IT-ontology vs dictionary(OS-26 discussion)

17 views
Skip to first unread message

Alex Shkotin

unread,
Apr 23, 2026, 5:52:59 AM (7 days ago) Apr 23
to ontolog-forum

There was a very useful discussion yesterday, but I don't think I got my point across.

An IT ontology is, first and foremost, a dictionary of subject-matter terms.

It is structured and contains not only definitions of terms, or at least descriptions, but also relationships between terms that are much richer than those typically found in dictionary definitions, such as synonymy and homonymy.

Terms in an ontology are either converted into identifiers in a formal language. For example, in Genotype Ontology (GENO), the term "biological sequence" could be converted into an identifier like biological_sequence or biologicalSequence. However, in OBO Foundry, each term receives an internal, special system identifier—in this case, GENO_0000702. The term itself is placed in the label attribute of this identifier. What looks like this in OWL2\FS:

AnnotationAssertion(rdfs:label obo:GENO_0000702 "biological sequence"@en)


Thus, the system of terms used for a subject area is the foundation of an IT ontology.

They should be found either in identifiers or in identifier labels.


Assertion: each IT ontology can be used to extract a glossary of the subject area it covers.

The contents of the GENO IT ontology can be found here: https://dashboard.obofoundry.org/dashboard/geno/dashboard.html 


And Janet, if you'd like, please submit your own definition of the term IT ontology.

We'll add it to https://ontologforum.com/index.php/Ontology(IT).

This collection is awaiting its analyst.


Long live conceptualization, relationshipization, attributization!


Alex


John F Sowa

unread,
Apr 23, 2026, 10:00:05 AM (7 days ago) Apr 23
to ontolo...@googlegroups.com
Alex,

I saw your notes during the session, and I agree.   The terminology used in any application for any purpose is a specialized ontology for that application.  But different applications will use the same terminology with different definitions.

This is a very strong reason for designing ontologies with a very general upper level and an open-ended variety of special cases for an open-ended number of applications.  If you look at any large dictionary for any natural language, you'll find multiple special cases for every word.  

For example, the word 'leg' when applied to a human, a mouse, and a table will have very different definitions.  There is no such thing as a single universal ontology of everything.  Any upper level will allways be vague, with just a few very general distinctions.  And there will always be an enormous number of special cases.

Even for something as precise as banking, a general ontology for bank transfers may be standardised.  But every bank has a large number of special cases with specifications that are incompatible with the same words used by other banks.

Some people will say that's why we need universal standardizations.  But that is impossible.  Even if you could magically create a universal standard for all banks, that standard would break down the next day when various banks revise or invent new options.

John
 

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

deddy

unread,
Apr 23, 2026, 2:41:57 PM (6 days ago) Apr 23
to ontolo...@googlegroups.com
John / All -

>
> the word 'leg' when applied to a human, a mouse, and a table will have very different definitions.
>

My OAD — same as yours? — is much more diverse.


Place to start here — unknown to ontologists? — is George Miller's classic "Ambiguous Words."
https://www.thekurzweillibrary.com/ambiguous-words

13 simple Robert Frost words... 3.6T combinations & permutations.

Duly note the obvious, WordNet ONLY includes natural language "words" ... ignoring the highly specialized jargon / unnatural language used inside software applications.

Just checked... WordNet does not even include JCL as a word.


It is expected the motivated student — welcome to use Claude's COBOL conversion facilities — will connect the dots.

______________________
David Eddy
>  -------------------------
>  
>  FROM: "Alex Shkotin" <alex.s...@gmail.com>
>  
>  There was a very useful discussion yesterday, but I don't think I got
>  my point across.
>  
>  An IT ontology is, first and foremost, a dictionary of subject-matter
>  terms.
>  
>  It is structured and contains not only definitions of terms, or at
>  least descriptions, but also relationships between terms that are much
>  richer than those typically found in dictionary definitions, such as
>  synonymy and homonymy.
>  
>  Terms in an ontology are either converted into identifiers in a formal
>  language. For example, in Genotype Ontology (GENO), the term
>  "biological sequence" could be converted into an identifier like
>  biological_sequence or biologicalSequence. However, in OBO Foundry,
>  each term receives an internal, special system identifier—in this
>  case, GENO_0000702. The term itself is placed in the label attribute
>  of this identifier. What looks like this in OWL2\FS:
>  
>  AnnotationAssertion(rdfs:label obo:GENO_0000702 "biological
>  sequence"@en)
>  
>  Thus, the system of terms used for a subject area is the foundation of
>  an IT ontology.
>  
>  They should be found either in identifiers or in identifier labels.
>  
>  Assertion: each IT ontology can be used to extract a glossary of the
>  subject area it covers.
>  
>  The contents of the GENO IT ontology can be found here:
>  https://dashboard.obofoundry.org/dashboard/geno/dashboard.html
>  
>  And JANET, if you'd like, please submit your own definition of the
>  term IT ontology.
>  
>  We'll add it to https://ontologforum.com/index.php/Ontology(IT).
>  
>  This collection is awaiting its analyst.
>  
>  Long live conceptualization, relationshipization, attributization!
>  
>  Alex
>  
>  --
>  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/8e037f1a248840b48480e5f0f432e6e7%40e40d66d3280d4ac2b27527465d717bf4.
>  

Alex Shkotin

unread,
Apr 24, 2026, 4:56:59 AM (6 days ago) Apr 24
to ontolo...@googlegroups.com

John,


I agree that term definitions are absolutely crucial and are, in fact, the only source of interoperability: two systems can borrow definitions from each other and begin to understand them, but only if both definition systems are based on the same primary terms, which constitute the so-called top-level ontology. Otherwise, if the systems don't agree on the basic, primary terms, they will have to resort to examples of how the terms are used for the same objects and processes.

The importance of definitions is evident in the OBO Foundry approach, where they are dedicated to a special principle: Textual Definitions (principle 6).

"Summary

The ontology MUST have textual definitions for the majority of its classes and for top-level terms in particular."

Which I would strengthen to "Every term except primary terms must have a definition."

For example, for GENO, we have https://dashboard.obofoundry.org/dashboard/geno/dashboard.html 

image.png

This shows that sometimes an ontology doesn't even measure up to a comprehensive dictionary. How can a dictionary contain a term entry without a definition?


Alex



чт, 23 апр. 2026 г. в 17:00, John F Sowa <so...@bestweb.net>:

Alex Shkotin

unread,
Apr 24, 2026, 5:14:52 AM (6 days ago) Apr 24
to ontolo...@googlegroups.com
David,

For me, beginning with C++ we have the same situation in formal languages but the term is "polymorphic functions".
And for variables with the same id we have a point like this "Variables with the same identifier must be in different scopes, in different contexts."

Define a term or show a valuable set of examples of usage.
In John's example it would be like this: this is a leg of a human, this one is a leg of a table. Show me a leg of this duck.

Alex

чт, 23 апр. 2026 г. в 21:41, deddy <de...@davideddy.com>:
To view this discussion visit https://groups.google.com/d/msgid/ontolog-forum/20260423184149.3978393.qmail%40server267.com.

Nadin, Mihai

unread,
Apr 24, 2026, 2:09:09 PM (5 days ago) Apr 24
to ontolo...@googlegroups.com
Dear and respected colleague:
I will **run** a marathon.
The faucet will **run** all day.
She will **run** for office.
The software will **run** in the background.
....much more.
The ambiguity of language explains what is called the expressive dimension. No precision. The language of 2 letters behind the Turing machine is the substrate of a language: High precision. No expressive dimension. The ontology engineering endeavor reflects the pragmatic need to translate from a one open-ended language to another closed language. The only way to accomplish is by adopting the pragmatic view: "the meaning of a concept lies in its conceivable practical effects or habits of action" cf. C.S. Peirce).
Once the context is defined, the ambiguity is reduced (or eliminated).
Mihai Nadin



From: ontolo...@googlegroups.com <ontolo...@googlegroups.com> on behalf of deddy <de...@davideddy.com>
Sent: Thursday, April 23, 2026 11:41

To: ontolo...@googlegroups.com <ontolo...@googlegroups.com>
Subject: RE: [ontolog-forum] IT-ontology vs dictionary(OS-26 discussion)
John / All -

>
> the word 'leg' when applied to a human, a mouse, and a table will have very different definitions.
>

My OAD — same as yours? — is much more diverse.


Place to start here — unknown to ontologists? — is George Miller's classic "Ambiguous Words."

>  
>  And JANET, if you'd like, please submit your own definition of the
>  term IT ontology.
>  

>  
>  This collection is awaiting its analyst.
>  
>  Long live conceptualization, relationshipization, attributization!
>  
>  Alex
>  
>  --
>  All contributions to this forum are covered by an open-source
>  license.
>  For information about the wiki, the license, and how to subscribe or

>  ---
>  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

>  

--
All contributions to this forum are covered by an open-source license.
For information about the wiki, the license, and how to subscribe or

---
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.

John F Sowa

unread,
Apr 24, 2026, 6:08:15 PM (5 days ago) Apr 24
to ontolo...@googlegroups.com
I have always emphasized the open-ended continuum of aspects of the universe and even the aspects of thoughts within each of us.  I apologize for only listing 3 examples in the first sentence of that paragraph::

"For example, the word 'leg' when applied to a human, a mouse, and a table will have very different definitions.  There is no such thing as a single universal ontology of everything.  Any upper level will allways be vague, with just a few very general distinctions.  And there will always be an enormous number of special cases."   

But if I had listed 50 example, that would also be an inadequate approximation to infinity.  In general, the more precise your statement, the more likely it is to be FALSE.   A goal of having an absolutely precise formal ontology of any aspect of the universe is absolutely impossible.  

However, there are many applications for which absolute accuracy is essential.  For your bank account, 99% accuracy woulld be a failure.  The amount must be precise to a fraction of a cent.  That precision is possible because even a one-bit error is considered a failure.  And computer systems have encoding methods that guarantee either (1) extremely high accuracy or (2) a warning of a failure.

There is much more to say about these issues.

John
 

From: "Nadin, Mihai" <na...@utdallas.edu>

Dear and respected colleague:
I will **run** a marathon.
The faucet will **run** all day.
She will **run** for office.
The software will **run** in the background.
....much more.
The ambiguity of language explains what is called the expressive dimension. No precision. The language of 2 letters behind the Turing machine is the substrate of a language: High precision. No expressive dimension. The ontology engineering endeavor reflects the pragmatic need to translate from a one open-ended language to another closed language. The only way to accomplish is by adopting the pragmatic view: "the meaning of a concept lies in its conceivable practical effects or habits of action" cf. C.S. Peirce).
Once the context is defined, the ambiguity is reduced (or eliminated).
Mihai Nadin
 
John / All -

>
> the word 'leg' when applied to a human, a mouse, and a table will have very different definitions.
>

My OAD — same as yours? — is much more diverse.


Place to start here — unknown to ontologists? — is George Miller's classic "Ambiguous Words."
https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.thekurzweillibrary.com%2Fambiguous-words&data=05%7C02%7Cnadin%40utdallas.edu%7C062162a4cf414267b8b008dea167fee3%7C8d281d1d9c4d4bf7b16e032d15de9f6c%7C0%7C0%7C639125665227996108%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=eyenCsQuzcsayX08hjY9Z4nqiAfMXxW6cqTkyhN7WTQ%3D&reserved=0

13 simple Robert Frost words... 3.6T combinations & permutations.

Duly note the obvious, WordNet ONLY includes natural language "words" ... ignoring the highly specialized jargon / unnatural language used inside software applications.

Just checked... WordNet does not even include JCL as a word.


It is expected the motivated student — welcome to use Claude's COBOL conversion facilities — will connect the dots.

______________________
David Eddy


>  -------Original Message-------
>  From: John F Sowa <so...@bestweb.net>
>  

Alex Shkotin

unread,
Apr 25, 2026, 4:12:25 AM (5 days ago) Apr 25
to ontolo...@googlegroups.com
John,

"The amount must be precise to a fraction of a cent." Give me an example of a fraction of a cent on a banking account.
I think "The amount must be precise to a cent." is what we need here as money on the bank accounts are natural numbers.
By the way, using rational numbers for bank accounts is a big mistake.
And if we are talking about general financial rules what about GAAP?

The ability of a person to guess the meaning in which a word is used depending on the situation is amazing.

Alex

сб, 25 апр. 2026 г. в 01:08, 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/332ce390d1f44f969020da61a4fe5830%40200ba14ffc5345c78af98e9a6b572ca4.

John F Sowa

unread,
Apr 25, 2026, 12:39:08 PM (5 days ago) Apr 25
to ontolo...@googlegroups.com
Alex,

In banking, computations are carried out with more precision than just dollars and cents.  But the final result is rounded to the nearest cent. 

Natural languages are highly redundant.  That enables people. especially children, to guess the meaning of words they don't know.  Sometimes they make mistakes.  They may need several examples before the guess the intended meanng.

As a result of invasions by the Romans, by Nordic tribes, and by the French, English lost the complex endings of IndoEuopean.  As a result, the same word form may be used in multipe ways.  The Calvin and Hobbes cartoon is a good example:

Alex Shkotin

unread,
Apr 26, 2026, 5:08:07 AM (4 days ago) Apr 26
to ontolo...@googlegroups.com

JFS:"In banking, computations are carried out with more precision than just dollars and cents.  But the final result is rounded to the nearest cent. "

Yes, it is a subtle issue. So let's ask FIBO https://spec.edmcouncil.org/fibo/ontology/FND/Accounting/CurrencyAmount/AmountOfMoney

where we have "editorial note

  • This is an actual sum of money, not the measure of a sum of money in monetary units, although it has the same basic properties (decimal number with a currenct unit).

"

and I asked DeepSeek: "In FIBO ontology, the amount of money is represented as a decimal number https://spec.edmcouncil.org/fibo/ontology/FND/Accounting/CurrencyAmount/AmountOfMoney compared to integers, can this be dangerous?"

The answer is here. 

I like this "Difficulties with currency conversion: Exchange rates often contain 4–6 decimal places. When multiplying by them, the risk of getting a number like 19.999... instead of 20.00 is very high if a specialized decimal data type is not used."

But initially I wrote only about banking accounts.


JFS:"...English lost the complex endings of IndoEuopean." This is why it's easier to recognize that (PDF) English is a HOL language message #1X.

For example,

22(5) John seeks a unicorn.

John seeks a unicorn.

John seek-s (a unicorn)

(John seek-s (a unicorn))  --binary infix operator expression: (seek)s

Construction completed.


Here we have two unary operator: "a", "s" and one binary "(seek)s"


Alex



сб, 25 апр. 2026 г. в 19:39, 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/0526d03c0f014057bf30ea17380e1c3d%40973e60e240fd46b397bf071055425d2e.
Reply all
Reply to author
Forward
0 new messages