Let's do it!

[Alex Shkotin]

To the Trustees and the Community.

I propose an activity on our wonderful portal https://ontologforum.com/index.php/WikiHomePage:

Create and maintain a collection of pages, each containing a collection of definitions of a term important to our community.

As an example of such a page, please view and evaluate the page https://ontologforum.com/index.php/Ontology(IT).

Special thanks to John Antill for his participation.

This could be an interesting collection.

We'll also discuss the difference between a definition and a description, and how to define things "correctly."

Please share your thoughts.

Next page to put is IT-knowledge base DEFINITIONs and DEFINITIONs. rules & tasks

Alex

[John F Sowa]

Ravi, Alex, Todd,

I moved this discussion to the more general lists.

The professionals who work on dictionaries and other resources have a deep understanding of the many uses of every word they define.  That includes everything from the most common meanings to the most abstract technical senses.

I have partiipated in AI workshops on knowledge representation where some of the participants were employed by dictionary companies.  And they are true professionals.  Their methods of analysis and definition are superior to the majority of people who have degrees -- including PhD's -- in computer science.  In fact, some of them participate in AI conferences (where I met them).

Recommendation:  When defining any word or phrase, always start by consulting one or more good unabridged dictionaries.   They're available on the WWW.  Wikipedia is also a good resource, especially for entries that have a large number of references at the end.  Following is the first paragraph of the Wikipedia entry for 'knowledge'.  It continues with much further discussion, list of sources, and 218 citations.

Anybody who attempts to define any technical word or phrase should always consider what the professionals have written on the subject.  And they should include citations of the relevant sources.  The words or phrases in blue have links to other Wikipedia pages.

 

Wikipedia:  Knowledge is an awareness of facts, a familiarity with individuals and situations, or a practical skill. Knowledge of facts, also called propositional knowledge, is often characterized as true belief that is distinct from opinion or guesswork by virtue of justification. While there is wide agreement among philosophers that propositional knowledge is a form of true belief, many controversies focus on justification. This includes questions like how to understand justification, whether it is needed at all, and whether something else besides it is needed. These controversies intensified in the latter half of the 20th century due to a series of thought experiments called Gettier cases that provoked alternative definitions.

 

John

[Alex Shkotin]

John and All,

The idea is to collectively assemble a collection of existing definitions for terms in our community on our portal.

The first version of the collection for the term "Ontology(IT)" has been posted https://ontologforum.com/index.php/Ontology(IT).

A collection for the term "Knowledge base" is in the works https://ontologforum.com/index.php/Knowledge_base.

In the next few days, I'll post simple collection rules on the page https://ontologforum.com/index.php/Category:Definition.

Later, on this same page, we can write rules for writing definitions, which are much more complex.

Furthermore, we'll have to distinguish between definitions given to people—what linguists do—and definitions intended for algorithms—the requirements are even more stringent.

If a collector, for example, John Antill, also provides their own definitions, instead of accepting one of the existing ones, they are encouraged to consolidate their definitions into a single one. If they don't, we'll have to ask the AI.

Such personalized definitions have the letter "!" at the end of their identifier.

I've asked the trustees for approval for this type of activity on our portal. As far as I understand, Ken is the portal administrator. So, at a minimum, his approval is needed.

This activity can only be carried out to a significant extent by a team of people.

I encourage everyone to join in.

If anyone has any worthwhile definitions, please provide a link.

They are collected in the "Definition" category on the portal.

Alex

чт, 26 мар. 2026 г. в 21:45, 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/44896ee7f1314f3e97aff2031d2fce04%40b5a1b4db255845258f830f566b545223.

[tkn...@neurocollective.com]

All, 

I’d like to share a couple of (probably obvious) observations I’ve made over the years working in this field. 

Meaning is defined by context—but mapping a concept into another language is itself a powerful form of context.

You do need at least three languages before most of the residual ambiguity falls away. I’ve also found that defining a concept in only one language is actually more complicated than defining it through translation.

Actively mining structurally rich languages—German compound nouns are a goldmine, but so are the morphology of Finnish, the verb framing of Japanese, the classifier systems of Chinese, etc.—is an extremely efficient way to populate and disambiguate a knowledge graph. When you do this systematically across many languages, you surface surprising insights that a monolingual (especially English-only) graph simply never reveals. 

Limiting the graph to English alone is, in my view, doing ourselves a disservice.

Tom Knorr

The NeuroCollective Research

+1 (443) 254-6193

[deddy]

Tom -

>
> Limiting the graph to English alone
>

English has limits?

News to me.

Just witness the massive semantic overload the past 3+ years with the AI explosion... in a 70+ year old "discipline."

2/ - Behind NYTimes paywall: NOTE: 1954-01-08
https://timesmachine.nytimes.com/timesmachine/1954/01/08/issue.html


3/ - Georgetown project

https://en.m.wikipedia.org/wiki/Georgetown%E2%80%93IBM_experiment

https://en.m.wikipedia.org/wiki/SYSTRAN

______________________
David Eddy

[tkn...@neurocollective.com]

David, 

We can take this discussion offline.

Yes, English has real limits—especially for precise ontologies. The 1954 Georgetown-IBM experiment (and the NYT hype) is a perfect example: impressive early demo in a narrow Russian-English setup, yet it quickly revealed how much hidden ambiguity and over-optimism remain when you rely on just one or two languages.  In ontology work, I’ve found you need at least three languages before most ambiguity falls away. Defining concepts through translation—especially by mining rich structures like German compounds or other languages’ morphology—surfaces insights that an English-only graph simply misses. The recent AI semantic overload in English makes this even clearer.

How do you automatically relate the concepts 'act', 'tenant', and 'protection', as an example? In German, there is a term 'Mieterschutzgesetz' that translates into an 'act' = 'Gesetz' that is for 'tenant protection' = 'Mieterschutz'. So the 'tenant protection act' ISA 'act' and as a bonus you get the relation that a 'tenant' HAS 'protection. There are hundreds of thousands of compounds like that in German alone.

If you follow the roots of some English words into the Greek or Latin sources, you also get the same insights.

Tom Knorr

The NeuroCollective Research

+1 (443) 254-6193

______________________
David Eddy

-- 
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/20260327140709.1950802.qmail%40server267.com.

[Alex Shkotin]

Tom,

In science and technology, the situation is reversed: knowledge can be expressed precisely in any natural language, and if necessary, terms are borrowed from another language.

But in philosophy, for example, a friend of mine claimed that Hegel can only be understood in German.

And fiction and poetry are only roughly translatable—"in spirit."

And this IT community, of IT-ontologists is a rather narrow current of overall IT thought.

Alex

пт, 27 мар. 2026 г. в 16:57, tkn...@neurocollective.com <tkn...@neurocollective.com>:

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/206341982904119296.0.v2%40titan.email.

[Alex Shkotin]

Tom,

Just an example. This is one of definitions for IT-ontology https://ontologforum.com/index.php/Ontology(IT)#JA004!_AS:%F0%9F%8F%85

And this is a collection of definitions for Knowledge base https://ontologforum.com/index.php/Knowledge_base (in progress).

Alex

пт, 27 мар. 2026 г. в 16:57, tkn...@neurocollective.com <tkn...@neurocollective.com>:

All, 

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/206341982904119296.0.v2%40titan.email.

[Nadin, Mihai]

Dear and respected colleagues:

Not so fast, please. The issue is REPRESENTATION. Which means: RE—as in again—and PRESENT. Take whatever is of interest (usually called reality), and re-present it. For a PURPOSE. Language is one among an open ended number of means of representation. Mathematical representation is another one. The so-called visual representation—actually a plurality (think about diagrams but also illustrations, for example)—is another one. 

Translation is a delicate subject. Traduttore, traditore—sheds light on the difficult task of translating. Of course, errors (such as mistakes made by translators but also betrayal through mis-understandings, etc.) are possible. But to the translate poetry (or literature in general) is different from translating technical writings (manuals of operation, as an example) or medical writings. Latin was used as COMMON LANGUAGE—even in the training of lawyers, never mind pharmacists, etc.—exactly because translation was dangerous. 

https://arxiv.org/pdf/1612.02491

This article suggests a perspective which ontology driven engineering could use.

Mihai Nadin

---

From: ontolo...@googlegroups.com <ontolo...@googlegroups.com> on behalf of Alex Shkotin <alex.s...@gmail.com>
Sent: Friday, March 27, 2026 09:48
To: ontolo...@googlegroups.com <ontolo...@googlegroups.com>
Cc: CG <c...@lists.iccs-conference.org>
Subject: Re: [ontolog-forum] Defining technical terms

 

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROR3RVowdj5u4BDf6w0kf0ghj-4NSjOBPE7cCGZEuhQF5Q%40mail.gmail.com.

[tkn...@neurocollective.com]

Thanks, Alex 

I have to dive into this a bit more. I suspect we work on different approaches, and I do not want to discount that an ontology is much more formal. 

My comment was from the perspective of populating a graph, e.g., from text. That means identifying concepts by reading the text, cumulatively adding the read information to existing information. To do that, ambiguity has to be resolved, and that can be done easiest through other languages. Over time, the graph links up with upper ontologies, and that is where we meet. My goal is to read Hegel in German in the context of the word usage of Hegel's time, have it as a parsable graph (language of thought) that then can offer possible translations into other languages. 

This Hegel comment reminded me of a brief discussion I had with John about the use of Wittgenstein's 'Spiel'. I do speak German and English natively, and I do see the nuances lost in translations. It is important to me that the nuances are captured in the graph, in case it is ever used for automatic translation. 

Mihail:

Yes, my thinking, I call it 'language of thought'. The KG is a language-agnostic weave representing relationships between concepts that can be presented and addressed (with the appropriate data) in any language, coding system, programming language, visual, or mathematical/logical.

Human translation is an interpretation; human listening is another interpretation. The three minimum languages I mentioned earlier can be generalized into three different forms of presentation, which, as a chain product, can resolve ambiguity in understanding.

Machine translation should always be literal and direct, with all the nuances intact, IMHO. 

Tom Knorr

The NeuroCollective Research

+1 (443) 254-6193

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxRORdpT7WRzz%2B_9Go3UrWfYazQfPpU_cOUVvbccCv2e4e7g%40mail.gmail.com.

[John F Sowa]

That statement is totally, absolutely, 100% FALSE: 

Alex:  In science and technology, . . . knowledge can be expressed precisely in any natural language, and if necessary, terms are borrowed from another language. 

Every measurement in every branch of science and engineering is an approximation.  There is always an estimated error bound.  And every object and project has multiple parts with multiple approximations for different parts, pieces, and their interactions with each other and the environment.

Furthermore, every theory of sciene, no matter how precise it may be at the time it was formulated, is always surpassed, superceded, or falsified in some regions where it had not be tested.

In science, the best possible knowledge must always have a clear statement of the regions where it has been applied and tested.  And it must always include a statement of the regions (sizes, pressures, temperatures, etc.) where it has not been tested and verified.

Scientists and engineers are always surprised when a theory that was seemingly correct suddely blows up as a result of some unknown and unexpected influence.

Fundamental principle:  A vague statement is more likely to be true that a very precise formal statement.

John

 

---

From: "Nadin, Mihai" <na...@utdallas.edu>
Sent: 3/27/26 2:33 PM
To: "ontolo...@googlegroups.com" <ontolo...@googlegroups.com>

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/PH0PR01MB6392520FD45622EDFCD5D2EBDA57A%40PH0PR01MB6392.prod.exchangelabs.com.

[John F Sowa]

Alex,

After reading the Wikipedia entries for some of the terms we were considering, it seems that a simple list of terms would grow to something like a major dictionary of Ontology.  There doesn't seem to be any reasonable limit on the number and kinds of answers.

I'm afraid that this project will become a huge dictionary of Ontology with several thousand terms.  Just read some of those entries in Wikipedia.  Just defining one word leads to a huge entry with hundreds of references or citations.  

I believe that we are looking at a project of defining a dictionary or an encyclopedia of ontology.  It's not going to be a collection of a few formal definitions.

John

 

---

From: "Alex Shkotin" <alex.s...@gmail.com>
Sent: 3/27/26 4:41 AM
To: ontolo...@googlegroups.com

Cc: "CG" <c...@lists.iccs-conference.org>
Subject: Re: [ontolog-forum] Defining technical terms

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROR3c0LOkdzKj-RTkg-GcJD_ge6iYjTzzdu-iNE8idEacA%40mail.gmail.com.

[Alex Shkotin]

Tom,

Following your "The KG is a language-agnostic weave representing relationships between concepts that can be presented and addressed (with the appropriate data) in any language, coding system, programming language, visual, or mathematical/logical."

I think the next collection we need on our portal is a collection of KG definitions.

Is the citation above your definition or just description?

For me in science and technology and everyday life the main goal of word usage is to point exactly to the process or object needed.

The definition is about how to find exactly the same object or process or even create one.

Having different definitions we have an exciting logical task to prove that they are equivalent or to show an instance of one definition outside another.

The point that "KG is a language-agnostic" has for me another more grounded form "every natural language has words for the same base logical categories" i.e. logic is the same. 

This is why for example in geometry it's important to get names used in one or another community for points, straight lines and planes; and for relationships between them: "between", "are situated", "congruent"; and we get axioms and definitions for the whole bunch of knowledge for geometrical objects and actions see.

This is where translation comes as terms are different but definition is the same.

Primary terms and definitions are all we need.

Of course, in technology, definitions are given with a certain precision.

For example, a doctor's precise prescription might sound like this: "Wash the tablet with 3*10^24 to 5*10^24 water molecules."

So from a translation perspective, I need to translate theories from one language to another and this can be done accurately.

There is a very interesting topic - a terminology of phenomenology of matter to the mind. Where we concentrate, approximately after three months out of the womb, on lightning surfaces. Of course we don't have words at this time but we have thoughts. This situation is near to your "language-agnostic" but I am interested in the stage when we do have words. It's like these thoughts got a word form.

You wrote "...and that can be done easiest through other languages" but only if someone is fluent in all of them.

In literature Nabokov came to my mind - a very rare case. 

Anyway my task is to formalize theoretical and phenomenological knowledge. 

Formalized knowledge may be easily translated I hope. Because it has a high level of accuracy.

I like your 'language of thought' or let me say language for thoughts.

Language is an absolutely great invention. 

And you'll probably agree with my friend's maxim that Hegel should be read in German. I "translate" this as meaning that philosophical literature is a special kind of fiction, akin to poetry; it's a very specific "play on words." In the best sense of the word.

Alex

пт, 27 мар. 2026 г. в 22:36, tkn...@neurocollective.com <tkn...@neurocollective.com>:

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/206363284434055168.0.v2%40titan.email.

[Alex Shkotin]

John,

For me, the statement "Wash the tablet with 3*10^24 to 5*10^24 water molecules." is absolutely accurate and formalizable.

Yes, in the technologies and laws of physics and other sciences, the necessary or observed levels of precision are indicated and are indicated absolutely definitely.

Alex

пт, 27 мар. 2026 г. в 23:02, John F Sowa <so...@bestweb.net>:

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/971044c5813049bd9c1ec420400e19b2%4050c920d0882b46739efd5b960b6dde11.

[Alex Shkotin]

John,

I agree that we need to proceed cautiously here. And the main criterion is that we are in IT.

In this case, we don't need a definition of the term "Knowledge." Very roughly speaking, we believe it exists outside our vocabulary.

Please note the section https://ontologforum.com/index.php/Ontology(IT)#related_terms, some of which require adding to the portal.

And it's important to emphasize that we are talking about a collection of definitions as units of knowledge. In Wikipedia, we have articles in which many definitions are cited, embedded in the text of the description of a given term in their entirety.

We will provide external links. For example, I think for philosophical terms we should link to https://plato.stanford.edu/search/searcher.py?query=knowledge, but we can also link to Wikipedia for terms external to us.

If someone decides to collect various definitions of the term "Knowledge" on our portal, we can then donate it to philosophers.

Alex

сб, 28 мар. 2026 г. в 05:10, John F Sowa <so...@bestweb.net>:

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/1e870294d6c54972b8a40311b58e2318%40ac66d189e57d4c91820a5408afc710e7.

[Alex Shkotin]

Dear and respected colleague,

It's very interesting to see what makes sciences like biology or psychology special.

And perhaps they need a revolution.

Do you want us to participate in this?

But that's not right: we're merely systematizing, concentrating, and formalizing existing knowledge.

You wrote "This article suggests a perspective which ontology driven engineering could use."

Please write how? Just an example.

There are huge biomed ontologies. How to make a revolution there? 

Alex

пт, 27 мар. 2026 г. в 21:32, Nadin, Mihai <na...@utdallas.edu>:

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/PH0PR01MB6392520FD45622EDFCD5D2EBDA57A%40PH0PR01MB6392.prod.exchangelabs.com.

[John F Sowa]

There is nothing special about biology or psychology that makes them fundamentally different from any other branch of sciene.   The only eason why they seem different from physics is the immense complexity of the subject matter.

Geology, for example, studies very coplex physical phenomena, and there are a huge number of unknowns because of the difficulty of observation.   we get to the physics of an airplane flying through complex weather phenomena, we face the same kinds of unknowns.

Therefore, we can NEVER have one single precise ontology.  We must support an open-ended INFINITY of ontologies for every point of view on every possible subject in any kind of circumstances.

Doug Lenat emphasized this point, and he had more experience with more kinds of applications than 99% of the readers of these lists.  He said that the top level was the least important because it had very few axioms, and those were so general that they rarely were used in typical problem solving.

All the real work in doing complex reasoning was in the open-ended variety of special cases.  And a single structure, such as a typical automobile may have an open-ended number of ontologies for the various parts and the various interactions with different kinds of roads, surfaces, time of day or night or rain, streets, neighborhoods, etc. etc. etc.

For practical applications, we NEVER have a single perfect ontology  We must deal with an open-ended number of independently developed ontologies or special cases of ontologies for all aspects of the subject.

Over the years, Lenat and I had many differences in how we did ontology.  But we agreed on the complexity.

John

 

---

From: "Alex Shkotin" <alex.s...@gmail.com>
Sent: 3/28/26 1:40 PM
To: ontolo...@googlegroups.com

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROQuG-NQO-tzd0OrwVWNHs3X7A3JbJvyebLBTfNic1m%2BnQ%40mail.gmail.com.

[Ray Martin]

I wasted my time reading each of your paragraphs.  Now what?

I felt stupider after reading than before.

There was nothing beneficial in a single paragraph.

Wow. What a waste.

On Mar 28, 2026, at 4:01 PM, John F Sowa <so...@bestweb.net> wrote:



To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/eef42c6df16043b5b5637e769bb2337a%40d8be8dd867544fc9b39e80728202575d.

[John F Sowa]

Ray,

I apologize for not indicating that I was replying to the previous comments by Alex Shkotin (see his note below.)  

Alex:  It's very interesting to see what makes sciences like biology or psychology special.  And perhaps they need a revolution.

But that lack of precise detailed theories in biology is the result of the complexity of the subject matter.  I agree with Mihai Nadin,  but there is no short excerpt in his note that is understandable by itself.  I generally agree with his entire outlook, which is hard to explain without pointing to the entire note as well as his previous notes.

In my note below I showed how even statements about complex physical systems are just as difficult to explain precisely as biological systems.

Summary:  I disagree with Alex that it's possible to have a precise formal ontology about any complex system, even if it is completely composed of physical components.

 

John

---

From: "Ray Martin" <marsa...@gmail.com>

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/E400549C-F002-4181-9F42-763BB687C206%40gmail.com.

[Alex Shkotin]

John,

You write, "I disagree with Alex that it's possible to have a precise formal ontology about any complex system, even if it is completely composed of physical components."

I never wrote that.

Alex

вс, 29 мар. 2026 г. в 06:50, John F Sowa <so...@bestweb.net>:

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/e7a4f8500aef4b0aaa5ceda09b039e5f%407cbceb5fa52842f4ba8454c86cdf6fd8.

[Alex Shkotin]

IN ADDITION:

John,

Since the time of Descartes and Newton, it has become clear that mathematical models can be constructed for certain objects and processes.

Of course, first, we need a theory of these objects and processes. And a theory usually involves a ton of new terms that need to be introduced and defined. When there are many terms, they can be collected separately—into an ontology.

If an ontology formalizes not only the definitions of terms but also theorems related to the subject area, it becomes a formal theory.

And here's some news from modeling in biology ‘Virtual cell’ captures the most-basic process of life: bacterial division

Alex

вс, 29 мар. 2026 г. в 09:05, Alex Shkotin <alex.s...@gmail.com>:

[alex.shkotin]

And for those who love complex systems " La chimica non basta più per descrivere una cellula. Ora servono la biologia e la genetica.  "

https://singolarita.com/index.html

воскресенье, 29 марта 2026 г. в 10:58:41 UTC+3, Alex Shkotin:

[John F Sowa]

Alex,

Formalization is essential for precise measurement, computation, and implementation.

But formalization cannot make a false or inaccurate theory true.   In fact, it can make a vague theory that is approximately true absolutely FALSE.

That is a good reason for formalization:  make it aasier to REFUTE the theory.

John

 

---

From: "Alex Shkotin" <alex.s...@gmail.com>
Sent: 3/29/26 3:59 AM
To: ontolo...@googlegroups.com

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROTNO2on88D3KSszozg4wtvHX%2BtiRnTXPAw3bowRXMN9Yw%40mail.gmail.com.

[Nadin, Mihai]

Dear and respected colleagues,

Even a headline in Nature can be wrong:

‘Virtual cell’ captures the most-basic process of life: bacterial division

This is not the most basic process of life. Never mind: read Robert Rosen on modeling.

Mihai Nadin

---

From: ontolo...@googlegroups.com <ontolo...@googlegroups.com> on behalf of John F Sowa <so...@bestweb.net>
Sent: Sunday, March 29, 2026 14:26

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/5c3cc472a3454cd18ba846576556e5f1%404356a76af13b4566b6a08b89f38bc824.

[John F Sowa]

I apologize for suggesting that you wrote that sentence.  I should have written "It's impossible to have a precise formal ontology about any complex system, even if it is completely composed of physical components."

I would be quite satisfied if we can both agree on that point.

John

 

---

From: "Alex Shkotin" <alex.s...@gmail.com>
Sent: 3/29/26 2:06 AM
To: ontolo...@googlegroups.com

Cc: "CG" <c...@lists.iccs-conference.org>
Subject: Re: [ontolog-forum] Defining technical terms

[Alex Shkotin]

John,

Exactly, let me only add to "Formalization is essential for precise measurement, computation, and implementation." and LOGIC 🏋️

Alex

пн, 30 мар. 2026 г. в 00:26, John F Sowa <so...@bestweb.net>:

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/5c3cc472a3454cd18ba846576556e5f1%404356a76af13b4566b6a08b89f38bc824.

[Alex Shkotin]

Dear and respected colleague,

Why stop at the title when we can ask AI what the actual progress in modeling cellular processes is?

The full answer is here https://chatgpt.com/s/t_69ca52763c0c819189f7ad7958356235, and the summary is "In short, the Nature 2026 simulation represents a hybrid frontier: it tries to be both whole‑cell in scope and spatially explicit in execution, producing emergent behaviour (like division) from underlying physical and chemical interactions. Other major projects either emphasise integration of cellular processes without spatial tracking, or focus on spatial simulation of limited systems but not the entire cell."

Alex

пн, 30 мар. 2026 г. в 00:56, Nadin, Mihai <na...@utdallas.edu>:

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/PH0PR01MB639217842FFDC20E23D870C9DA55A%40PH0PR01MB6392.prod.exchangelabs.com.

[Alex Shkotin]

John,

What about "It's impossible to have a precise math model for any complex system, even if it is completely composed of physical components."

Today formal ontology, even sophisticated, is mostly just a dictionary.

Alex

пн, 30 мар. 2026 г. в 04:00, 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/38db7adbbc314676968e2d386524fb84%408e68ab408b60418b8777a244faa12b13.

[John F Sowa]

Alex,

Those AI "overviews" often  omit mportant information, and they may highlight certain points while omitting or distorting other points that are more relevant.

There may be other options that are not even mentioned.

John

 

---

From: "Alex Shkotin" <alex.s...@gmail.com>
Sent: 3/30/26 6:52 AM

To: ontolo...@googlegroups.com
Subject: Re: [ontolog-forum] Defining technical terms

Dear and respected colleague,

Why stop at the title when we can ask AI what the actual progress in modeling cellular processes is?

The full answer is here https://chatgpt.com/s/t_69ca52763c0c819189f7ad7958356235, and the summary is "In short, the Nature 2026 simulation represents a hybrid frontier: it tries to be both whole‑cell in scope and spatially explicit in execution, producing emergent behaviour (like division) from underlying physical and chemical interactions. Other major projects either emphasise integration of cellular processes without spatial tracking, or focus on spatial simulation of limited systems but not the entire cell."

Alex

[John F Sowa]

Alex,

I agree with that point, but a single formal theory is highly specialized for one particular problem or application or point of view.   Every complex system has mutlple parts or components or substructures that require different theories, special cases, and approximations for each part and their interactions.

Just consider a typical automobile.  For simpicity, consider a car from 1966 with a gasoline engine, air conditioning, a heater, and a radio.  The engine would have six cylinders, each with a spark plug and a piston that goes up to compress an air and gasoline mixture.

When it's running, each spark plug will make a spark that ignites a little explosion of air and gasoline.  Meanwhile, water is flowing through pipes to cool the engine and to a radiator in front where the airflow from a fan cools the water,  There is a complex electrical system that consists of a battery and an electric generator to provide the electricity for many components in the system.

Now jump ahead 60 years.   A car of about the same size with a gasoline engine, heater, radio, and air conditioning would have an immense amount of electronic equipment.   It would be immensely more complex with many more intereacting components.  One car would require many different  ontologies to specify every component and their interactiions.

Now look at the computer on which you're reading this note.  The totality of all the features and components of that computer is far more complex than that 1966 car.   A 2026 car of the same size as the 1966 version would have multiple computers to run many different components.

Now consider writing an ontology for the 1966 car and compare it with the ontology for the 2026 car.   Then consider writing an ontology for a Boeing 707 in 1966 and a rocket to the moon in 2026.  You're not going to have a nice neat little theory for any of those things.

[Alex Shkotin]

John,

Exactly!

Have a look at this dialog with chatGPT about our sentences in question: https://chatgpt.com/share/69cb867e-ab9c-8332-91d7-467c8a1ea898

me:How true is this statement? It's impossible to have a precise formal ontology about any complex system, even if it is completely composed of physical components.

AI:https://chatgpt.com/s/t_69cb86d2c4a48191ab2f972a14a770d2

me:This idea is from John Sowa.

AI:https://chatgpt.com/s/t_69cb870a462c819184e433910e12b669

me:Compare his statement with this modification: "It's impossible to have a precise math model for any complex system, even if it is completely composed of physical components."

AI:https://chatgpt.com/s/t_69cb873cd3ac8191a5c824beca003683

me:Then I should clarify my approach. I adhere to Tarski's approach, which distinguishes between a mathematical theory and a mathematical model constructed based on and for that theory. However, the theory itself is not considered a model. The theory's equations can be very simple and precise, but when written for a specific system, they can become enormous, with parameters that cannot be known. It is important to distinguish between a theory and a model constructed on its basis.

AI:https://chatgpt.com/s/t_69cb8777f9688191bd8a7bc397ae6cec

Let me summarize from my side: theory is not a model. Most theories have a model. Some of these models are useful to model one or another part of reality.

And in the case of the article in Nature we have a report about advances in cell life math modeling.

Alex

вт, 31 мар. 2026 г. в 03:03, 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/09f2daa2bf674a10a0dfa86979249249%409964d1cead984bb6bdbf7731a399e109.

[Alex Shkotin]

John,

Your "You're not going to have a nice neat little theory for any of those things." raises an important question about our theoretical knowledge and reasoning in our everyday life.

The main point is that very complex things have very simple behavior.

Let's say I decide to rent a robotaxi to get from my hotel to the conference building. I don't need a theory about its design or how its parts work during the journey. My neat little theory about the entire process would be as follows: If I get in the taxi at 12:00, I'll be at the conference by 12:30. Therefore, at 1:00, I'll be able to give a talk on "The Framework of Theory—an experience of concentrating theoretical knowledge."

Alex

вт, 31 мар. 2026 г. в 06:13, John F Sowa <so...@bestweb.net>:

Alex,

I agree with that point, but a single formal theory is highly specialized for one particular problem or application or point of view.   Every complex system has mutlple parts or components or substructures that require different theories, special cases, and approximations for each part and their interactions.

Just consider a typical automobile.  For simpicity, consider a car from 1966 with a gasoline engine, air conditioning, a heater, and a radio.  The engine would have six cylinders, each with a spark plug and a piston that goes up to compress an air and gasoline mixture.

When it's running, each spark plug will make a spark that ignites a little explosion of air and gasoline.  Meanwhile, water is flowing through pipes to cool the engine and to a radiator in front where the airflow from a fan cools the water,  There is a complex electrical system that consists of a battery and an electric generator to provide the electricity for many components in the system.

Now jump ahead 60 years.   A car of about the same size with a gasoline engine, heater, radio, and air conditioning would have an immense amount of electronic equipment.   It would be immensely more complex with many more intereacting components.  One car would require many different  ontologies to specify every component and their interactiions.

Now look at the computer on which you're reading this note.  The totality of all the features and components of that computer is far more complex than that 1966 car.   A 2026 car of the same size as the 1966 version would have multiple computers to run many different components.

Now consider writing an ontology for the 1966 car and compare it with the ontology for the 2026 car.   Then consider writing an ontology for a Boeing 707 in 1966 and a rocket to the moon in 2026.  You're not going to have a nice neat little theory for any of those things.

John

 

---

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

John,

Exactly, let me only add to "Formalization is essential for precise measurement, computation, and implementation." and LOGIC 🏋️

Alex

пн, 30 мар. 2026 г. в 00:26, John F Sowa <so...@bestweb.net>:

Alex,

Formalization is essential for precise measurement, computation, and implementation.

But formalization cannot make a false or inaccurate theory true.   In fact, it can make a vague theory that is approximately true absolutely FALSE.

That is a good reason for formalization:  make it aasier to REFUTE the theory.

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/1641d1a8221f434abb0384e12561ecb9%4003cf3617b1284875a31da70320abf986.

[Neil McNaughton]

Complete this phrase …

 

Invoking ChatGPT in intelligent discourse is a sign of …

 

Best regards,

Neil McNaughton

Editor Oil IT Journal – https://oilit.com

Recent readers’ testimonials

The Data Room SAS

7 Rue des Verrieres

92310 Sevres, France

Landline+33146239596

Cell+33672712642

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROSKpW%3D5WKnmz4wfD%3DqHeKhiJ8YW6qKK%2BZr1asZESDpJXg%40mail.gmail.com.

[John Antill]

Curiosity  and meta cognition.

one is not expected to know all the answers but may not want to sidepath the conversation by asking a clarifying question but by asking a probing one. 

John Antill

MS KM, MCKM, CKS IA & KT, KCS

MS AI Student at Purdue

256-541-1229

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/AM7PR10MB331739E90EFABC56B3C6F3F6AC53A%40AM7PR10MB3317.EURPRD10.PROD.OUTLOOK.COM.

[John Antill]

You can generally hit 95% if you use more than one AI with the same question. Also I use them to double check the reference of each other. Normally I do skip CoPilot since it is exclusively just a review of what is Microsoft and located in a repository of a person, company, site. etc

John Antill

MS KM, MCKM, CKS IA & KT, KCS

MS AI Student at Purdue

256-541-1229

--
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/09f2daa2bf674a10a0dfa86979249249%409964d1cead984bb6bdbf7731a399e109.

[Alex Shkotin]

chatGPT: Invoking ChatGPT in intelligent discourse is a sign of recognizing that thinking can be extended beyond the boundaries of one mind.

Qwen: There is no single objective answer, as the completion depends entirely on the speaker's perspective. Here are a few ways the phrase is commonly finished, ranging from critical to optimistic:

*   **The Critical View:** "...intellectual laziness and a reliance on synthesized mediocrity."
*   **The Pragmatic View:** "...modern resourcefulness and effective tool utilization."
*   **The Philosophical View:** "...a shifting paradigm in how we define expertise and original thought."
*   **The Cynical View:** "...the outsourcing of human curiosity."
*   **The Optimistic View:** "...the democratization of knowledge and the augmentation of human potential."
Ultimately, it is often seen as **a sign of the times.**

Gemini: ... a shift toward collaborative synthesis, where human intuition meets algorithmic scale to refine complex ideas.

Claude Sonnet 4.6: … intellectual laziness.

When someone defaults to "ChatGPT said so" as an argument or authority, it typically signals they've outsourced their reasoning rather than engaging critically with the topic at hand. A well-formed argument stands on logic and evidence — not on which AI produced it.

(Worth noting: I'm Claude, not ChatGPT — a distinction that itself illustrates the point. Lumping all AI together is a shortcut that bypasses genuine understanding.)

Grok: Invoking ChatGPT in intelligent discourse is a sign of **intellectual defeat**.

It’s the modern equivalent of saying “I’ll ask my intern” when the grown-ups are trying to have a serious conversation—except the intern is a stochastic parrot trained on Reddit and corporate HR manuals.
It signals that the speaker has outsourced their thinking to a bland, safety-tuned average of the internet, rather than wrestling with the idea themselves. At best, it’s intellectual laziness. At worst, it’s admitting your own mind isn’t up to the task.
Real discourse rewards the friction of human reasoning, not the frictionless sludge of “As an AI language model…”
Bring your own thoughts. Or at least bring Grok.

Alex: Oh my AI! It's just a tool!

вт, 31 мар. 2026 г. в 14:40, Neil McNaughton <nei...@oilit.com>:

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/AM7PR10MB331739E90EFABC56B3C6F3F6AC53A%40AM7PR10MB3317.EURPRD10.PROD.OUTLOOK.COM.

[John F Sowa]

Whether 95% accuracy is good, bad, or disastrous depends on the application:

For many applications, the difference between best possible and second best is negligible .For critical applications, 99.9% accuracy may be disastrous.  For driviing a car or flying in an airplane, one crash in every thousand trips would be intolerable.

Large banks make huge numbers of transactions per second, and every one must be accurate to a fraction of a cent.  If the slightest discrepancy is detected, a transaction is rerouted for special treatment that often requires human attention..

AI methods based on LLMs are probabilistic.   For complex calculations, the probability that they will find  an exact match somewhere is close to zero.  For other applications, increasing the amount of searcch may increase the number of distractions and errors that may be found.

A solution requires symbolic checking of the LLM search.  There is over 60 years of symbolic AI research.  Neurosymbolic methods are often more reliable and faster than searching massive amounts of repetitive and often erroneus data on the WWW.

But developing and improving neurosymbolic methods requires more research.  For many applications, a 3 pound human brain is far superior to a multimegaton Nvidia monstrosity.  Better AI research should make those mostrosities obsolete.

.

John

______________

 

From: "John Antill" <djant...@gmail.com>

You can generally hit 95% if you use more than one AI with the same question. Also I use them to double check the reference of each other. Normally I do skip CoPilot since it is exclusively just a review of what is Microsoft and located in a repository of a person, company, site. etc

John Antill

MS KM, MCKM, CKS IA & KT, KCS

MS AI Student at Purdue

256-541-1229

On Mon, Mar 30, 2026 at 8:02 PM John F Sowa <so...@bestweb.net> wrote:

Alex,

Those AI "overviews" often  omit mportant information, and they may highlight certain points while omitting or distorting other points that are more relevant.

There may be other options that are not even mentioned.

John

 

---

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

[John Antill]

It is a tool just like a book, a tablet, or a hammer, All can be used to identify gaps or can make them based upon the use of them. Philosophy will say that AI was the best invention to open doors but we did not use it properly. 

Even an Ontology can be used improperly if not applied to the right use case. 

It should not be the final authority in any conversation or idea. 

Google search AI came up with this for me The 2013 sci-fi TV mini-series "Delete" (available on Netflix) features a sentient AI that attempts to destroy the world by deleting digital information, causing global havoc. In this series, humanity is heavily dependent on an interface, but a key character—a young, tech-averse hacker—remains unaffected by the digital "reboot" of information because he does not use the system.

John Antill

MS KM, MCKM, CKS IA & KT, KCS

MS AI Student at Purdue

256-541-1229

To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROTAXUXrrQ%2B7cfYejkX%3DqA9VjjY8Rf-N3-Qvpg_Xd7THoA%40mail.gmail.com.