantics of the lone data element do not change.

It seemed to me that one might argue that it differs for pragmatic reasons as John Sowa (JFS)  discussed in this exchange:
 
JFS>> The pragmatics, of course, is how the sheet is supposed to be
 >> interpreted by a performer.

JAS> Not quite. The pragmatics is how the sheet music is in fact
 > performed.  It is almost never clear from spots on a page how
 > the music is to sound.

John continued ---
As I said in previous messages, the terms 'syntax', 'semantics',
and 'pragmatics' are watered-down versions of Peirce's semiotics.

CSP actually made very precise distinctions about the 'interpretant'
of a sign as intended by the speaker (or person who wrote, drew,
built, arranged, sculpted, etc.) and 'interpretant' by the listener
(or the one who views, feels, smells, tastes, etc.) the sign.

CSP was very clear about the fact that the speaker and listener
can interpret the same sign in very different ways -- and
sometimes the interpreter may have a much more elaborate
interpretation than the one who generated the sign.
Gary Berg-Cross

From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx on behalf of Duane Nickull
Sent: Wed 9/5/2007 12:12 PM
To: Pat Hayes
Cc: [ontolog-forum]
Subject: Re: [ontolog-forum] Current Semantic Web Layer pizza (was ckae)

Pat:

Pet Hayes Schreibt:
>>Semantics is always contextual.

>Nonsense. Fortunately; because if it were true, semantics would be impossible.
>This assertion is like the idea that all
>assertions are contexual. It sounds plausible,
>but that 'all' makes it (literally) incoherent.

In general, I suspect I should have not used the "always" word but I still
believe that the context in which an instance of something occurs has a
great deal of impact on the semantics (what the thing means), albeit perhaps
not in 100% of instances.  My grammar is generally bad.

My memories are actually relevant in the subject of semantics given it is I
who was the recipient of the signal (diver down flag).  I see and interpret/
infer something from the flag (BTW - I did study semaphore and do know the
signal from a mariner perspective as well).


>> Semantics cannot exist without including the aspect of context.
>
> That is simply obviously false, since it does so
> exist and has done for maybe 70 years now.

Perhaps I worded this wrong.  Context is perhaps part of inference.  When
one (human, computer, dog, other) encounters a symbol/syntax instance, there
is an inference process.  We all draw meaning of some form at some level of
awareness (raising in tuple store based on impact perhaps).  That is perhaps
not part of declared semantics but I would still think if is relevant.


> This word "context" has been the cause of more
> wasted time in KR than almost any other. It is a
> deep tar-pit, not the least reason being that it
> is itself thoroughly contextual, ie it can be
> used to mean almost anything. There are many,
> many problems with the idea that ontology
> meanings are contextual.
Disagree that it is a complete waste of time given there are important
lessons learned.  Several papers and efforts have declared that it is an
open ended rathole and computationally too expansive to contemplate.  Even
the UN/CEFACT CCTS project which had a mere 8 context modifiers in its
framework could potentially have a very large number of semantic values.  It
is not even known at what level of granularity to determine a difference.
The CCCTS work could have 8 modifiers, each with values of up to 3000
enumerations or an extremely large number.

In general, I have no idea how to solve this issue but have worked enough on
XML artifacts to know that semantic declarations without context do not work
for all purposes.

An example - imagine a data element named "FirstNameOfParty".  An instance
is slightly different if it occurs within a hierarchy of
//PurchaseOrder/BuyerParty vs. //PurchaseOrder/SellerParty even though the
semantics of the lone data element do not change.

The semantics of each data element were declared in an attempt to allow
cognitive applications to automatically suggest mappings from one format to
another (in this case EDI to CBL).  The UBL effort immediately recognized
that context is utterly essential to the process as the context in which
something occurs affects many aspects of its semantics.


>There are extremely good
> reasons why any encoding of meaning in a
> formalism *intended to be used and archived on a
> network* should NOT be contextual (see
> http://www.ihmc.us/users/phayes/ikl/guide/guide.html#LogicForInt
> ). Moreover, we have found in the IKRIS work that
> any formalism which incorporates contexts can be
> mapped without loss of meaning into the
> context-free IKL formalism (by treating contexts
> as entities in the ontology itself, if necessary.
> But even that is usually more trouble than it is
> worth.)
Thanks for this - I'll take a look as I have an interest in this.
>
> KR work has been grappling with contextuality and
> contextual representations for over a decade now.
> The results are in, IMO, and they are negative.

I agree that the results are less than optimistic.  CCTS, UBL and several
researchers have all noted similar issues with the context.

Duane





--
**********************************************************************
"Speaking only for myself"
Blog - http://technoracle.blogspot.com
Community Music - http://www.mix2r.com
My Band - http://www.myspace.com/22ndcentury
MAX 2007 - http://technoracle.blogspot.com/2007/07/adobe-max-2007.html
**********************************************************************

_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/  
Subscribe/Config: 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 Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx    (01)

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