Computable Contracts for Insurance: Establishing an Insurance-Specific Controlled Natural Language

[John F Sowa]

Resending a note from Edward Zalta

 

---

From: "Edward Zalta" <edward...@gmail.com>

Hi John,

   I sent this reply to Alex Shkotin, and cc’ed everyone. But since I’m not a member of ontolog-forum, it wasn’t distributed to you. In case you are interested, here is my reply.

   All the best,

Yours,

Ed

Begin forwarded message:

From: Edward Zalta <edward...@gmail.com>

Subject: Re: [ontolog-forum] Computable Contracts for Insurance: Establishing an Insurance-Specific Controlled Natural Language - InsurLE

Date: July 22, 2025 at 3:34:25 PM PDT

To: Alex Shkotin <alex.s...@gmail.com>

Cc: ontolo...@googlegroups.com, "Norbert E. Fuchs" <fu...@ifi.uzh.ch>, r...@doc.ic.ac.uk, Edward Zalta <za...@mally.stanford.edu>

Dear Alex,
  Thanks for looping me into your discussion and for referencing my unpublished monograph, in which I’ve been compiling theorems proved in object theory over the past 40 years. I looked up your discussion on groups.google.com and I see that your last message seems to be a reaction to the remark by John Sowa, who wrote:

I strongly support projects to develop special-purpose ontologies for various industries. And I'm sure that the one you describe is good for its purpose.  But I had seen so many projects, that I am extremely skeptical about any attempt to develop a universal ontology of everything.

Given what he says, I doubt John will find that the system developed in Principia Logico-Metaphysica is a universal ontology of everything.  Indeed, that wasn’t what it was designed to be, though it was designed to start with the highest level ontological categories of objects and relations.

If you do look at that monograph, I would characterize it as an attempt to systematize and analyze the abstract objects, such as mathematical objects,  that are presupposed in the natural sciences and in the "a priori" sciences (logic, linguistics, etc.).  I’m not a Pythagorean and so I don’t think that mathematical objects are fundamental elements of the universe. So to account for the meaningfulness and content of mathematical language, we need a philosophical analysis of that language. That is one application of object theory.  It might come as a surprise just how much mathematics can be derived from a system which doesn’t assume any mathematics but only logical principles and a few high-level axioms about the two fundamental categories of objects and relations (I use second-order quantified modal logic with just one extra mode of predication for the base system, but then re-express the theory using relational type theory).  

In general, goal is to use object theory to *define* abstract objects that others postulate, and *derive* the governing principles that are asserted of them.  The List of Important Theorems and the Table of Contents will tell you what I mean by this.

But I haven’t attempted to work on any applied ontologies of the kind being discussed in this thread on ontolog-forum.  To that extent, I don’t have much to offer.

In any case, thanks for pointing out my work.
All the best,
Yours,
Ed

On Jul 22, 2025, at 1:41 AM, Alex Shkotin <alex.s...@gmail.com> wrote:

John, I am sending a copy to Prof. Edward Zalta just for a chance to get his point

Any philosophical doctrine is a general-purpose ontology.  We may choose any to formalize.
We need just one or another enthusiast for this exciting work.
A great example is Edward Zalta's formalization of metaphysics Principia Logico-Metaphysica (Draft/Excerpt).
Interconnection of domain specific theories and philosophical doctrine for me lies in Phenomenology of Matter (PhoM). Which maybe is a part of phanerology.
For us, the main theme of PhoM is the geometry of luminous surfaces of different colors in different places. Just look around you to start applying this general-purpose theory.

Alex

пн, 21 июл. 2025 г. в 19:53, John F Sowa <so...@bestweb.net>:
Alex,

I've known Robert Kowalski for many years, and he is an expert in developing formal theories for various purposes.

But the point I keep stating is that  independently developed special-purpose theories make special-case assumptions (axioms) that are almost always incompatible (inconsistent) with the special cases for other purposes.

For these reasons, a collection of multiple special-case theories will NOT give you a general-purpose ontology.

John

From: "Alex Shkotin" <alex.s...@gmail.com>

John,

I am sending a copy to Prof. Robert Kowalski for possible feedback.
Any formal ontology keeps formal theory and a model of this theory. In terminology of Description Logics (OWL2) formal theory is T-box, and model is A-box.
If somebody is using InsurLE or Common Logic the situation will be the same, but with different terminology.
What is funny with the OWL2\DL approach is that they keep all terms (for classes, and properties) as primary. And all T-box formulas as axioms. They actually call them "axioms".
Step forward to the formal theory is to give definition for most terms.
And you are right: applicability of a theory may be very narrow, for example just an organization, or even family. Where we have our own special terms and definitions for objects and processes.
So, definitions in the Insurance area is the first step to theory.  

Alex

вс, 20 июл. 2025 г. в 21:17, John F Sowa <so...@bestweb.net>:
Alex,

That's an important point:  "Creating a formal theory for Insurance area of practice may be next."  

Various industries and various domains of interest have specialized interests and requirements.  But the specializations for one domain are likely to ignore or oversimplify the requirements for other domains.

Warning #1:  Special cases are likely (almost certainly) inconsistent with one another and with any complete, general
system.

Warning #2: ISO requires every standard to be updated or reaffirmed every 5 years in order to accommodate new developments.

Warning #3:  Nobody knows how many new warnings will be created by ongoing R & D in AI.

By the way, I had been working with ISO standards organizations when I was employed by IBM, and I continued that work for a number of years afterwards.   Those three warnings are my short summary of some very difficult and challenging issues.

When I say that Ontolog Forum does not have the resources to develop any kind of standards, I am speaking from experience.  ISO is a huge international organization compared to Ontolog Forum.  Anybody who wants to work on standards should start by working with the ISO subgroup of their country

John


--
All contributions to this forum are covered by an open-source license.
For information about the wiki, the license, and how to subscribe or
unsubscribe to the forum, see http://ontologforum.org/info
---
You received this message because you are subscribed to the Google Groups "ontolog-forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ontolog-foru...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/b8f49ebc3ff143388b25156af169921e%40551ba8aeac044aeab5bc11a74f84ea83.

[Alex Shkotin]

John,

Thank you for pointing out the fantastic work of EZ once more.

It's a pity he is not on ontolog-forum.

EZ:"I’m not a Pythagorean and so I don’t think that mathematical objects are fundamental elements of the universe."

IMHO these objects and many others exist in our minds. So the universe consists of matter and a finite number of minds interacting with each other through matter. 

Theory of the objects in the mind should be supplemented with theory of material objects and processes. Beginning from phenomenology of matter.

Part of this theory is everyday geometry.  The base of our civilization.

Observation. Why is the world knowable? Largely because the subsystem behaves within the system in a simpler way than it is structured.

Alex

ср, 29 окт. 2025 г. в 22:32, John F Sowa <so...@bestweb.net>:

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/c0f283c649bf45589132e21531e02fad%4061b1b34adc6f4f3b92a3f769a8c2cab7.

[John F Sowa]

Alex,

I very strongly disagree with that point:  "Observation. Why is the world knowable? Largely because the subsystem behaves within the system in a simpler way than it is structured."

Our sensory perception is far more fundamental than any language (natural or artificial).   And the continuous infinity of perceptions must be mapped to a different continuous infinity of possible viewpoints or purposes.  The viewpoints include every imaginable purpose or intention or application.  The sizes include continuous magnification or telescoping from the tiniest subatomic particles to the largest and farthest aspects of the universe.

There is no such thing as one size or one purpose fits all possible applications.  And then we have to consider the mappings to the languages (multiple natural and artificial) that are processed near the frontal lobes.   That involves mappings and transformations from the inputs in the brain stem to the computations in the cerebellum to the perceptions in the back of the brain, the processing in the middle, and the interpretations in the frontal lobes.

For just a very short overview of the immense complexity, I have repeatedly mentioned the YouTube summary "Without Ontology LLMs Are Clueless."   To find it, just search for "John Sowa, You Tube" and click on the one with that title.  Those slides include many references for much, much more information.

I have also pointed to many other references for other purposes that supplement or extend or support those.  For some, see https://jfsowa.com/ikl .   But there are many more before and since those.

John

 

---

From: "Alex Shkotin" <alex.s...@gmail.com>

John,

[Simon Polovina]

Hi everyone.

You may find The Limits of AI: Generative AI, NLP, AGI, & What’s Next? On YouTube, of interest. It gives a balanced view of the subject, in my view 🤔

I also posted a short remark on my LinkedIn page: hashtag#GenA-hashtag#GOFAI.

NB: the YouTube video is from IBM Technology, an informative site covering multiple topics, including AI.

Thoughts welcome.

Thanks,

Simon

 

Dr Simon Polovina

Simile Logics Limited

Profile │ Email

 

 

[Alex Shkotin]

John,

on your `I very strongly disagree with that point:  "Observation. Why is the world knowable? Largely because the subsystem behaves within the system in a simpler way than it is structured."`

This rule is based on the fact that a subsystem interacts with other system components only along its boundary.

A cell in an organism may be very complex internally, but it exchanges molecules with other cells.

However, recent research seems to indicate that cells also exchange information. This has not yet been studied.

Alex

пт, 31 окт. 2025 г. в 02:34, John F Sowa <so...@bestweb.net>:

--
All contributions to this forum are covered by an open-source license.
For information about the wiki, the license, and how to subscribe or
unsubscribe to the forum, see http://ontologforum.org/info
---
You received this message because you are subscribed to the Google Groups "ontolog-forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ontolog-foru...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/4ea04b28812343b69fd9d67e43e0e032%4002ba44f5f31b47c882144104ac96a02c.

[Simon Polovina]

Hi Jack.

Thanks for your observations, which resonate with me. You and I (and many others) have been in this AI game for a long time, with many false dawns, and we've learnt from it. (As a side note, I wish Prolog was much more in the mainstream than it ended up being; wrong place, wrong time, but that’s another story.)

My view is that Gen AI is a game-changer. I’ve done a lot of work with SAP, and, ‘used right’, it’ll make a significant impact, e.g. Business AI. Indeed, I also first heard the word ‘ontology’ mentioned at SAP’s leading business events recently. Maybe I may even hear Conceptual Graphs being mentioned, too, as one of the few that still use them (refer to my latest paper, for example). Apologies, I’m straying off-topic for this thread.

Going back to that video and taking on board your comments is why I liked that  video.

Cheers,

Simon

 

Dr Simon Polovina

Simile Logics Limited

Profile │ Email

 

From: Jack Park <jack...@topicquests.org>
Sent: 31 October 2025 17:50
To: si...@similelogics.ltd
Cc: ontolo...@googlegroups.com; CG <c...@lists.iccs-conference.org>
Subject: [CG] Re: The Limits of AI: Generative AI, NLP, AGI, & What’s Next?

 

Simon,

 

Thanks for sharing this.  In fact, I have mixed thoughts about it.  In brief:

 

The production values are stunning - likely done with an AI overlay  - unless he's skilled at writing backwards 🤠

 

He occasionally referred to the machine "understanding" long before he introduced that term and the issues surrounding it. Machine understanding likely falls into the deep philosophical camp and may never be resolved; that some computer says the surprising X in response to a prompt: does that really means it "understood" the prompt?  Here's what I wrestle with: a huge amount of symbolic processing in humans is pattern recognition, which has been at the roots of AI all along, but what does it mean to understand what you recognize?  There are so many dimensions to that. For example, the earliest cave person who saw a predator and, not recognizing it as such, died. Those who witnessed the event formed memory patterns - models - which allowed them to equate that creature with loss of life.  Does the Tesla that ran over some child in a crosswalk modify its model with such rigor? I think Minsky co-authored a book The Emotion Machine which wrestled with some of those dimensions.

 

Finally, the video comes from an agency which has a financial stake in the game; notice that it does not promote Watson in that video to the degree it did in the past.  If Watson is that powerful, why resort to GenAI?  We live in a time when public expectations are driven by the necessity to raise funding through means which tend to over-describe the possibilities.  Recall the hype about expert systems, and all the funding behind them, and now, extinct businesses and business models.  Will history repeat itself?

 

I remain hopeful, in the Engelbartian sense, that we find legitimate ways to use these tools to augment human capabilities.  In some narrow domains, we are seeing that.  About AGI, I remain skeptical.

 

Cheers,

 

Jack

 

_______________________________________________
CG mailing list -- c...@lists.iccs-conference.org
To unsubscribe send an email to cg-l...@lists.iccs-conference.org