top layer ontologies, domain layer, application layer and data models?

[Michael DeBellis]

I was talking to someone today who is interested in: "literature about the difference between top layer ontologies, domain layer, application layer and data models?" I was wondering if anyone has any relevant papers. I told her that IMO data modeling isn't really what ontologies are for, that should be done with SHACL or UML. Interested in any papers or opinions.

Michael

[Alex Shkotin]

Michael,

Sorry, to begin I can't afford not to ask Claude3S like this:

 Is there any literature about the difference between top level ontologies, domain specific ones, application level ontologies and data models?  

C3S: 

Yes, there is literature that discusses the differences between top-level ontologies, domain-specific ontologies, application-level ontologies, and data models. These concepts are related to knowledge representation and organization, and they play different roles in various domains and applications.

1. Top-level ontologies (also known as upper ontologies or foundational ontologies):
Top-level ontologies aim to provide a general, domain-independent conceptualization of the world. They define the most fundamental categories and relations that are common across all domains. Examples of top-level ontologies include the Basic Formal Ontology (BFO), the Descriptive Ontology for Linguistic and Cognitive Engineering (DOLCE), and the Suggested Upper Merged Ontology (SUMO). Top-level ontologies serve as a starting point for building more specific ontologies and provide a common framework for knowledge integration and interoperability.

2. Domain-specific ontologies:
Domain-specific ontologies are focused on representing the concepts, relationships, and knowledge within a particular domain or subject area. They build upon the general concepts defined in top-level ontologies and provide more detailed and specialized knowledge for a specific field. Examples include ontologies for medicine, biology, engineering, finance, and many other domains. Domain-specific ontologies are essential for capturing and representing domain-specific knowledge, enabling knowledge sharing, and supporting reasoning and decision-making within that domain.

3. Application-level ontologies:
Application-level ontologies are ontologies that are designed and developed for specific applications or use cases. They often combine and extend concepts from both top-level ontologies and domain-specific ontologies, tailoring them to the requirements of a particular application or system. Application-level ontologies are closely tied to the implementation and functionality of the application they are designed for, such as knowledge management systems, decision support systems, or information integration systems.

4. Data models:
Data models are conceptual representations of the data structures and relationships within a specific system or application. They define the entities, attributes, and relationships that are relevant for storing and managing data in a particular context. Data models are typically used in database design, software engineering, and data integration tasks. While data models share some similarities with ontologies in terms of representing concepts and relationships, they are primarily focused on the logical and physical organization of data rather than capturing and representing domain knowledge.

The literature on these topics often discusses the distinctions between these different types of ontologies and data models, their roles in knowledge representation and management, their levels of abstraction and generality, and their relationships and interactions. Some key references in this area include works by Nicola Guarino, Guus Schreiber, Michael Gruninger, and the various ontology engineering methodologies and standards developed by organizations like the World Wide Web Consortium (W3C) and the Object Management Group (OMG).

AS: Could you give some references to begin with this topic?

C3S:

Sure, here are some useful references to begin exploring the differences between top-level ontologies, domain-specific ontologies, application-level ontologies, and data models:

1. "Ontologies: Principles, Methods and Applications" by Asunción Gómez-Pérez, Mariano Fernández-López, and Oscar Corcho (2004). This book provides a comprehensive introduction to ontologies and discusses the different types of ontologies, including top-level, domain, and application ontologies.

2. "Ontology Engineering in a Networked World" by Mari Carmen Suárez-Figueroa, Asunción Gómez-Pérez, and Mariano Fernández-López (2012). This book covers various aspects of ontology engineering, including the distinctions between different levels of ontologies and their relationships.

3. "An Introduction to Ontology Engineering" by Rilana Gracia, Johanna Völker, and Holger Lewen (2020). This book dedicates a chapter to discussing the taxonomy of ontologies, including top-level, domain, and application ontologies.

4. "Ontology Development 101: A Guide to Creating Your First Ontology" by Natalya F. Noy and Deborah L. McGuinness (2001). This highly cited paper provides a practical guide to ontology development and touches upon the different types of ontologies.

5. "A Methodology for Ontology Integration" by Nicola Guarino (1998). This paper discusses the integration of different ontologies, including top-level and domain ontologies, and their role in knowledge representation.

6. "Data Model Patterns: A Metadata Map" by David C. Hay (2006). This book explores the relationships between data models and ontologies, providing insights into their differences and complementary roles.

7. "Ontology-Based Data Access" by Diego Calvanese, Giuseppe De Giacomo, Domenico Lembo, Maurizio Lenzerini, and Riccardo Rosati (2012). This paper discusses the use of ontologies as a conceptual view over data sources, bridging the gap between ontologies and data models.

These references cover various aspects of ontologies, their different types, and their relationships with data models, providing a solid foundation for understanding the distinctions among these concepts.

Alex

чт, 25 апр. 2024 г. в 03:16, Michael DeBellis <mdebe...@gmail.com>:

I was talking to someone today who is interested in: "literature about the difference between top layer ontologies, domain layer, application layer and data models?" I was wondering if anyone has any relevant papers. I told her that IMO data modeling isn't really what ontologies are for, that should be done with SHACL or UML. Interested in any papers or opinions.

Michael

--
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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/5cf0bd73-e00e-4625-8d7c-23c4b6f788c5n%40googlegroups.com.

[Chris Partridge]

WRT: "IMO data modeling isn't really what ontologies are for, that should be done with SHACL or UML. "

IMHO, what makes something ontological is that it is aiming to picture what one thinks of as the real world explicitly. So if the data model is picturing the real world, it is in some sense ontological.

I realise that some people would say if it is in OWL (or somesuch) it is automatically an ontology - otherwise not.

However, we have built ontologies (pictures of the 'real' world) in UML, C++, JSON and XML. For us the base language is driven by the particular need at the time, it is the intended interpretation that (for us) makes it an ontology.

The ontological problems we wrestle with arise when we try to clearly represent in some computer (so formal in some sense) language the world.

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxRORNcgetb50gxU1Vdc%2BMu7BB6VSnEEVwESzcXTOgP_CDEg%40mail.gmail.com.

[Michael DeBellis]

IMHO, what makes something ontological is that it is aiming to picture what one thinks of as the real world explicitly. So if the data model is picturing the real world, it is in some sense ontological.
I realise that some people would say if it is in OWL (or somesuch) it is automatically an ontology - otherwise not.
However, we have built ontologies (pictures of the 'real' world) in UML, C++, JSON and XML. For us the base language is driven by the particular need at the time, it is the intended interpretation that (for us) makes it an ontology.
The ontological problems we wrestle with arise when we try to clearly represent in some computer (so formal in some sense) language the world.

The problem is that all these terms such as ontology and data model (also knowledge graph) are so vaguely defined and used to mean different things. So by an ontology I mean a formal model defined using set theory and logic. Given that definition I don't see how you could define an ontology in C++. You could use an ontology as a starting point for a data model, what people used to call the analysis model and then transform it into UML (which I think of as being at the design level) and from UML to code IMO that's not the same as saying you built the ontology in C++. The critical issue is the transformation. There are concepts in UML you don't have in OWL (I'll just use OWL as a shorthand for any language using a subset of FOL) and details in C++ that you don't worry about in UML. 

If you say an ontology is just a model of the real world than IMO you are saying  that a C++ object model is an ontology. One of the principles of good OOP design is that you base your classes on the business objects. I.e., you model the world. 

Michael

You received this message because you are subscribed to a topic in the Google Groups "ontolog-forum" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ontolog-forum/83loKPZIPzk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ontolog-foru...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAMWD8MqLgn0KW46gaUYEpWod7gVHLUuA-tG_w1mOmbBjcUi_VQ%40mail.gmail.com.

[Michael DeBellis]

Alex, that is great stuff, I'll pass that on to my client. Thanks!

Michael

You received this message because you are subscribed to a topic in the Google Groups "ontolog-forum" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ontolog-forum/83loKPZIPzk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ontolog-foru...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxRORNcgetb50gxU1Vdc%2BMu7BB6VSnEEVwESzcXTOgP_CDEg%40mail.gmail.com.

[Alex Shkotin]

Michael and Chris,

It should be noted that formal ontology is designed to store theoretical knowledge about one or another type of entities and processes. We usually store this knowledge in textbooks written in natural and mathematical languages. We store knowledge about specific objects (for example, the Moon) and processes (for example, an eclipse) in tables and databases. A formal ontology, for example in OWL2 or YAFOLL, can store both types of knowledge: theoretical and factual. 

What kind of knowledge processing can we do with this formalized knowledge? That is the question.⚖️

Alex

чт, 25 апр. 2024 г. в 17:51, Michael DeBellis <mdebe...@gmail.com>:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CALGFikco0b9W3%2BuapU_4wQNS8BQOnDcS2dXofTn7iRYaHq-QqQ%40mail.gmail.com.

[Alex Shkotin]

Hi, Anatoly!

For me every philosophical doctrine has TLO inside. But what about system engineering? Please remind me what you think about TLO? Do you use any?

Alex

[Chris Partridge]

Hi Michael,

WRT: So by an ontology I mean a formal model defined using set theory and logic. Given that definition I don't see how you could define an ontology in C++.

In what way is C++ not formal? Strictly speaking, it is probably more formal than most handwritten logic.

And you can easily treat your C++ model classes as sets, with the appropriate checks.

WRT: One of the principles of good OOP design is that you base your classes on the business objects. I.e., you model the world. 

In the MDA structure, they might start out based on the world.

We work a lot with operational systems, I cannot think of a case where the operational systems had a particularly good match with the world.

I think there is probably something like Friederike Moltmann's distinction between natural language ontology and foundational ontology going on here. The OOP is based upon the OOP language world.

I'd be pretty surprised to find a C++ system that was a good match with the 'real' world. In fact, I think the big challenge of semantic interoperability (and why we don't have it) is because the match is not good enough. A good example here is the way that SAP. Maximo, etc. treat replaceable part. They don't have a good enough match.

Chris

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CALGFikco0b9W3%2BuapU_4wQNS8BQOnDcS2dXofTn7iRYaHq-QqQ%40mail.gmail.com.

[Michael DeBellis]

Chris, all these things depend how you define your terms so when you ask:

In what way is C++ not formal? 

If by formal you mean it has a rigorous syntax and you can always tell a well formed formula and parse it without ambiguity (something not true about natural languages) then okay C++ is formal by that definition. In my experience though that isn't what people mean by formal. They usually mean and for sure I always mean that there is some well defined semantics. When you say Dog is a subclass of Mammal that has meaning in OWL based on logic and set theory (i.e., that the set of dogs is a subset of the set of mammals). That isn't the case with C++. You could make Mammal a subclass of Dog in C++ and there are no axioms or reasoners to tell you that is wrong. The guy who reviews your code will tell you you are an idiot but not the same. In fact, at one point I spent a lot of time on projects where guys who had been programming in COBOL for decades were learning to use OOP and one of the most common mistakes was to confuse a class hierarchy with a part of hierarchy. Of course some new OWL users do that as well but if they define their axioms correctly the OWL reasoner will realize they are using the subclass relation incorrectly. 

 I'd be pretty surprised to find a C++ system that was a good match with the 'real' world. 

Again, depends how you define a model. I can tell you that every OOP methodology and every OOP project I supervised or took part in started with people getting in a room and writing on the white board the basic business objects and the use cases they took part in. So things like Account, Order,  Bill of Materials, etc. I agree that you often end up with a lot of detail in a C++ (or for that matter Smalltalk, Java, CLOS, or Python) OOP model that involves technical objects as well as business objects. Things like queues, load balancers, schedulers, etc. That's what I mean when I  say that IMO an OOP model usually starts as a model at what software engineers call the design level. Where you are integrating business objects and objects that are part of the technical design. Whereas an OWL model stays at the analysis level. 

One last point: just as with the concept "species" in biology all these definitions are useful fictions. It isn't like there is a clear boundary between what is design and what is analysis that everyone would agree on. 

Michael

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAMWD8MrCs-i1UyqC9WTQCjLXFx0G2PmfiRSaXFthd4Ok5AXWuw%40mail.gmail.com.

[Chris Partridge]

Hi Michael,

I'm not sure where to start.

WRT:  When you say Dog is a subclass of Mammal that has meaning in OWL based on logic and set theory (i.e., that the set of dogs is a subset of the set of mammals).
As noted by Pat here (https://www.semanticuniverse.com/blogs-intensional-and-extensional-owl/) "So in an extensional logic (like OWL-DL), if you know that classes A and B have exactly the same members, then you know that A is the same class as B. In an intensional logic (like OWL-Full or RDFS) you don’t know this, and it might not be true. In those logiocs, two classes can have exactly the same members and still be different classes."

So either dogs and mammals are not sets or OWL-FULL is not formal - according to you.

Pellet is written in Java. So does this mean when you use it that the OWL file is no longer formal, or that Java is formal and C++ isn't. Or that a reasoner cannot be written in C++?

So something like C++ (or Java) needs to be in the stack.

WRT: They usually mean and for sure I always mean that there is some well defined semantics. When you say Dog is a subclass of Mammal that has meaning in OWL based on logic and set theory (i.e., that the set of dogs is a subset of the set of mammals).

But, I assume not that the well-defined semantics gives you the intended interpreation of dogs and mammals. They could just as easily have a good model that has them referring to sheep and goats.

If your well-defined semantics doesn't support higher-order types, does this mean they aren't real. Then what happens when you work with a formal theory that does. Do they suddenly become real? Ditto modality.

WRT: I can tell you that every OOP methodology and every OOP project I supervised or took part in started with people getting in a room and writing on the white board the basic business objects and the use cases they took part in. So things like Account, Order,  Bill of Materials, etc.

For me the issue here is: what is the intended interpretation. If you think they are talking about the same thing then all the systems you look at will have the same data structures for these. And mapping data for these business objects from one system to another will be trivial and simple. In my experience neither of these are true.

For me, the point of the ontology is to get a good enough picture of the world so that the data becomes interoperable. 

I could go on. But hopefully you see where this is going.

Of course, if you and others want to call every OWL file an ontology, then you have every right to do so. But just don't expect it to then magically refer to the intended interpretation or be trivially interoperable.

I prefer to work with a more traditional philosophical notion of ontology that offers some traction on semantic interoperability.

Hope that helps.

Chris

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CALGFike4GjZee2%2BjUjjO8g0-s0snCUthL27fwLCkUaBRbxYF8w%40mail.gmail.com.

[Ken Baclawski]

Is it necessarily wrong for Mammal to be a subclass of Dog?  You might look at my article about issues like this:

K. Baclawski and B. Indurkhya. The notion of inheritance in object-oriented programming. Comm. ACM 37(9):118-119. (September, 1994) https://bit.ly/3rw9tAe

While my article was published nearly 30 years ago, the points it makes are IMHO still relevant.

Ken

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CALGFike4GjZee2%2BjUjjO8g0-s0snCUthL27fwLCkUaBRbxYF8w%40mail.gmail.com.

[Mike Peters]

This is a great discussion that gets to the heart of things IMO.

Alex S., thanks for using Claude to generate a list of useful readings; well, it's a good start, anyway. I read the book (s) by David C. Hay - all his books are excellent.

Chris P., you make an excellent point about data models being able to describe the world, so in that sense, they are ontological.

The SaaS project I'm working on uses relational data models that include these key entities or "business objects": account, order, party, biological cell, atom, protein, ecosystem, planet, sub-atomic, etc. Curiously, some of these are also included in BFO-COB. There were others mentioned in Chris M's GitHub Project document.

They can and need to be defined in something like OWL and used in databases and code to do anything useful.

I think the key to the puzzle is that the key "business objects" must exist in the real world, be defined in a TLO like BFO-COB, and represent some key intersection of material reality, e.g., an Atom, A Protein, a Biological Cell, or an Organism, etc. Then everything else can be hung off these key definitions via Formal Ontologies, taxonomies, etc.

Mike Peters
Ajabbi
https://blog.ajabbi.com

New Zealand

[Ravi Sharma]

Substantial discussion, I am learning, appreciate.

Would have liked John's opinion also!

While I have to assimilate info in this thread a lot, I am thinking of "fit for the purpose" models!

This leads to:

• Different models of Ontology (definition has been discussed, also architecture)?
• Which model is fit for what purpose?
• Do not all models of "real things and concepts" have Knowledge Representation in common?
• If so, how does TLO and others derive from this or an accepted model of KR?
• Programming level models (based on use cases and requirements), in my opinion, have limitations based on both the programming language related limitations and also domains-and-knowledge connection related limitations modeled!

Hope I have not derailed the focus, but just an input that programming environment itself has inherent limits, hence reality or models are further constrained. Not to reduce the importance of Programming which is enabling the IT world running Banks, supply chain and LHCs of today!

😀

Thanks.

Ravi

(Dr. Ravi Sharma, Ph.D. USA)

NASA Apollo Achievement Award

Former Scientific Secretary iSRO HQ

Ontolog Board of Trustees

Particle and Space Physics

Senior Enterprise Architect

SAE Fuel Cell Standards Member

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/274b4a00-115d-404b-bc1e-5ecabd00717en%40googlegroups.com.

[Burkett, William [USA]]

Wrt to Chris P’s point that Mike P repeats:  “So if the data model is picturing the real world, it is in some sense ontological”

 

I agree with that.  However, IMHO, If the model is a model of the “real world” then is it not a “data” model.  I would rather call things what they are: if it’s a “data model” you’re modelling data structures.  If you’re modelling the “real world” then it is not a “data” model – because it’s not a model of data.  (The term “conceptual data model” should be banished from everyone’s vocabulary!)  If you are modelling the real world, it could be called an “ontology”. 

 

Mike: I’ve read David Hay’s books and they’re okay.  He tends to get caught up in malady I call “meta-itis” – modeling the model in the model.  (Mathew West, RIP, did this a lot too.) 

 

So many, if not all, data modelling books talk about modelling “business objects” – “things important to the business.”  That is - again IMHO and with all due respect – so lame!  All of these books miss the important dimension of the why certain entities that are perceived to exist in the real work (i.e., “business objects”) are or should be represented in the model.  It is not just that the “account, order, party, biological cell, atom, protein, ecosystem, planet” are there and are recognized and are therefore represented in the model.  There is a business (usually process) requirement to capture information about these things.  If you don’t have this connection, why is it in your model?  That’s an issue I have with a lot of ontology work in the abstract – the ontologies just “represent objects in the real world” with no reason for why they’re in the model.

 

Bill Burkett

 

 

From: ontolo...@googlegroups.com <ontolo...@googlegroups.com> On Behalf Of Mike Peters
Sent: Friday, April 26, 2024 1:21 PM
To: ontolog-forum <ontolo...@googlegroups.com>
Subject: [External] Re: [ontolog-forum] top layer ontologies, domain layer, application layer and data models?

 

This is a great discussion that gets to the heart of things IMO. Alex S. , thanks for using Claude to generate a list of useful readings; well, it's a good start, anyway. I read the book (s) by David C. Hay - all his books are excellent. Chris

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/274b4a00-115d-404b-bc1e-5ecabd00717en%40googlegroups.com.

[Ravi Sharma]

John Sowa said at one time that logical data models contain ontology information because they relate entities?

Thanks.

Ravi

(Dr. Ravi Sharma, Ph.D. USA)

NASA Apollo Achievement Award

Former Scientific Secretary iSRO HQ

Ontolog Board of Trustees

Particle and Space Physics

Senior Enterprise Architect

SAE Fuel Cell Standards Member

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/MW4PR06MB8219A55D45FBCAF0DAAE338591162%40MW4PR06MB8219.namprd06.prod.outlook.com.

[Burkett, William [USA]]

> John Sowa said at one time that logical data models contain ontology information because they relate entities?

 

I don’t disagree with that, Dr Sharma, but begs the question of what those terms mean.   (It’s always back to “depends on what you mean by x”.)

 

B.

 

 

 

 

From: ontolo...@googlegroups.com <ontolo...@googlegroups.com> On Behalf Of Ravi Sharma
Sent: Friday, April 26, 2024 9:08 PM
To: ontolo...@googlegroups.com
Subject: Re: [External] Re: [ontolog-forum] top layer ontologies, domain layer, application layer and data models?

 

John Sowa said at one time that logical data models contain ontology information because they relate entities? Thanks. Ravi (Dr. Ravi Sharma, Ph. D. USA) NASA Apollo Achievement Award Former Scientific Secretary iSRO HQ Ontolog Board of TrusteesParticle

[John F Sowa]

Bill and Ravi,

I said many things about many things at many times.

Basic point:  It's impossible for any program on a digital computer to be informal.  Therefore, there must be a formal ontology of some kind buried in it.  But that does not imply that it's accurate or good for its purpose.

Engineers have an excellent saying:  "All models are wrong, but some are useful."  That can be generalized: "All ontologies are wrong, but some are useful."

Since I worked at IBM for 30 years, I understand the IBM phrase "functionally stabilized" very well.   It means that the product is obsolete (or at lease obsolescent) and no further updates or modifications will be made.  IBM would continue of sell it to anyone who depends on that particular product type.  And IBM would continue to fix bugs, but they would urge customers to upgrade to newer technology.

I won't say anything about anybody's pet ontologies, but I believe that a lot of them are or should be "functionally stabilized".

John

[Ravi Sharma]

John

I have also done Enterprise level Data Modeling and I say and hope you do not disagree that

logical data models contain ontology information because they relate entities.

Regards.

Thanks.

Ravi

(Dr. Ravi Sharma, Ph.D. USA)

NASA Apollo Achievement Award

Former Scientific Secretary iSRO HQ

Ontolog Board of Trustees

Particle and Space Physics

Senior Enterprise Architect

SAE Fuel Cell Standards Member

--

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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/0aa79ffb8fef472e8ad9af3f12887587%40bestweb.net.

[Alex Shkotin]

Michael,

WRT:Of course some new OWL users do that as well but if they define their axioms correctly the OWL reasoner will realize they are using the subclass relation incorrectly.

Let me point out the importance of scientific definitions to be used in our formal ontologies, i.e. science and theory first and formalization second.

In this particular case we are in Biology where the terms "mammal" and "dog" have definitions. And the challenge for OWL2 is if we can formalize these definitions in OWL2 as its expression power is not very good for real world modeling. But if these two definitions are formalized we are expecting to get a proof that dog implies mammal. If all terms have definitions then class hierarchy is just derived automatically. 

We have shown this for definitions of igneous rocks here.

And formalization is a little bit of a restricted area: we took definitions of experts to formalize, but it's hard to resist giving your own definition. Therefore, each definition must contain a reference to the original source: a book or person - the author of the definition. 

When we get different definitions inside the same theory, the task arises to prove equivalence or to present an entity which splits these definitions to check for authors their correctness.

Consider the same Biology area to define what it means to be a living being. And maybe one definition says bacteriophage is living but other says not.

From this scientific point of view C++ program is such a very special entity to prove its property of any kind. For example to satisfy specifications 😀

And for example, there is no general algorithm for proving a halting, so for each class of programs we will have to prove it separately.

WRT:One last point: just as with the concept "species" in biology all these definitions are useful fictions. It isn't like there is a clear boundary between what is design and what is analysis that everyone would agree on.

I don't think any biologist would agree with your "in biology all these definitions are useful fictions". Is that you can define better or from your point of view in biology can not be nonfiction definitions?

And we did not usually need "everyone would agree on", but inside of a working team every one must. The presentation of knowledge in the form of theory is absolutely necessary for the correct use of terms when they go beyond the naming of specific things and actions. Currently, precise definitions are what AI lacks. This is what they strive to replace with billions of examples.

Alex

пт, 26 апр. 2024 г. в 17:31, Michael DeBellis <mdebe...@gmail.com>:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CALGFike4GjZee2%2BjUjjO8g0-s0snCUthL27fwLCkUaBRbxYF8w%40mail.gmail.com.

[Alex Shkotin]

In addition, an inaccurate definition is better than its absence, since it can be clarified. I will also note that precise definitions are very necessary for robots, since it is better not to let robots into areas where we do not have them. Or you will have to let them in “at your own peril and risk.” What is happening now on a global scale.

вс, 28 апр. 2024 г. в 12:29, Alex Shkotin <alex.s...@gmail.com>:

[Alex Shkotin]

Mike,

WRT:The SaaS project I'm working on uses relational data models that include these key entities or "business objects": account, order, party, biological cell, atom, protein, ecosystem, planet, sub-atomic, etc

It's better to say "objects of interest". And in any way is there a chance to point to a science or technologies from which these terms are? Just look in what form we have theoretical knowledge there.

Like this: account from GAAP, party - ???, biological cell, protein - biology, planet - astronomy; atom, subatomic - atomic physics, ecosystem - ???

What is an idea to keep all of these terms in one place? Encyclopedia?

What kind of knowledge processing do you keep in mind?

Maybe you should cooperate with Azamat Abdoullaev ontop...@gmail.com?

If something exists there always exists somebody who scientifically studies this kind of thing. i.e. she systematizes experimental knowledge about them 🏋️

Alex

пт, 26 апр. 2024 г. в 22:21, Mike Peters <mi...@redworks.co.nz>:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/274b4a00-115d-404b-bc1e-5ecabd00717en%40googlegroups.com.

[Anatoly Levenchuk]

Hi, Alex,
1. Most system engineers do not know about the existence of TLO (upper ontology) and the difference of TLO and middle ontology and domain ontology. Moreover, they have significant problems in thinking about types, although they are better at this than “people from the street.” System engineers, of course, use ontics - middle ontology, the types of which are in no way connected through any TLO.

2. But those who manage system configuration (they are usual systems engineers) and are engaged in system description (multi-domain multi-model usually in some PLM database) are forced to think about ontology by their profession. There appears the problem of “working English” when you want to communicate with a Spaniard, a Chinese and a Malaysian, same is true for engineering CAD/CAM/CAE software that speaks it own language usually. Another thing is that at this moment all three are claiming “working English”, although the meeting is being held in English, which is already have all the words necessary for translation. In systems engineering, PLM people connects all sorts of CAD/CAM/CAE and other modeling tools to PLM databases (such as Siemens Teamcenter or Dassault Systemes 3DXperience). They know about TLO, but not knowing this name, they know it as a «neutral data model». E.g. such a model is ISO 15926, where TLO is in ISO 15926-2 part of the multi-part standard. This approach is failed, because ISO 15926-2 treated not as a «working English» but as a «very exotic Esperanto». Therefore TLO exists as a native data model to PLM systems, each vendor has its own variant that is far from academic notion of TLO.

3. The most advanced systems engineers, of course, think about TLO, and there they are already moving towards constructivism (as in constructive mathematics: not objects and relationships, but objects and construction operations). I have a review on this topic: https://ailev.livejournal.com/1653296.html (it in Russian, you can use Google Translate or similar service) - Mathew West helped organize all this. When John Sowa said that he was interested in what Mathew thought about this topic, then you can just go to the materials there and find out what he thinks. 15 years ago it was TLO HQDM (similar to ISO 15926-2). But a couple years ago Mathew sent me a link to these materials when I was interested in system approach applicable to descriptions, not only to 4D objects. English speakers can go strait to https://gateway.newton.ac.uk/event/tgmw80/programme and bypass may text in Russian. Matthew West said that this should not be seen as separate disparate talks, but starting with his own talk as six parts of one story. It's April 2021, so something else has probably been done there over the past year and a half.

4. We have experience in training system engineers. And our teachers come to the conclusion that it is useless to teach systems engineers if they are not first given knowledge of ontology and be fluent with types and relationships. Here are two recent reports on this topic (last weekend conference, sorry it in Russian)
-- Alexander Luchkov, “Teaching architectural practices at a university.”, https://www.youtube.com/watch?v=TdD7mKMEngA&t=8964s

--  Kirill Gaydamaka, “Experience in teaching system modeling at a university.”, https://www.youtube.com/watch?v=TdD7mKMEngA&t=11775s

In our System Management School (Aisystant) we teach students about TLOs, but I have already said in my Ontolog Forum talk (https://www.youtube.com/watch?v=G4S-Zrc5qUk&t=2s):
-- First we teach the types of objects and relationships: that they are important
-- Then we give the main ideas of 4D TLO ISO 15926 and point to the BORO book by Chris Partridge.
-- We teach to use multi-level multi-domain ontology in engineering projects: many levels of hierarchy in relation to classification at once. Let's say, here is just one of the courses where TLO (under two names: upper ontology and meta-meta-model, while we say that they are usually described in text, and not in formal language, and this is important!) actually learns how be used by engineers.

Best regards,
Anatoly

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROT%2Bqyandq3wbJh7HN2EHHJU-XQNeVzmYJ_qYKa2Fann2A%40mail.gmail.com.

[Alex Shkotin]

Hi, Anatoly!

Thank you very much for your detailed answer! It seems to me that ontologists should know the opinion of systems engineers.

Alex

вс, 28 апр. 2024 г. в 20:03, Anatoly Levenchuk <ai...@asmp.msk.su>:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/000901da998e%2407f47d00%2417dd7700%24%40asmp.msk.su.

[Mike Peters]

Hi Anatoly

I too discovered those fascinating sessions at Newton, when Matthew put me onto them. I was very impressed with the work both he and Chris Partridge did, and as result, the architecture of Pipi 9 was heavily modified. Such a pity Matthew has passed away, I had a lot of questions to ask him.

I read your fascinating Live Journal articles. Google Translate does a surprisingly good job. My SAAS project Pipi is a complex adaptive system in its own right designed to host complex adaptive systems as a SaaS. I have found the Quanta Magazine and Santa Fe Institute full of a lot of thinking about Emergent properties and Systems thinking..

Alex: I actually call these "Pipi Privatives" (a film animation industry term). Yes and they are intended to be directly linked to scientific "laws".

Most of them came from this man George F R Ellis

https://www.blog.ajabbi.com/2023/12/george-ellis-and-emergence.html

and a smaller number confirmed by BFO-COB

You can see my draft list here
https://docs.google.com/spreadsheets/d/1sP97vXlPHX2l7t-qX_rgq6N47W3pcF1SDqPUOGItmRc/edit?usp=sharing

Mike Peters
https://www.blog.ajabbi.com
New Zealand

[Chris Partridge]

Hi,

Am travelling, so responses will be delayed.

WRT: However, IMHO, If the model is a model of the “real world” then is it not a “data” model.  I would rather call things what they are: if it’s a “data model” you’re modelling data structures.  If you’re modelling the “real world” then it is not a “data” model – because it’s not a model of data.  (The term “conceptual data model” should be banished from everyone’s vocabulary!)  If you are modelling the real world, it could be called an “ontology”. 

I realise this (or something close) is the orthodoxy - and I think I kinda assumed this in one of my early papers. The idea being that one starts with a conceptual model of the domain and this mrphs at some stage into a model of the data - and then back into a system that again refers to the domain.

However, for quite a while now I've struggled with this story - as it doesn't seem to fit well with the facts (the facts of our (BORO) practices).

There are (at least) two senses of model - one used in the quote where the model is a specification of something and refers to it, the other where it is a version of something. A scale model would be an example of the second. I'm coimg round to the idea that we are actually doing somethhing close to the second.

In our legacy, reengineering projects we import the data (and schema) from the source system into our working area. we can (and do) map it to various formats including UML. We may work directly on the UML model, when we refactor things. As part of our bCLEARer process, this dataset undergoes a series of transformations eventually ending up as the finished product. So there is no point were it shifts from a conceptual model to a data model to a system. It seems to us that it is a system through the whole life cycle. Nevertheless,we do the kinds of things that people do with conceptual and data models. 

I wonder whether the orthodoxy reflects historic constraints. I remember building models in the late 80s using graphics with a team of support staff - and being so happy to move to a repository based tool that could just about support a model of the entity types. I think technology constraints in those days meant we could not load all the data (and schema). And we just work the same way now. To see this, ask yourself whether you oad all the data into you UML or entity models, when you have it. I don't see people doing this - if they do, it would be good to know.

Of course, if people want to adopt the orthody as their practice, I'm not saying it is impossible - just that it seems a little clunky.

Best,

Chris

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/68d49c75-dc01-436b-8571-17d8c155eaf2n%40googlegroups.com.

[Alex Shkotin]

Mike,

I should get Ecology, but my other questions are not answered: "party - ???

What is an idea to keep all of these terms in one place? Encyclopedia?

What kind of knowledge processing do you keep in mind?"

Alex

пн, 29 апр. 2024 г. в 12:14, Mike Peters <mi...@redworks.co.nz>:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/68d49c75-dc01-436b-8571-17d8c155eaf2n%40googlegroups.com.

[John F Sowa]

I agree with Chris on the issues he discusses below.

But I very, very strongly agree with the engineering motto:  "All models are wrong, but some are useful."

I would generalize that:  "Every ontology is a model of some aspect of the universe.  Therefore, all ontologies are wrong, but some are useful."

This point is especially important for detailed ontologies of "everything"..  Physicists know that every theory of physics is wrong in its most fundamental underlying principles.  But physicists also know that many, many theories are highly accurate on important aspects that have been studied in detail.  They also know that whenever they explore new areas at finer levels of detail, they run into amazing new principles that they had not suspected or even dreamed about.

ISO has a fundamental guideline that takes these issues into account:   Never consider any standard to be permanent.  Every five years, every ISO standard must be reconsidered and either revised, reaffirmed, or abandoned.  The same principles must be applied to ontologies.

John

 

---

From: "Chris Partridge" <partri...@gmail.com>

[Burkett, William [USA]]

Chris –

 

> There are (at least) two senses of model - …

 

Models have purposes – models are developed for a reason(s).  These purposes can be divided into two major kinds: (1) the domain task at hand – the application objective; and (2) the model intent or function.  The first is easy to see and understand.  The second is what you’ve seen with your “two senses”.  Model Intent is the general, domain-independent function of the model and I think there are six senses:

 

• Inquiry (Inquisitive);
• Analysis (Descriptive);
• Prediction (Predictive);
• Conveyance of Knowledge (Informative);
• Specification and Control (Prescriptive);
• Representation and Display (Representative).

 

These purposes overlap and Model Intent may shift in emphasis over time.  In fact, most (data) models go through all of them except Predictive.  The two senses you’ve identified are Specification/Control and Representation/Display (e.g., “scale model”)

 

I’ve been a systems engineer in the data modelling world for my whole career and I’m sensitive to what models can and can’t do.  John Sowa recently quoted George Box (a statistician): “All models are wrong.  Some are useful.”  There’s another one I like and quote often from Alfred Korzybski that conveys a similar message: “The map is not the territory.”

 

Bill

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAMWD8MqdjTd4-OY9pXPipjM1pbVMH_G3RRSUzDapXiQTFKAGqQ%40mail.gmail.com.

[Chris Partridge]

Hi Bill,

I'm not quite sure I see how these neat analysis/categories capture exactly what I was suggesting.

Let's assume my purpose is to build a system. I think I'm suggesting that I can (to use your terms) prescribe what system I want to build (control what is built) by representing what I want built. Or, at least, use the representation to engage in discussion about what I want to build - so engage in both categories - they overlap as you mention. I'm probably also doing other things on your list. 

What bothers me is how we rationalise the task we do when we engage in doing what is sometimes called data modelling. Should we see our data model itself as a model of the domain, and our building it is modelling the data. Or are we building a model of the data in the system, that then does the referring? It seems to me that the first of these is a cleaner way to think about what we are doing. I don't quite see how these two different characterisations of the same process fall neatly (or otherwise) under different categories - they both seem to fall under much the same ones. I agree however that one can use these categories to pick out different aspects of our description of what is going on - maybe that was your point.

I'm only being picky here because I think one way things move forward is by looking carefully at our assumptions, and revising them when things don't add up.

Chris

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/MW4PR06MB82191AF7AEA930C700DC9202911B2%40MW4PR06MB8219.namprd06.prod.outlook.com.

[John F Sowa]

Michael and Alex,

I agree that the terminology defined below is useful.

But the principle "All models are wrong, but some are useful" must be kept in mind.

However, older models aren't necessarily bad, and newer models aren't necessarily better.

All major breakthroughs occur when somebody finds a novel way of rejecting some widely accepted model(s) -- new or old.

John

 

---

From: "Michael DeBellis" <mdebe...@gmail.com>
Sent: 4/25/24 10:52 AM
To: ontolo...@googlegroups.com
Subject: Re: [ontolog-forum] top layer ontologies, domain layer, application layer and data models?

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CALGFikfykMw-WQ0yVtwt2pzf3stPhb3ajqaEu8VOVnVJjEdK4w%40mail.gmail.com.

[Burkett, William [USA]]

Thanks, Chris – I get your point and agree that the “first of these is a cleaner way to think about what we are doing”.  I see the two steps as distinct and separate, however – entirely separate.  They have different, though related, purposes.  I acknowledge that the term “data modelling” often refers to tasks focused on creating a “model of a domain” and I think that that is a misnomer.  Understanding and representing (in a model) the entities I see in a domain (that are important toward a purpose/goal of mine) is a task and an end in itself.  Look at it as a “requirements statement” – a “Descriptive” or “Informative” model.  The /data/ model is a separate model.  I think you’d agree that many different and distinct (“Prescriptive”) models of data (actual data structures) can encode/represent/provide the same information about the entities modelled in the informative/descriptive model (with possibly differing degrees of fidelity).  The prescriptive data model is a means; the informative/descriptive domain model is an end.  Presumably, the informative/descriptive model will be less likely to change over time while data structures may change regularly.  See Ron Ross’s “Business Knowledge Blueprints” for a published version of this idea.

 

Am I demonstrating an understanding of your point?  Or am I still missing something?

 

Bill

 

 

From: ontolo...@googlegroups.com <ontolo...@googlegroups.com> On Behalf Of Chris Partridge

Sent: Monday, April 29, 2024 8:41 AM
To: ontolo...@googlegroups.com

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAMWD8MqNVy-WQKKa8M%2By7U7OvFHKmDcFiFr0ZoJKMjCo0batyw%40mail.gmail.com.

[paul.w...@btinternet.com]

On a point of information, and to give credit where it is due, I believe the quote “All models are wrong, but some are useful”, is generally attributed to the statistician George Box.

 

Although, of course, the quote can apply to models in any field.

 

Regards,

 

Paul Warren

 

From: ontolo...@googlegroups.com <ontolo...@googlegroups.com> On Behalf Of John F Sowa
Sent: Monday, April 29, 2024 3:17 PM
To: ontolo...@googlegroups.com
Subject: Re: [ontolog-forum] top layer ontologies, domain layer, application layer and data models?

 

I agree with Chris on the issues he discusses below.

--

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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/1e10db53dbfd49bf96009e63de564b2d%40bestweb.net.

[Mike Peters]

Hi all

JS: "But I very, very strongly agree with the engineering motto:  "All models are wrong, but some are useful."

I would generalize that:  "Every ontology is a model of some aspect of the universe.  Therefore, all ontologies are wrong, but some are useful."

This point is especially important for detailed ontologies of "everything"..  Physicists know that every theory of physics is wrong in its most fundamental underlying principles.  But physicists also know that many, many theories are highly accurate on important aspects that have been studied in detail.  They also know that whenever they explore new areas at finer levels of detail, they run into amazing new principles that they had not suspected or even dreamed about.

ISO has a fundamental guideline that takes these issues into account:   Never consider any standard to be permanent.  Every five years, every ISO standard must be reconsidered and either revised, reaffirmed, or abandoned.  The same principles must be applied to ontologies."

MP: That's spot on John. I'm going to use that somewhere in the Pipi Docs.

AS: "I should get Ecology, but my other questions are not answered: "party - ???

What is an idea to keep all of these terms in one place? Encyclopedia?
What kind of knowledge processing do you keep in mind?"

MP:  Alex, Party is a data model entity I got from David C. Hay and Len Silverston - it means something that is party to, and often used by data modelers. Can be used for people, organisations and user agents which are sub types. It makes a great "who" primative.

Ecosystem is one of the primative's.

At the moment, I'm using a simple blog to share with people. However behind the scenes, I'm getting Pipi to use templates to render a public facing website that will openly document all assumptions, parameters, primatives, complex systems, relationships to ontologies (eg Snowmed), API's (eg (Protein Data Bank) etc. Most properties are fully emergent and can also be documented in HTML. The code will not be publicly documented. Maybe sometime this year. :) :)

After that get a demo version on-line using the free cloud tier infrastructure so people can try it out (will require booking a time slot). Happy to give a live demo to this forum sometime, when you can all use it.

Mike Peters
https://www.blog.ajabbi.com
New Zealand

[Chris Partridge]

Hi Bill,

I think we are close.

You are right that we will alter (data?) models to improve say performance.So, there are the before and after models - driven by different purposes. And one only makes the performance change when one is concerned about performance of the system. So talk about purpose has a purpose :)

I want to say both the models are models of the domain, just in the second some compromises have been made in the directness of reference. So they are not materially different in what kinds of things they are at some level of abstraction. Though they are different and the differences are driven by 'different' configurations of purposes - and the similarities by similar urposes..

One thing that this approach seems to encourage is dropping the different way schema and data are treated in the development process. For example, it seems to make it more likely for developers to include the data in their 'conceptual' schema.

Best,

Chris

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/MW4PR06MB82195FA886FA22F2D063EC51911B2%40MW4PR06MB8219.namprd06.prod.outlook.com.

[Ravi Sharma]

Let us keep converging, but also leave room for future inputs.

For example today's particle physics encompasses not only Math, Statistics, Quantum information and fields but also uses ML, AI inputs and all of the above to identify new Beyond Standard Model Physics.

Big Bang and Standard Models are therefore becoming obsolete in the new physics that is emerging!

Regards.

Thanks.

Ravi

(Dr. Ravi Sharma, Ph.D. USA)

NASA Apollo Achievement Award

Former Scientific Secretary iSRO HQ

Ontolog Board of Trustees

Particle and Space Physics

Senior Enterprise Architect

SAE Fuel Cell Standards Member

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAMWD8MoQH%3DOT4cZ%2BqSGL9ahONebpdLuY2rh%3DeyR6nHonm29h1A%40mail.gmail.com.

[Alex Shkotin]

John,

We use theories to solve tasks. For example, in math we have the theory of undirected graphs. We use it to solve tasks to get one or another property of one or another graph, see Specific tasks of Ugraphia on a specific structure (in development, open for comments) The Russian version is here.

The first rule of thinking for physicists creating a model for a real system is What can we neglect? What of what we know about the physical system is not essential for solving our task?

The joke "All models are wrong, but some are useful." is from people who know how to create task oriented models for real systems.

How can we correctly pose a problem without having a theory? Every theory is about some object and processes. What kind of tasks and problems about objects and processes can we solve using this theory? 

That is the question.

ABSTRACTION is a way of thinking we should pose before any abduction, deduction, induction.

Alex

пн, 29 апр. 2024 г. в 17:17, 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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/1e10db53dbfd49bf96009e63de564b2d%40bestweb.net.

[Alex Shkotin]

Mike,

The idea of your Pipi system looks like all kinds of systems crawler and catalog.

Alex 

пн, 29 апр. 2024 г. в 22:33, Mike Peters <mi...@redworks.co.nz>:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/842324eb-54bb-4cc0-b344-333b0cc72ffen%40googlegroups.com.

[Mike Peters]

AS:  The idea of your Pipi system looks like all kinds of systems crawler and catalog.

MP: It was a description of the planned technical documentation system to be provided for developers and computer scientists. I intend it to be very open about the parameters, primitives', scientific laws, entities, domains, ontologies, API's and rules in place, etc. A lot could go on GitHub. I built and tested this thing but didn't do any documentation, not even version control, lots of drawings though, so relying on Pipi to self document itself. It is very fast, can render 1200 Wikipedia sized web pages in 5 seconds using a laptop. So I'm the hold-up figuring out what would be useful for people. Going for a cross linked "Javadoc's" + Wikipedia + Adobe Help look and format. Will need to autogenerate lots of embedded data model diagrams and I'm not sure how to do that.

Given that Pipi can evolve by learning from the users, subject to the rules and parameters previously mentioned, and also write its own code, the internal workings will not be documented. The ML involved it is largely a black box. No LLM anywhere. 

Best to wait till this CAS SaaS platform is working and then you can try it out and test it. 

Mike Peters
https://www.blog.ajabbi.com
New Zealand

[Alex Shkotin]

Mike,

You know how I like to ask LLMs :-)

topic-1

AS: What kind of technical documentation systems do we have?

C3S: I don't actually have information about what specific documentation systems your organization uses. As an AI assistant without access to details about your company's internal systems and processes, I can only provide a general overview of common technical documentation systems and practices.

Some typical systems and tools used for technical documentation include:

- Documentation portals/wikis (e.g. Confluence, MediaWiki)
- Static site generators (e.g. Sphinx, Jekyll, Hugo)
- Markup languages (e.g. Markdown, reStructuredText, AsciiDoc)
- Component content management systems (e.g. DITA OT, easyDITA)
- API documentation tools (e.g. Swagger UI, Postman)
- Diagramming tools (e.g. Lucidchart, Visio, PlantUML)
- Version control systems (e.g. Git) for documentation source files

Modern technical writing teams often use a docs-as-code approach, treating documentation like software with version control, review processes, automated builds and publishing. Popular workflows integrate tools from the list above.

If you can provide more specifics about your organization and documentation needs, I'd be happy to give a more targeted response. Otherwise, I've given a general overview of common documentation systems and practices.

Meta.AI: There are several types of technical documentation systems, including:

1. Help Authoring Tools (HATs)

• Examples: MadCap Flare, Adobe RoboHelp, Confluence
• Features: Single-source authoring, content management, output to multiple formats (HTML, PDF, etc.)

2. Wiki Systems

• Examples: MediaWiki, Confluence, SharePoint
• Features: Collaborative editing, version control, linking, and categorization

3. Documentation Generators

• Examples: Doxygen, Javadoc, Pydoc
• Features: Automatic generation of documentation from code comments and structure

4. Content Management Systems (CMS)

• Examples: WordPress, Drupal, Joomla
• Features: Web-based content creation, management, and delivery

5. Version Control Systems (VCS)

• Examples: Git, SVN, Mercurial
• Features: Source control, versioning, and collaboration

6. Single-Source Documentation

• Examples: LaTeX, Markdown, reStructuredText
• Features: Write once, publish to multiple formats (PDF, HTML, etc.)

These systems help create, manage, and publish technical documentation, such as user manuals, guides, and reference materials, making it easier for users to understand and work with software, hardware, and other technical products.

AS2MP: Do you think we can integrate all of them in one?

topic-2

AS:Will need to auto generate lots of embedded data model diagrams and I'm not sure how to do that.

C3S: Automatically generating data model diagrams from code or data definitions is a common requirement for technical documentation. There are several tools and approaches that can help with this:

1. **Code/Schema-Driven Diagramming**
    - Tools like SchemaSpy, SchemaCrawler, or Schemaverse can reverse engineer database schemas and generate entity-relationship diagrams.
    - Libraries like DataMapper, ERAlchemy, or Django Extensions support model graphing for code-first ORMs.

2. **Documentation Generators**
    - Sphinx can integrate extensions like sphinx.ext.graphviz or sphinxcontrib-er2dtoo to pull models from docstrings and generate diagrams.
    - JSDoc can use plugins like jsdoc-diagram to visualize class diagrams from JavaScript code.

3. **General Diagramming Tools**
    - Tools like Lucidchart, PlantUML, or Python GraphViz can import data definitions and models via scripts to programmatically generate diagrams.
    - Some IDEs like IntelliJ IDEA have diagramming plugins that can reverse engineer models.

4. **Documentation-as-Code**
    - Treating documentation as code allows integrating documentation updates alongside data model changes.
    - CI/CD pipelines can run documentation builds that regenerate diagrams on each release.

The best approach depends on your technology stack, existing tooling, and how tightly you want to couple documentation with the data models themselves. Code-driven approaches keep diagrams more up-to-date but require integration. Documentation generators provide more flexibility in styling.

I'd recommend evaluating your current documentation workflow and trying out some proof-of-concepts with the relevant tools for your environment. Having an automated way to generate up-to-date data model visualizations can greatly improve technical documentation quality.

Meta.AI: Automating the generation of embedded data model diagrams can be a challenging task, but there are several options to consider:

1. Graphviz

• A popular open-source tool for visualizing graphs and diagrams
• Supports various input formats (e.g., DOT, SVG)
• Can be integrated with scripting languages like Python or Ruby

2. PlantUML

• A tool for creating UML diagrams from plain text descriptions
• Supports various output formats (e.g., SVG, PNG, PDF)
• Can be integrated with documentation generators like Doxygen or Javadoc

3. yFiles

• A commercial library for diagramming and graph visualization
• Supports various programming languages (e.g., Java, C#, JavaScript)
• Offers advanced features like automatic layout and styling

4. Mermaid

• A markup language for generating diagrams and flowcharts
• Supports various output formats (e.g., SVG, PNG, PDF)
• Can be integrated with documentation tools like Markdown or AsciiDoc

5. SQL Diagram Generators

• Tools like SchemaCrawler, DbSchema, or SQL Server Management Studio
• Can generate diagrams from database schema information

To get started, consider the following steps:

1. Define your data model using a formal notation (e.g., UML, ERD)
2. Choose a diagramming tool or library that fits your needs
3. Write a script or program to generate the diagrams automatically
4. Integrate the generated diagrams into your documentation or reporting workflow

Remember to explore each option's documentation and examples to find the best fit for your specific use case.

Alex

вт, 30 апр. 2024 г. в 13:30, Mike Peters <mi...@redworks.co.nz>:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/b7517129-2f91-44f2-8fa8-aec4cfc08834n%40googlegroups.com.

[Mike Peters]

AS:  Do you think we can integrate all of them in one?

MP: Because Pipi is a CAS, all the technical documentation is generated by Pipi from its own dynamic and evolving config using its internal CMS static site generator. Editing the documentation directly changes the config. So unfortunately existing tools will not work in this situation. However the CMS can now create wiki, docs and help files that xref each other in HTML, with PDF later.

AS: Automatically generating data model diagrams from code or data definitions ...

MP: Alex, thank you for using LLM to dig up this information about data model diagram generation. 

Everyone: top layer ontologies, domain layer, application layer and data models

MP: I think it might have been Lee Smolin (Perimeter Institute) who wrote
"Nature is a book to be read, not written"

IMHO,  all these science models, mathematics, philosophy, top layer ontologies, domain layer, application layer, taxonomies, schema, and data models, and drawings are just a human representation of reality, not reality itself. Useful to a point. However only of any use if they help understand, make and document useful tools and systems for humanity. Usefulness is the key word.

Pipi 9 uses all of the above in that context regardless of what they are called technically.

Mike Peters
https://www.blog.ajabbi.com
New Zealand

[John F Sowa]

Alex,

The statement "All models are wrong. Some are useful," as Paul said, is by the statistician George Box.  It is most definitely NOT a joke.  It is a fact.  ALL currently known (or guessed) theories of the universe are WRONG in their predictions of what would be observed by the new James Webb telescope.

Whenever scientists and engineers refine their instruments to a deeper or finer or farther distance, they discover new facts that show the limitations of their earlier observations, instruments, and theories.  Just look at the latest results from the James Webb telescope.  All previously "known" or "believed" theories are wrong.

AS:  ABSTRACTION is a way of thinking we should pose before any abduction, deduction, induction.

No.  Abstraction is just one way of looking at the data prior to making a guess about the meaning.   And guessing is just a commonsense word for 'abduction'.  If you can find a better way of guessing, it is, by definition, another method of abduction.

John

 

---

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

[Alex Shkotin]

Mike,

How do you see the Pipi 9 user interface on the WWW? Is that like QAS (query answering system)?

Please, just describe a simple use case. 

Is that right, that Google QAS (aka Search), Chat GPT, Claude 3 etc. are CAS in your classification?

The description of a Pipi 9 use case, even before implementing it, would be very useful.

Self-described system is a dream of a developer :-)  Usually we had requirements for internally documented code, like Java doc.

Alex

вт, 30 апр. 2024 г. в 23:50, Mike Peters <mi...@redworks.co.nz>:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/9db02387-0159-4c1d-b671-8493934afab8n%40googlegroups.com.

[Mike Peters]

Hi Alex,

AS: How do you see the Pipi 9 user interface on the WWW? Is that like QAS (query answering system)?

Please, just describe a simple use case. 

Is that right, that Google QAS (aka Search), Chat GPT, Claude 3 etc. are CAS in your classification?

The description of a Pipi 9 use case, even before implementing it, would be very useful.

Self-described system is a dream of a developer :-)  Usually we had requirements for internally documented code, like Java doc.

MP: Pipi 9 is a platform for hosting highly complex systems, that are important for society, with lots of moving parts, highly expensive, a lot of bureaucratic waste and unintended consequences. An important problem to solve. A good example would be big health systems.

A Complex Adaptive System (CAS) is one where the whole is greater than the sum of the parts, because of the interactions between the parts. A CAS emerges when there are a lot of parts interacting, the randomness in mass becomes statistical giving rise to scientific laws, and emergent properties emerge. 

• https://www.blog.ajabbi.com/2023/12/george-ellis-and-emergence.html

A biological cell is a perfect example.  

Markus Covert, PI at Stanford's Covert Lab came up with the first biological cell simulator which illustrates this.

• https://www.blog.ajabbi.com/2023/12/markus-covert-and-mycoplasma.html

Pipi 9 is a CAS and has emergent properties that can be harnessed. To do this the internal architecture is wildly different from other computer systems.

• https://www.blog.ajabbi.com/2024/02/evolutionary-computing.html
• https://www.blog.ajabbi.com/2023/12/how-to-deal-with-parameters-for-whole.html

The documentation/Admin UI interface exists to make PIPI answerable to humans. While it is a black box, it can be contained by parameters, coached and trained.

• https://www.blog.ajabbi.com/2023/12/using-coldfusion-as-static-site.html

Pipi 9 is a low code/no code platform. When working on Pipi 6 some 7 years ago, I made up some widely different sample applications to test on the platform. Space, Film, Ecological Restoration, Farming, etc. Seemed to work well.

So users will work with a application running in the cloud on this platform. It will be a work tool to run something.

Developers, Scientists, etc., can play with the parameters, ontologies etc., in the middle layer. Pipi will respond and learn from both.

Hope that helps. It will be easier to understand when you try it. The back end is like nothing else out there.

Mike Peters
https://www.blog.ajabbi.com
New Zealand

[Alex Shkotin]

John,

I am now on the way to formalize Static theory as a part of Mechanics. And not only theory, where we have tractatus to begin with (Principia, and later) but also formulation and solution of tasks using the theory of static mechanical systems.

I will give an example of setting up a specific problem that has to be dealt with using concretization and abstraction. These knowledge transformations can probably be called differently.

initial problem statement:

“The weightless beam is held in a horizontal position by a hinged-fixed support in point A and a vertical rod BC.

At point D, a concentrated force F=30kN is applied to the beam at an angle of 50°."

concretization 

“The weightless beam is held in a horizontal position by a hinged-fixed support at point A of the basic structure and a vertical rod BC, rigidly attached at point C to the basic structure.

At point D, a concentrated force F=30kN is applied to the beam at an angle of 50°."

HERE we had to add a basic structure and "rigidly attached…" to get a situation with a full composition of bodies.

more concretization and geometric abstraction

"points: A, B, C, D.

segments: A-D, B-C.

ray F: from point D in the direction 50° clockwise from A>D.

point B belongs to the segment A-D.

B-C is perpendicular to A-D in the left half-plane from A>D.

the ratio of segments A-B to B-D is 3/2."

where A>D means a ray from A in the direction of D.

HERE we get a geometric figure for the problem of finding the reaction of supports at points A and B will be solved.

I just wanted to share how concretization and abstraction are used when solving engineering problems using theoretical knowledge.

It is interesting that the texts of Euclid’s Elements that came to Europe did not contain drawings. I wonder if they were in Euclid's original treatise? Probably not.

I am looking forward to getting your terminology about this kind of knowledge processing.

We use theories to solve tasks with accuracy, precision needed for our goals.

We formalize theories and tasks to get support from a computer system.

The number of rules we apply during task solving is a little bit amazing.

Like this [1] from here. Please, keep in mind that it is under development.

Alex

[1]DATA

Rules for transforming propositions and phrases

In what follows, for brevity, the word otext will denote the proposition or phrase being processed.

There are many ways to get from one text to another or several others. It is assumed that the processing of the original text is replaced by the processing of those obtained from it, and we know how to process the received ones, i.e. having received their values, get the value of the original one. We will call such processing methods rules. 

The texts resulting from applying the rules are called subtasks.

The following are descriptions of different rules.

"subtask"

A rule named “subtask” only indicates that this subtask is separated into a separate task, i.e. must have its own solution block - if it exists, it is indicated in the “parameters” column (see below), and if the cell keeps "???", then the corresponding task has not yet been added to the framework! And strictly speaking, the task block has not been completed to the end.

"interpretation"

This rule makes the transition from terms of theory to terms of structures.

For example, in "There exists x in U such that _e1 is incident to x." the term "incident" is used, introduced in the theories of Biria and used in Ugraphia for inc global variable. Knowing its “binding” with inc we can interpret the statement as “There exists x in U such that (_e1 x) in inc.”, where the term “incident” is not used.

It is important to emphasize that in the resulting statement no terms of theories are used!

"substitution" 

This rule refers to the substitution of the definition of a term at the place of its use. Typically, a definition consists of a precondition formulated in a sentence beginning with the word “Let” and the definition itself, consisting of a phrase using the term, a syntactic connective (for example, “if and only if”) and a determinant - a phrase that specifies the meaning of the term. Definitions of terms are given in one theory or another - in our case it is Ugraphia. From a programming point of view, a definition is a macro command, and a “substitution” is a macro substitution, and sometimes the substituted text itself (preconditions and determinant) is modified.

The substitution comes down to the fact that the statements of the precondition form a linear block, and the determinant - a node of the decision tree.

For example, consider the statement

_e1 parallel to _e4.

it uses the term from Ugraphia - “parallel”.

Applying his definition:

eng	Let e1, e2 denote edges. e1 is parallel to e2 if and only if e1 and e2 denote different edges and e1 has the same endpoints as e2.

we get two precondition statements:

_e1 is an edge.

_e4 is an edge.

which need to be checked for truth, 

and a subtask:

_e1 and _e4 are different and _e1 has the same endpoints as _e4.

It is easy to see that the actual parameters of the place where the term is used: _e1 and _e4, are substituted in place of the formal parameters e1, e2 of the “macro command” of the definition.

"quantifier expansion"

A quantifier always runs over some set, sequence, and is expanded by running into an operation on statements or phrases written for each element of the set - a good move to finitism.

For example, consider the expansion of the second quantifier "every" in

"every member of every pair of inc belongs to U."

we need to run by inc and in our case, see the __inc block of the framework, we will get from inc for the first element:

"every member of (_e1 _v1) belongs to U." 

etc. for each pair in inc.

Notes. The finitistic way for the quantifier expansion in mathematical logic can be found, for example, in Esenin-Volpin works. And it can be stated using an example like this

Let S be {e1 e2} and p() be a unary predicate on S. Then

"∀x:S p(x)" expands into "p(e1)∧p(e2)"

the generalization to the case of any finite number of elements in non-empty S is obvious.

Thus, the quantifier statement is expanded into several more specific statements and the meaning of the original statement is obtained by applying the operation ∧ or +, etc. to the expanded values.

"split"

This is a situation when in the original statement a conjunction such as “and” or the words “equals”, “plus” and the like are applied to two specific statements or phrases.

For example, in 

"_e1 and _e4 are different and _e1 has the same endpoints as _e4." 

we split along the second “and”, obtaining two statements - subtasks: 

"_e1 and _e4 are different."

"_e1 has the same end vertices as _e4."

"reformulation"

According to this rule, various free texts in NL are converted into equivalent but more regular ones.

For example, 

"_e1 is incident to some element from U."

becomes

"there exists x in U such that _e1 is incident to x."

"choice"

Execution of a rule consists of SEARCHING in a set and determining the presence or absence of an element that satisfies a given condition.

for example, applying the "choice" rule to a statement

" there is x in U such that (_e1 x) in inc. "

consists of running through U and checking for the current element that it, paired with _e1, is present in inc. If such an element is found, then the statement is considered true, otherwise - false.

"selection"

From a certain set, a subset of elements is selected according to some criterion.

for example, in the phrase 

" number of elements of U such that it's an edge and simple and incident _v1 "

before counting the quantity, it is necessary to obtain a subset of U with the described elements and then the phrase is converted into

" number of elements in list (_e1, _e2, _e4) ".

"obviously"

With text, the mental action of obtaining its meaning is performed.

For example, it is obvious that the following statements are true:

"_e1 and _e4 are different."

"_e1 is an element of U."

"(_e1 _v1) в inc."

although in the last two cases you need to look at U and inc, respectively.

"count"

The calculation formulated in the phrase is performed in the mind.

For example, 

counting "number of elements in list (_e1, _e2, _e4)" will give the result 3.

ср, 1 мая 2024 г. в 01:27, 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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/935921a317a54203b56bb155256b864c%40bestweb.net.

[Alex Shkotin]

Mike,

Fine. You know I am ready to test your system if you need this.

Alex

ср, 1 мая 2024 г. в 11:47, Mike Peters <mi...@redworks.co.nz>:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/d4040a42-99c8-48ca-80d4-e9bdc1e5ec11n%40googlegroups.com.

[Neil McNaughton]

>> However, IMHO, If the model is a model of the “real world” then is it not a “data” model.  I would rather call things what they are: if it’s a “data model” you’re modelling data structures.  If you’re modelling the “real world” then it is not a “data” model – because it’s not a model of data.  (The term “conceptual data model” should be banished from everyone’s vocabulary!)  If you are modelling the real world, it could be called an “ontology”. 

 

I like this… do others? Could it be considered as an accepted distinction? Canonical even?

 

Neil

 

Neil McNaughton

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

The Data Room SAS

 

 

From: 'Burkett, William [USA]' via ontolog-forum <ontolo...@googlegroups.com>
Sent: Saturday, April 27, 2024 12:40 AM
To: ontolo...@googlegroups.com
Subject: RE: [External] Re: [ontolog-forum] top layer ontologies, domain layer, application layer and data models?

 

Wrt to Chris P’s point that Mike P repeats:  “So if the data model is picturing the real world, it is in some sense ontological”

 

I agree with that.  However, IMHO, If the model is a model of the “real world” then is it not a “data” model.  I would rather call things what they are: if it’s a “data model” you’re modelling data structures.  If you’re modelling the “real world” then it is not a “data” model – because it’s not a model of data.  (The term “conceptual data model” should be banished from everyone’s vocabulary!)  If you are modelling the real world, it could be called an “ontology”. 

 

Mike: I’ve read David Hay’s books and they’re okay.  He tends to get caught up in malady I call “meta-itis” – modeling the model in the model.  (Mathew West, RIP, did this a lot too.) 

 

So many, if not all, data modelling books talk about modelling “business objects” – “things important to the business.”  That is - again IMHO and with all due respect – so lame!  All of these books miss the important dimension of the why certain entities that are perceived to exist in the real work (i.e., “business objects”) are or should be represented in the model.  It is not just that the “account, order, party, biological cell, atom, protein, ecosystem, planet” are there and are recognized and are therefore represented in the model.  There is a business (usually process) requirement to capture information about these things.  If you don’t have this connection, why is it in your model?  That’s an issue I have with a lot of ontology work in the abstract – the ontologies just “represent objects in the real world” with no reason for why they’re in the model.

 

Bill Burkett

 

 

 

From: ontolo...@googlegroups.com <ontolo...@googlegroups.com> On Behalf Of Mike Peters
Sent: Friday, April 26, 2024 1:21 PM
To: ontolog-forum <ontolo...@googlegroups.com>
Subject: [External] Re: [ontolog-forum] top layer ontologies, domain layer, application layer and data models?

 

This is a great discussion that gets to the heart of things IMO. Alex S. , thanks for using Claude to generate a list of useful readings; well, it's a good start, anyway. I read the book (s) by David C. Hay - all his books are excellent. Chris

--

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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/5cf0bd73-e00e-4625-8d7c-23c4b6f788c5n%40googlegroups.com.

--
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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxRORNcgetb50gxU1Vdc%2BMu7BB6VSnEEVwESzcXTOgP_CDEg%40mail.gmail.com.

--
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 a topic in the Google Groups "ontolog-forum" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ontolog-forum/83loKPZIPzk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ontolog-foru...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAMWD8MqLgn0KW46gaUYEpWod7gVHLUuA-tG_w1mOmbBjcUi_VQ%40mail.gmail.com.

--
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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/CALGFikco0b9W3%2BuapU_4wQNS8BQOnDcS2dXofTn7iRYaHq-QqQ%40mail.gmail.com.

--
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 a topic in the Google Groups "ontolog-forum" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ontolog-forum/83loKPZIPzk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ontolog-foru...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAMWD8MrCs-i1UyqC9WTQCjLXFx0G2PmfiRSaXFthd4Ok5AXWuw%40mail.gmail.com.

--
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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/CALGFike4GjZee2%2BjUjjO8g0-s0snCUthL27fwLCkUaBRbxYF8w%40mail.gmail.com.

--
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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/274b4a00-115d-404b-bc1e-5ecabd00717en%40googlegroups.com.

--
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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/MW4PR06MB8219A55D45FBCAF0DAAE338591162%40MW4PR06MB8219.namprd06.prod.outlook.com.

[Barry Smith]

On Tue, Jun 11, 2024 at 4:06 AM Neil McNaughton <nei...@oilit.com> wrote:

>> However, IMHO, If the model is a model of the “real world” then is it not a “data” model.  I would rather call things what they are: if it’s a “data model” you’re modelling data structures.  If you’re modelling the “real world” then it is not a “data” model – because it’s not a model of data.  (The term “conceptual data model” should be banished from everyone’s vocabulary!)  If you are modelling the real world, it could be called an “ontology”. 

 

I like this… do others? Could it be considered as an accepted distinction? Canonical even?

You hit the nail on the head: https://philpapers.org/archive/SMIBCO 

Barry

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/AM0PR10MB37151473A1ADFF89A8D229D7ACC72%40AM0PR10MB3715.EURPRD10.PROD.OUTLOOK.COM.

[John F Sowa]

Neil, Barry, List,

I agree that the word that modifies the word 'model' should describe whatever you're modeling.

A data model would be a model of various kinds of data structures, such a bits, bytes, lists, matrices. etc.

A model train would model a big thing that travels the rails.

But I would say that an ontology would classify, describe, and relate anything and everything.   In effect, it would be a model of models.

John

 

---

From: "Barry Smith" <ifo...@gmail.com>

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAN_82SSvPXSy-4w4iULjN1EDbeqxjwYN%3D1OMbkzT3-cQcTXW6g%40mail.gmail.com.

[Neil McNaughton]

On Tue, Jun 11, 2024 at 4:06 AM Neil McNaughton <nei...@oilit.com> wrote:

>> However, IMHO, If the model is a model of the “real world” then is it not a “data” model.  I would rather call things what they are: if it’s a “data model” you’re modelling data structures.  If you’re modelling the “real world” then it is not a “data” model – because it’s not a model of data.  (The term “conceptual data model” should be banished from everyone’s vocabulary!)  If you are modelling the real world, it could be called an “ontology”. 

I like this… do others? Could it be considered as an accepted distinction? Canonical even?

You hit the nail on the head: https://philpapers.org/archive/SMIBCO 

Barry

 

Thanks for this. It’s quite wordy though. Can I say that data is about databases and ontology is about graphs? Or maybe the other way around?

Neil

 

From: ontolo...@googlegroups.com <ontolo...@googlegroups.com> On Behalf Of Barry Smith

Sent: Wednesday, June 12, 2024 1:08 AM
To: ontolo...@googlegroups.com

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAN_82SSvPXSy-4w4iULjN1EDbeqxjwYN%3D1OMbkzT3-cQcTXW6g%40mail.gmail.com.

[Dan Brickley]

On Wed, 12 Jun 2024 at 08:46, Neil McNaughton <nei...@oilit.com> wrote:

On Tue, Jun 11, 2024 at 4:06 AM Neil McNaughton <nei...@oilit.com> wrote:

>> However, IMHO, If the model is a model of the “real world” then is it not a “data” model.  I would rather call things what they are: if it’s a “data model” you’re modelling data structures.  If you’re modelling the “real world” then it is not a “data” model – because it’s not a model of data.  (The term “conceptual data model” should be banished from everyone’s vocabulary!)  If you are modelling the real world, it could be called an “ontology”. 

I like this… do others? Could it be considered as an accepted distinction? Canonical even?

You hit the nail on the head: https://philpapers.org/archive/SMIBCO 

Barry

 

Thanks for this. It’s quite wordy though. Can I say that data is about databases and ontology is about graphs? Or maybe the other way around?

When you express things in a data-centric language (xml Schema systems, json schemas etc.) you are generally defining data structures. This can be used directly to validate data/docs against those definitions. The contrast is that when you express yourself in an ontology-centric language (including rdf, however "toy" it seems to semanticists) you're better considered as defining genralizations about the world represented by various pieces of data. These can be used to check for contradictions in data/docs, infer missing implied information etc. But it limits the direct use for validation in that from an ontology perspective there is nothing wrong with a piece of data with stuff missing from it, or added to it. Or written in a different syntax or with its claims written out in a different order.

This is why RDF folk came up with additional validation systems (shack, shex etc.).  It all can be confusing because both traditions give you machinery to say things about "Person", "Shipping Order" etc., and at a certain level data structure rules are as much things in our world deserving description as anything else. 

Dan

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/AM0PR10MB3715D101BEA0AA36EC56F9BBACC02%40AM0PR10MB3715.EURPRD10.PROD.OUTLOOK.COM.

[Chris Partridge]

Guys,

>> However, IMHO, If the model is a model of the “real world” then is it not a “data” model.  I would rather call things what they are: if it’s a “data model” you’re modelling data structures.  If you’re modelling the “real world” then it is not a “data” model – because it’s not a model of data.  (The term “conceptual data model” should be banished from everyone’s vocabulary!)  If you are modelling the real world, it could be called an “ontology”. 

Bill and I have been having a long conversation about this offline.

Bill is , of course, quite right about the mainstream orthodox view of modelling in the IS community. 

This envisages a starting point where one develops a model of the real world (the model represents/refers to the real world)
At some stage (the data modelling stage) one then uses exactly the same model as a data model to represent the data structure of the system you are building. Where, of course, some of the data then represents the real world - so a two level model. 

I have, in the last decade or so, come to think that there is a different and better way to describe what is going on - one that opens up new ways of working.

The kind of conceptual models we build using things like UML modelling tools, it seems to me, can be regarded as representing the real world, but also (at the same time) showing how this representation can be captured as data.

(For more on showing see Macbeth, D. (2012). Seeing How It Goes: Paper-and-Pencil Reasoning in Mathematical Practice. Philosophia Mathematica, 20(1), 58–85. https://doi.org/10.1093/philmat/nkr006)

So the shift from conceptual to data models can be seen as not a shift in representation but different roles for the same model - one of representing the other of showing.

And that during the SDLC this model evolves becoming a closer and closer match for how it shows the data in the system - and then (with modern MDA tools) it becomes the system.

Interestingly, this then gives these IS models a similar sense to models elsewhere, e.g. in architecture, where a scale model is built to show what the building would look like.

Also, closer to the sense of model in science that is explored by e.g. Margaret Morrison and Mary Morgan.

There is a lot more to say, but just two quick points.
If one is reverse engineering, so  travering the SDLC in the opposite direction, then this showing perspective is a much better match for the practice we follow in bCLEARer. So we start with a system and then we 'evolve' the whole system to one where the 'conceptual' underpinning (the real world representing) is much clearer and explicit. 

If one accepts that one is both representing the real world and showing the system, then the current practices of only modelling (representing and showing) entity types at the conceptual modelling stage seems odd - why not model the individual entities - the particulars.

On Wed, 12 Jun 2024 at 08:46, Neil McNaughton <nei...@oilit.com> wrote:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/AM0PR10MB3715D101BEA0AA36EC56F9BBACC02%40AM0PR10MB3715.EURPRD10.PROD.OUTLOOK.COM.

[Burkett, William [USA]]

Chris and I have indeed been having a long, offline conversation about this topic.  I appreciate his patience and engagement as we’ve tried to understand each other.  (I haven’t quite reached a full understanding of his position on “showing” versus “being a model about/of”, but I’m working on it – I suppose I should ready the Macbeth material. 😊)

 

A persistent point of non-agreement between our points of view is reflected in Chris’s statement:

 

“At some stage (the data modelling stage) one then uses exactly the same model as a data model to represent the data structure of the system you are building.”

 

It is entirely plausible and reasonable to do this, but I feel that a model of the real world (“concept model” or “ontology”) and a model of data (“data model”) are two separate models with distinct purposes (or perhaps distinct primary purposes.)  The model of the real world should (ideally) be more robust and stable over time, presenting a “picture of the world as I/we understand it”.  Many differently-structured models of data (and the data they govern) can provide the same information (with degrees of semantic approximation) as described/stated/asserted/represented by the model of the real world.  Models of data (In addition to representing/encoding/conveying information about the real world) are also strongly and primarily driven by data processing considerations – requirements that do not (or should not) encumber models of the real world.  The relationship between the model of the real world and one or more models of data is established by mapping.  (The mapping effectively says: “these bits of data provide information about these things in the model of the real world.”)

 

Our positions are just different and our non-agreement is perfectly okay.  Maybe we just haven’t yet uncovered our motivations or drivers for preferring our positions.

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAMWD8MoP%3D%3D5MyySqbpFC72x2M5pX6tDyKcPB91SCjkeSd0Q-DA%40mail.gmail.com.

[Chris Partridge]

@bill - I need to say I appreciate the accuracy with which he is thinking of the issue.

Just to be clear, I was suggesting that the standard view (as documented in textbooks) is that one makes this shift. See the extract below from an earlier paper of mine (before I adopted the show/represent approach)

The challenge of epistemic divergence in IS development, Communications of the ACM Volume 52 Issue 6. pp 127–131

https://www.borosolutions.net/challenge-epistemic-divergence-development - doi: 10.1145/1516046.1516079 - "This way of thinking of the development of the Information specification is now generally accepted and can be found in standard textbooks. For example, [7] state that at the first stage a model is constructed by asking the customers what are “…the “things” that the application or business process addresses” (the business UoD model). Their work goes on to say “These “things” evolve into a list of input and output data objects as well as external entities that produce or consume data” (the system UoDD model). This description suggests a single specification where the analyst acts as an interpreter, who makes the shift in interpretation between the UoD and the UoDD. To complete the process, there is a subsequent stage where the technological concerns are taken into account and the system UoDD model is transformed into a physical model to be implemented. "
[7]. Pressman, R.S. and Ince, D. Software Engineering: A Practitioner’s Approach. McGraw-Hill, London, 2000.

This reflects (maybe roughly?) Bill's view I think.

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/PH0PR06MB81443951575F70829B38836591C02%40PH0PR06MB8144.namprd06.prod.outlook.com.

[Burkett, William [USA]]

John –

 

Your statement:

 

> In effect, it would be a model of models.

 

Would literally mean that an ontology is a “metamodel”.  Does that mean every ontology is a metamodel?  (I would think not.)   And if an ontology is a model of a model, then it would be a “model of a model” and not a “model of the real world” unless there’s some transitivity going on that my ham-handed brain has some trouble wrapping itself around.  I myself have trouble understanding models that model the model in the model (which in the past I’ve referred to as a malady called “meta-itis”.)

 

Bill

 

 

 

 

From: ontolo...@googlegroups.com <ontolo...@googlegroups.com> On Behalf Of John F Sowa
Sent: Tuesday, June 11, 2024 7:43 PM
To: ontolo...@googlegroups.com

Subject: Re: [External] Re: [ontolog-forum] top layer ontologies, domain layer, application layer and data models?

[Alican Tüzün]

Isn't the "data" just an observation of an artifact or human, hence the data structure is the structure of the observation?

In other case, isn't the ontology how the model of reality is?

We already know the observation is mere observation however not the reality. However an ontology is the human understood reality hence a model of it. For example meta-physics cannot be observed hence a data model of it is not possible. However an ontology of the metaphysics is possible.

Alican

--

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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/PH0PR06MB8144106A66A606F614AAC56591C02%40PH0PR06MB8144.namprd06.prod.outlook.com.

[Nadin, Mihai]

Data is the outcome of measurement.

Mihai Nadin

Sent from planet earth

On Jun 12, 2024, at 11:17 AM, Alican Tüzün <tuzun...@gmail.com> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAKTnhHyq2f0HvU3KviF-637sZpSKGwO1SYg42hYzfb9TSKebDA%40mail.gmail.com.

[Steve Ray]

Bill,

Wasn't it the group from Nottingham University that built that cool editor that loaded a meta-model that then let you populate a model, then you could start it up again using the model as the meta-model and populate another level down the recursive meta hierarchy? This was back in the early 1990's STEP days.....

Steve

--

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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/PH0PR06MB8144106A66A606F614AAC56591C02%40PH0PR06MB8144.namprd06.prod.outlook.com.

[Gillman, Daniel - BLS]

Random numbers are data. Are they measurements? What about the characters printed in a book version of a novel?

 

It seems to me relying on measurement is too restrictive. I presented this idea to the list several years ago, but Frank Farance and I arrived at a definition of datum as follows:

Designation (in the sense as described in ISO 704) of a value, where a value is a concept with a notion of equality defined.

 

One nice part of this definition in my mind is the lack of reliance on where data might come from.

 

Yours,

Dan

 

 

Dan Gillman

Information Scientist

Office of Survey Methods Research

US Bureau of Labor Statistics

Washington, DC 20212 USA

Work +1.202.691.7523

Cell    +1.410.624.9582

Email Gillman...@BLS.Gov

--------------------------------------------

“Whatever it is, I’m against it.

No matter what it is or who commenced it,

I’m against it!”

   ~Prof Quincy Adams Wagstaff

 

 

 

From: ontolo...@googlegroups.com <ontolo...@googlegroups.com> On Behalf Of Nadin, Mihai
Sent: Wednesday, June 12, 2024 3:03 PM
To: ontolo...@googlegroups.com
Subject: [SOCIAL NETWORK] Re: [External] Re: [ontolog-forum] top layer ontologies, domain layer, application layer and data models?

 

CAUTION: This email originated from outside of BLS. DO NOT click (select) links or open attachments unless you recognize the sender and know the content is safe. Please report suspicious emails through the “Phish Alert Report” button on your email toolbar.

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/34693C3D-55EE-4F2D-8CC1-3470EB30DDF0%40utdallas.edu.

[Steve Ray]

Now I remember. It was Alison McKay at Leeds University...

Steve

--

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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/PH0PR06MB8144106A66A606F614AAC56591C02%40PH0PR06MB8144.namprd06.prod.outlook.com.

[Alican Tüzün]

@dan the definition you provided cuts off the observation-measurement-data chain, which in my opinion is more general definition than cybernetics grounded definition of data.

@nadin If you are talking about the measurement after the observation, we are talking about the same thing.

Best,
Alican

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/SA1PR09MB830248A69A61C796EA9FA3C69EC02%40SA1PR09MB8302.namprd09.prod.outlook.com.

[Michael DeBellis]

>> However, IMHO, If the model is a model of the “real world” then is it not a “data” model.  I would rather call things what they are: if it’s a “data model” you’re modelling data structures.  If you’re modelling the “real world” then it is not a “data” model – because it’s not a model of data.  (The term “conceptual data model” should be banished from everyone’s vocabulary!)  If you are modelling the real world, it could be called an “ontology”. 
 
I like this… do others? Could it be considered as an accepted distinction? Canonical even?

What does it mean to "model the real world"? IMO any model we build is "just a model" and that applies to ontologies, mathematics, logic, etc. A model is a tool that abstracts away from some details of the world in order to focus on others.  E.g., if you model the solar system you typically ignore the gravitational effect that each planet has on all the others because the n body problem is very difficult to solve and the impact of the Earth on the orbit of Neptune is so small that it is probably less than your rounding error. So a data model focuses on modeling the data of a large system. That's part of the world. The standard model of physics models how particles interact and a predator/prey model models how the distribution of predators and prey are cyclic. All of these things are modeling "the real world" but just some aspect of it. They abstract away from some details to focus on others. 

I also don't see the reason to ban "conceptual data model". A common distinction in software engineering is between analysis and design. The way I see it a "conceptual model" is just another name for an analysis model and one of the strengths of OWL and SHACL is that they allow you to model and reason about data at the analysis (or conceptual) level and unlike traditional CASE tools you don't have to transform from the analysis model into the design model. I.e., you don't have to worry about what normal form you are going to use or how to structure the data so that the most common SQL queries can be done efficiently. 

Michael

https://www.michaeldebellis.com/blog

You received this message because you are subscribed to a topic in the Google Groups "ontolog-forum" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ontolog-forum/83loKPZIPzk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ontolog-foru...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAKTnhHwvezvzYJM%3DW%2B8pvrrDuhH1QNMzptu70SuTUQOVaQR2ew%40mail.gmail.com.

[Gillman, Daniel - BLS]

Alican,

 

I just don’t see that observation (as the word is used say in statistics or science) and measurement are needed. How are my 2 examples either from measurements or not data?

 

The  “cybernetics” definition takes each datum as it is and answers the question “what is this?” without caring where it came from. That is a separate issue and is covered under provenance considerations.

 

Each datum is a representation of some underlying meaning, hence a designation. The need for a notion of equality supports computation, for every data system involves moving data around, and this requires making copies. Copying has to be faithful.

 

In my experience data arise in many ways. Some of them come from secondary considerations: classification (creating categorical data from free text), recoding (creating categorical data from combinations of other values), allocation (using an algorithm to fill in a missing value), or imputation (using a statistical model to generate missing data). There are others. The more processing applied to some data, the less like the original they are. Do they always remain observations or measurements?

 

What about storing a value for Pi when needing to calculate the circumference of circles after measuring the diameters? Pi is a constant. It isn’t observed. It isn’t measured, either. So, the stored number is not a datum?

 

Moreover, when we generate a text file, say when we write a paper, we are creating characters. There’s no fundamental difference between the bytes needed to store the characters in a text document versus the characters in a file recording the results of a survey or scientific experiment. Even if we record the data using pencil and paper, the only real difference is how we organize the characters. And we can compute with both kinds.

 

One difference is the datatype, which tells us what operations we are allowed to use on some data. Statistical and scientific data look a lot different than text from that perspective. Datatypes apply to data, though.

 

I am very hard pressed to see that observation or measurement is an essential characteristic of data.

 

Yours,

Dan

 

 

 

From: ontolo...@googlegroups.com <ontolo...@googlegroups.com> On Behalf Of Alican Tüzün
Sent: Wednesday, June 12, 2024 3:33 PM
To: ontolo...@googlegroups.com
Subject: [SOCIAL NETWORK] Re: [External] Re: [ontolog-forum] top layer ontologies, domain layer, application layer and data models?

 

CAUTION: This email originated from outside of BLS. DO NOT click (select) links or open attachments unless you recognize the sender and know the content is safe. Please report suspicious emails through the “Phish Alert Report” button on your email toolbar.

@dan the definition you provided cuts off the observation-measurement-data chain, which in my opinion is more general definition than cybernetics grounded definition of data.

@nadin If you are talking about the measurement after the observation, we are talking about the same thing.

Best,
Alican

 

On Wed, 12 Jun 2024, 9:26 pm 'Gillman, Daniel - BLS' via ontolog-forum, <ontolo...@googlegroups.com> wrote:

Random numbers are data. Are they measurements? What about the characters printed in a book version of a novel? How about storing an approximation of Pi so you can calculate the areas of circles you observed the radii for ? Pi isn'

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAKTnhHwvezvzYJM%3DW%2B8pvrrDuhH1QNMzptu70SuTUQOVaQR2ew%40mail.gmail.com.

[Nadin, Mihai]

Data has no meaning. Read Shannon. Only referenced to it represents does it conjure meaning in the act of interpretation.

Mihai’s Nadin

Sent from planet earth

On Jun 12, 2024, at 3:53 PM, 'Gillman, Daniel - BLS' via ontolog-forum <ontolo...@googlegroups.com> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/SA1PR09MB83020F975C7EEF5764B7B6889EC02%40SA1PR09MB8302.namprd09.prod.outlook.com.

[Chris Partridge]

WRT: I also don't see the reason to ban "conceptual data model". A common distinction in software engineering is between analysis and design. The way I see it a "conceptual model" is just another name for an analysis model and one of the strengths of OWL and SHACL is that they allow you to model and reason about data at the analysis (or conceptual) level and unlike traditional CASE tools you don't have to transform from the analysis model into the design model. I.e., you don't have to worry about what normal form you are going to use or how to structure the data so that the most common SQL queries can be done efficiently. 

I think that one could reframe this in terms of showing. In the case of OWL and SHACL the data structure that they are showing is both the one used for conceptual modelling and implementation. So (as Micheal says) there is no need for the showing data structures to be transformed as part of the SDLC.

And if, as this implies, the analysis model and the design model are the same thing - then one way of explaining this is there are two ways the model is working - representing the world and showing the data.

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CALGFikeAVGiagPPEzB4O%2BOZ5cG3UzKAxTdJ9UFZFZZjzy3UGWg%40mail.gmail.com.

[Burkett, William [USA]]

MD> What does it mean to "model the real world"? IMO any model we build is "just a model"

 

You’ve got a good point here, Michael.  A model is a model regardless of what’s modelled and the “real world” is everything, so a “model of the real world” is kind-of meaningless.  And models do what you describe them to do.

 

The point is about what is modelled.  Your example is a “model of the solar system”; the subject of the model is the solar system.  So perhaps the original statement should be modified as “model a Universe of Discourse” to nominally circumscribe some portion of “the real world”.

 

MD> I also don't see the reason to ban "conceptual data model".

 

My argument is simple: the textbook and conventional understanding of the term “conceptual data model” is “a model of the objects important to the business”, or some such similar description.  Where and how is the model, then, any kind of model of “data”?  The whole intent (as I read the “conventional wisdom”) is to create a model that describes objects in a UoD that are important to a business process or software system.  It is an information requirements specification rather than a system design specification. 

 

Also, I’ve got this decidedly unconventional position that as soon as you “write down” a model in OWL, you’ve got a “data model” regardless of what you might want to call it.  OWL has a formal structure and software processes consume and execute based on those structures – hence it is objectively “data model”.  Whether is an OWL structure, nested XML elements, JSON objects, RDF, or relational table – they’re all data models because they model the structure of data elements in a collective whole. 

 

Bill Burkett

 

 

 

From: ontolo...@googlegroups.com <ontolo...@googlegroups.com> On Behalf Of Michael DeBellis
Sent: Wednesday, June 12, 2024 4:10 PM
To: ontolo...@googlegroups.com
Subject: Re: [External] Re: [ontolog-forum] top layer ontologies, domain layer, application layer and data models?

 

 

 

ZjQcmQRYFpfptBannerEnd

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CALGFikeAVGiagPPEzB4O%2BOZ5cG3UzKAxTdJ9UFZFZZjzy3UGWg%40mail.gmail.com.

[Burkett, William [USA]]

MN> Data has no meaning. Read Shannon. Only referenced to it represents does it conjure meaning in the act of interpretation.

 

I agree with you on this, Mihai.  But let’s include the other side:  we speak, write, and create data with the intent of capturing and conveying “meaning” to an audience with the hope and intent that their interpretation process produces an approximately equivalent body of “meaning” (so we thereby “communicate”.)

 

Shannon was all about mathematics and mechanics.  His use of the term “information” is devoid of “meaning” and should be thought of as “information capacity” (IMHO).

 

Bill

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/7AFB0922-6412-4B30-87B1-4DA02C82D799%40utdallas.edu.

[Michael DeBellis]

Also, I’ve got this decidedly unconventional position that as soon as you “write down” a model in OWL, you’ve got a “data model” regardless of what you might want to call it.  OWL has a formal structure and software processes consume and execute based on those structures – hence it is objectively “data model”.  Whether is an OWL structure, nested XML elements, JSON objects, RDF, or relational table – they’re all data models because they model the structure of data elements in a collective whole. 

Exactly. 

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/PH0PR06MB8144FBF55ECAC221C4FE5BEA91C12%40PH0PR06MB8144.namprd06.prod.outlook.com.

[Gary Berg-Cross]

Some people make a distinction about information so there is Shannon information which gets some measurements of a size.

 Another type of information is semantic information which is aboutness -  what  the information is about.

An example of people's cite is a container where Shannon information tells you the size of the container and semantic information tells you what's inside the container.

Gary Berg-Cross 

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/PH0PR06MB8144C3233853CBE4295C442591C12%40PH0PR06MB8144.namprd06.prod.outlook.com.

[Chris Partridge]

@bill WRT:  Also, I’ve got this decidedly unconventional position that as soon as you “write down” a model in OWL, you’ve got a “data model” regardless of what you might want to call it.  OWL has a formal structure and software processes consume and execute based on those structures – hence it is objectively “data model”.  Whether is an OWL structure, nested XML elements, JSON objects, RDF, or relational table – they’re all data models because they model the structure of data elements in a collective whole. 

Is the OWL a data model in the sense it is referring to (representing) the data (schema and data)? Or, as I'd like to argue, that it shows you what the data looks like?

If I built my conceptual model using, say, SQL DDL where my intention was to represent the business, then it looks like your 'decidedly unconventional position' is that this automatically becomes a data model as it "has a formal structure and software processes consume and execute based on those structures". (Whereas if I used UML it would not be?)

I'd suggest an equally "unconventional position" that this conceptual model represents the business and shows the data structure in which it is represented allows it to be both a conceptual model and in the Burkett-sense a data model at the same time - one is not forced to choose one or other.

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/PH0PR06MB8144FBF55ECAC221C4FE5BEA91C12%40PH0PR06MB8144.namprd06.prod.outlook.com.

[John Bottoms]

Reddit's AskScience forum is a good chance to pose questions to a group of noted scientists.

[ This is the title for the current askscience: ]

AskScience AMA Series: Hi Reddit - we are back again! We are group of engineers, scientists, innovators, technologists, digital experts, and designers with a collected 45 PhDs / Professors and 35 members representing national science or engineering institutions.

[ Next is the question I posed. My apologies for the long text, but I feel this is important territory and getting access to this group of professions is rare. I'm not sure which "Roger" responded, but we may be able to track him down if deemed sufficiently important.  ]

It seems like the companies developing LLM systems are wandering about trying to figure out how to get to the next stage. Why haven't the Academies worked to build a roadmap forward? The E.U. has done that, and appear to be exceeding the U.S. in development collaboration, while the U.S. is promoting a competitive approach among the early startups. It seems when you want to build a town there is a Leather District and a Garment District; that seems to make more sense. Should we work to bring back an Office of Technology Assessment?

[ And finally we have the response from one moderator ]

chelmsfordses

•29 min. ago

Ok, a caveat here - I (Roger) am biased. Which way will soon become obvious. It is possible another of our team will present a different facet in due course.

For the uninitiated (as I was until a little while ago) the acronym LLM in this context stands for Large Language Models – a form of artificial intelligence that can interact with humans in common language, compiling the bones from text from non-techie folk in order that a machine can process them on the one hand, and on the other hand, commute states/activities etc. into easy-to-understand text to deliver to non-technical folk (either end of course can be fitted with text/speech conversion).

First, the bad (as I see it)

Large Language Models have started to invade business help pages like a cancer. If one wanted help that was not available on the website, historically one would have initiated a web chat, sent an email or, heaven forbid, picked up the telephone. Such pleas would be handled by real people who had adequate knowledge of the company for whom they worked to triage, ensuring the enquiry received appropriate attention and if necessary refer/transfer to the right department or individual for onward processing. The up-side of this model is that customers and prospects felt – and were – looked after. The down side, because they were so good, people would use these routes instead of surfing/browsing/searching the company website – a victim of their own success if you like. The front-facing staff had to increase in number to meet demand, increasing business cost, reducing efficiency (but catching up with growth brought about by increasing clientele thanks to a good experience and proliferating recommendations).

Along comes LLM. Real human LLM sales people demonstrate them to those in the ivory towers, and present them as the bees’ knees for pointing people to the right places on the website, but do so in an ever-improving tone so that the uninitiated may not appreciate that the LLM isn’t human. Great – money saved by redundant front end staff, with the customer still feeling like and being looked after. But what if the enquiry needs information not available to the LLM or on the website? This is where the LLMs come unstuck. Instead of referring the matter to the correct human colleague, they are not programmed to deal with it. No matter how polite the LLM is and no matter what lexicon they articulate, the client feels – and is – given the run-around. And of course now most of the front end staff are gone, so if eventually you do need to get a hold of a human, at best it will be a long wait ("all our representatives are busy helping other customers...."), at worst you'll end up diverted to a call centre handling multiple companies with scripts from which they can't deviate - or get cut off.

Customer Service ratings have slowly but surely declined since the genesis and proliferation of LLMs.  A HINT:  Whenever you encounter one, keep typing or saying “Speak to human”. If it recognises it you’ll get back on a good road. If it doesn’t…  find another company to deal with.

Now the good:

I'm grateful to my colleagues for this. LLMs are increasingly finding their way into companies to ease interfaces between humans and machines – Star Trek style. EG, “Please make the STL hole 4mm diameter” – with confirmation dialog may save having to redraft a CNC programming script.

For decades I have predicted a new computer language – English. Basically, with correct use of grammar, nouns and definitions, I think it should be possible to “write” a computer program in English and get predictable outcomes, with an LLM converting the prose to, say, C++. We’re actually getting there with things like Alexa and Google Home.

Now to the roadmap

For the customer services replacements, give them all to the nearest racoon and tell the racoons to run for the hills.

For the places where they really can improve things – man machine interfaces – yes let’s have a good roadmap. Perhaps the best-to-date LLM can compile one?

[ FIN, feel free to comment ]

-John Bottoms

Virus-free.www.avg.com

[John F Sowa]

To everybody who commented on this topic or read any of the email notes:

The various comments have raised valid points, but they don't converge on a clear, simple, definition that covers all the cases.  As a starting point, the following comment is a good place to begin:

Mihai Nadin:  Data is the outcome of measurement. 

That is consistent with sense #1 of 'data' in the Merriam-Webster dictionary:

1 factual information (such as measurements or statistics) used as a basis for reasoning, discussion, or calculation

2 information in digital form that can be transmitted or processed

3 Information output by a sensing device or organ that includes both useful and irrelevant or redundant information and must be processed to be meaningful

Note that all three senses include the word 'information'.   I suggest an extension of sense #1 that is based on the etymology of the word 'data' from the original Latin, "something given".  Sense #1 and the word 'give' suggest  the definition of data as information given to some agent about something for some purpose."  

In Latin, that agent may be a human or other animal.  But it might also be  a mythological being.  Robots and computers could be considered modern mythological beings  -- but only if the recipient could be considered an agent that uses the information for some purpose.  

In general, whenever anybody is trying to write definitions for new innovations, it's important to check with a good dictionary about (a) what those words mean for  more traditional things, and (b) what is the etymology that led to those meanings.

Specialized dictionaries that deal with technical specializations are also important,  But most readers will never look at any dictionary.  Therefore, it's important to keep the modern technical senses compatible with (in the same ballpark as) the common senses,

Fundamental principle:  Always consult a general-purpose dictionary before writing a specialized technical definition for any new concept.

John

 

[David Eddy]

John -

On Jun 14, 2024, at 3:05 PM, John F Sowa <so...@bestweb.net> wrote:

But most readers will never look at any dictionary.

Correct.

Now & again I will ask someone with an iPhone if they’re aware of & know how to get to the dictionary(s) on their device.

VERY uncommon knowledge.  

Occasionally they can associate it with the annoying spell-checker.

My admittedly limited experience is that Androids do NOT come with a dictionary & so far no one has asked to be directed to where one can get a US$9 OED/OAD.

I’ve been told, but not tried to verify that sometimes who / where you buy an Android determines if your “plan” comes with dictionary.

Going “off device” for dictionary exploration, is obviously unacceptable.

- David

[doug foxvog]

I always go off-device for a dictionary. My go-to is OneLook.org . It
will give a set if links: general, then specialized dictionaries for
special domains. I will often pick OED for normal words, Webster's 1912
to avoid modern definitions, an acrronym dictionary, if that's what i'm
looking for, or a domain (law,medicine, music, comp sci, ...) if i'm
looking for a definition in that domain. Sometimes, i'll go to multiple
references.

-- doug f

> --
> 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 on the web visit

> https://groups.google.com/d/msgid/ontolog-forum/10E19AE6-FFF4-4D56-B702-F59629789DE7%40davideddy.com.
>

[Leo Obrst]

+1.

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/c6e3d2f0f6d1fab8a5840c95a53a6474.squirrel%40emailmg.ipage.com.

--

Leo Obrst, lob...@gmail.com

[Alex Shkotin]

Doug,

Thanks. This is a good aggregator. And for the term "data model" they refer to Wiktionary [1] .

This reminds me of an old nice book [0] 1982. Where the first three and next 4 data models are carefully described.

First three are relational, network, and hierarchical. 

Next 4 are ER, binary, semantic network, infological.

For every data model they described  data structures, constraints on data structures, and operations with data structures.

And they have a definition for data model: "An intelligent tool that allows you to implement the interpretation of data in accordance with the specified requirements will be called a data model."

=translated back to English from Russian by Google :-)

Just to recall the beginning of the game. When we created our DBMS Kompas: an expanded subset of CODASYL, as our project leader V.I. Filippov said. 

Database is another nice term. 

Alex

[0] Tsichritzis D., Lochovsky F. Data models 1982 https://books.google.ru/books?redir_esc=y&hl=en&id=F6omAAAAMAAJ 

[1] 

сб, 15 июн. 2024 г. в 03:19, doug foxvog <do...@foxvog.org>:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/c6e3d2f0f6d1fab8a5840c95a53a6474.squirrel%40emailmg.ipage.com.

[alex.shkotin]

Actually the correct name of our great DBMS is COMPAS = COdasyl Mars PAScal after main ideas and tools.
Where MArS = Micro Archiving System.

суббота, 15 июня 2024 г. в 13:50:11 UTC+3, alex.shkotin:

[Kingsley Idehen]

--
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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/014e88424a09439b84aef36321016fb2%40bestweb.net.

Hi John and other interested parties,

There was a thread on this forum a while back that ultimately lead to my publication of a presentation titled “Understanding Data” to slideshare, which also has a safer PDF variant on our website.

Basically, I defined data as observation in reusable form.

​

-- 
Regards,

Kingsley Idehen	      
Founder & CEO 
OpenLink Software   
Home Page: http://www.openlinksw.com
Community Support: https://community.openlinksw.com
Weblogs (Blogs):
Company Blog: https://medium.com/openlink-software-blog
Virtuoso Blog: https://medium.com/virtuoso-blog
Data Access Drivers Blog: https://medium.com/openlink-odbc-jdbc-ado-net-data-access-drivers

Personal Weblogs (Blogs):
Medium Blog: https://medium.com/@kidehen
Legacy Blogs: http://www.openlinksw.com/blog/~kidehen/
              http://kidehen.blogspot.com

Profile Pages:
Pinterest: https://www.pinterest.com/kidehen/
Quora: https://www.quora.com/profile/Kingsley-Uyi-Idehen
Twitter: https://twitter.com/kidehen
Google+: https://plus.google.com/+KingsleyIdehen/about
LinkedIn: http://www.linkedin.com/in/kidehen

Web Identities (WebID):
Personal: http://kingsley.idehen.net/public_home/kidehen/profile.ttl#i
        : http://id.myopenlink.net/DAV/home/KingsleyUyiIdehen/Public/kingsley.ttl#this

[Michael DeBellis]

Kingsley, that's a nice presentation. Thanks for posting it. Are there any restrictions on using it? If I wanted to insert a few slides into my own presentation can I do that as long as I have proper attribution as to where the slides came from? Assuming  you are okay with people reusing parts of the presentation, is there any specific attribution you want us to include?

Cheers,

Michael 

https://www.michaeldebellis.com/blog

You received this message because you are subscribed to a topic in the Google Groups "ontolog-forum" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ontolog-forum/83loKPZIPzk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ontolog-foru...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/54c1e5f4-8a4d-4438-ae36-cb0823464190%40openlinksw.com.

[Kingsley Idehen]

Hi Michael,

On 6/17/24 11:03 AM, Michael DeBellis wrote:

Kingsley, that's a nice presentation.

Thanks.

Thanks for posting it. Are there any restrictions on using it?

No, just basic attribution as one would expect.

If I wanted to insert a few slides into my own presentation can I do that as long as I have proper attribution as to where the slides came from?

Yes, of course.

Assuming  you are okay with people reusing parts of the presentation, is there any specific attribution you want us to include?

Just as you've already outlined :)

Kingsley

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CALGFikeK4hFzR%3DK%3DHzi5o0XU-0Ozg63zRLbAGHNNLzfEukqydg%40mail.gmail.com.

[Nadin, Mihai]

Measurement: the dynamics of the measured affects the dynamics of the meter (this covers from measuring devices to sensorial perception and non-sensorial evaluations).

Rosen, R. 1978. Fundamentals of Measurement and representation of Natural Systems. New York :North-Holland

In this sense, observation is measuring.

The data is the outcome. The theory (based on thermodynamics) of how data is transmitted is Shannon’s contribution (with the unfortunate use of the word information as a substitute for data).

 

Mihai Nadin

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/54c1e5f4-8a4d-4438-ae36-cb0823464190%40openlinksw.com.

[Kingsley Idehen]

Hi Nadin,

On 6/17/24 1:30 PM, Nadin, Mihai wrote:

Measurement: the dynamics of the measured affects the dynamics of the meter (this covers from measuring devices to sensorial perception and non-sensorial evaluations).

Rosen, R. 1978. Fundamentals of Measurement and representation of Natural Systems. New York :North-Holland

In this sense, observation is measuring.

The data is the outcome. The theory (based on thermodynamics) of how data is transmitted is Shannon’s contribution (with the unfortunate use of the word information as a substitute for data).

 

Mihai Nadin

Observation is basically existential quantification (measurement). i.e., that something exists with certain discernible attributes. Naturally, this is situation and observer oriented.

Kingsley

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/BL3PR01MB689762EACD20B6BC01687308DACD2%40BL3PR01MB6897.prod.exchangelabs.com.

[John F Sowa]

I suggest some adjustments of terminology:

Mihai Nadin:  In the sense [of Robert Rosen], observation is measuring. 

Kingsley Idehen:   Observation is basically existential quantification (measurement). i.e., that something exists with certain  discernible attributes. Naturally, this is situation and observer oriented. 

Robert Rosen (and many other philosophers, psychologists, and neuroscientists) have studied those issues in detail.   Measurement is the result of comparing a percept to some standard as a result of selective attention and interpretation. What is selected depends on an open-ended variety of factors, of which situation and observer are significant.

MN:  The data is the outcome. The theory (based on thermodynamics) of how data is transmitted is Shannon’s contribution (with the unfortunate use of the word information as a substitute for  data). 

The word 'data' (what is given) is inappropriate for what is observed (taken and processed by some observer).  It would be a good word for what is passed along to another agent (human, animal, machine, or mythological being).

Unless anybody can suggest a better word for what is stored in the brain (or computer storage), I would say that 'information' is as good as any, and better than most.

Almost every word expresses it as the object of some verb.  We could use the Greek word 'logos',    As Heraclitus wrote and John the Evangelist repeated  (about 400 years later),   "In the beginning was the Logos."   That word is logical.

And by the way, I do not consider the word 'myth' to be a derogatory term.  Some of the most fundamental truths began as myths -- and they are transmitted in the most intelligible way in a literary presentation.

John

__________________

[Gary Berg-Cross]

John on some adjustments of terminology:>" And by the way, I do not consider the word 'myth' to be a derogatory term.  Some of the most fundamental truths began as myths -- and they are >transmitted in the most intelligible way in a literary presentation."

I guess I am less comfortable with the use of the concept-term myth here.  

Looking at a few dictionary definitions for example you can see why:

a usually traditional story of ostensibly historical events that serves to unfold part of the world view of a people or explain a practice, belief, or natural phenomenon

creation myths

b

: PARABLE, ALLEGORY

Moral responsibility is the motif of Platos myths.

2

a

: a popular belief or tradition that has grown up around something or someone

especially : one embodying the ideals and institutions of a society or segment of society

seduced by the American myth of individualism

the utopian myth of a perfect society

b

: an unfounded or false notion

the myth of racial superiority

3

: a person or thing having only an imaginary or unverifiable existence

Perhaps the term Hypothesis might be a better one to use  an adaptive concept over time.

Gary Berg-Cross 

Potomac, MD

240-426-0770

--

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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/9cb91fc83f8d49b9b0cf0b4ae07de12d%40bestweb.net.

[Michael DeBellis]

Unless anybody can suggest a better word for what is stored in the brain (or computer storage), I would say that 'information' is as good as any, and better than most.

I agree and would add one other example: the codings represented in DNA and RNA are also information. John Maynard Smith was one of the leaders of the Neo-Darwinian movement and he wrote a fascinating paper on this called The Concept of Information in Biology: http://mechanism.ucsd.edu/teaching/philbio/readings/maynardsmith.information.2000.pdf  From the abstract:

In both protein synthesis and in later development, genes are symbols, in that there is no necessary connection between their form (sequence) and their effects. The sequence of a gene has been determined, by past natural selection, because of the effects it produces. In biology, the use of informational terms implies intentionality, in that both the form of the signal, and the response to it, have evolved by selection. Where an engineer sees design, a biologist sees natural selection.  

--

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 a topic in the Google Groups "ontolog-forum" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ontolog-forum/83loKPZIPzk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ontolog-foru...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/9cb91fc83f8d49b9b0cf0b4ae07de12d%40bestweb.net.

[Gary Berg-Cross]

>protein synthesis and in later development, genes are symbols,

Yes, genes require an entire evolved cell system to work.  Their message is transcribed and carried out as part of that living, information processing system.

Gary Berg-Cross 

Potomac, MD

240-426-0770

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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/CALGFikfh0WvVaDokDmRwLPSVQtV17%3DUGJN_uVcFLPCiZykQPzw%40mail.gmail.com.

[Nadin, Mihai]

https://www.nadin.ws/wp-content/uploads/2011/09/computation_information_meaning.pdf

An article (vol. 2, Issue 1, January—April 2011, IJARITAC)—on computation, information, meaning.

Stored in computer memory—data

Brain: living matter interprets data continuously.

Information: data interpreted

Data:--of quantitative nature (see also the units of data, the measure)

Information: meaning (see also referencing data to what it means)

The genetic aspect—deserves a bit more nuance than in declaring genes to be symbols.

What is the meaning of labeling them as symbols? They are sequences of DNA—chemical compounds.

 

Mihai Nadin

 

From: ontolo...@googlegroups.com <ontolo...@googlegroups.com> On Behalf Of Gary Berg-Cross
Sent: Monday, June 17, 2024 5:15 PM
To: ontolo...@googlegroups.com
Subject: Re: [ontolog-forum] top layer ontologies, domain layer, application layer and data models

 

>protein synthesis and in later development, genes are symbols,

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAMhe4f3nHGNojWzA4v9GyU2XPBaT705h4Q0N0vJgX2NjMsf29w%40mail.gmail.com.

[Michael DeBellis]

What is the meaning of labeling them as symbols? They are sequences of DNA—chemical compounds.

The same reason we think of things such as the mapping of bits to characters in ASCII to be symbols. And just as there are dictionaries and rules of grammar for defining how to construct words and sentences out of characters, there are rules for how to interpret a genotype and use it to generate amino acids and more complex structures such as proteins.  There is a "Rosetta Stone" that defines the genetic code. See the table in this article: https://www.khanacademy.org/science/ap-biology/gene-expression-and-regulation/translation/a/the-genetic-code-discovery-and-properties  This shows the mapping from codons (sequences of 3 nucleotides) to amino acids. Then look at the other figures. Those describe how to transform genes into higher level structures. E.g., there are start and stop codons that define a higher level structure (called a frame) than individual amino acids. Over 20 years ago I saw a presentation at the Stanford Medical School where the researchers handed out little vials of water to the audience as we entered. They told us that we had a PDF version of their paper stored in RNA in the vials. They were able to write to and read from RNA and store arbitrary information such as a PDF file. 

Since then there has been significant work on how one can do general computation with DNA and RNA: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3758167/  I don't know if this will ever be practical for general computing but the point is that in theory its equivalent. It's also why I think people are short sighted to think that human cognition is only about neurons. 

BTW, another fascinating thing about the way DNA/RNA represent information is that Rosetta Stone mapping is the same for all known living organisms, from single celled bacteria to humans. That's one reason discovering alien life would be so interesting. It would help us understand if there are other ways to model information in biology besides RNA/DNA. One hypothesis for abiogenesis is that there was originally a simpler code than RNA but that RNA evolved from that and all the simple little proto-life organisms made up by proto-RNA were gobbled up by RNA and DNA organisms. 

Michael

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/BL3PR01MB6897D0C0D92826EDF24F2D85DACD2%40BL3PR01MB6897.prod.exchangelabs.com.

[Nadin, Mihai]

Peirce made us aware of the ethics of terminology. Using labels without sharing in the meaning is an exercise in futility.

You can do computation in any medium. Read Feynman about this idea.  DNA—you seem excited but not  really well informed: 1994--Leonard Adleman performed the first ever truly molecular-level computation using DNA combined with the tools and techniques of molecular biology. 

PLEASE: ontology without a clear understanding of the concept used degrades very fast. Even faster than the so-called AI….

 

Mihai Nadin

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CALGFikeZJnY7oGR_egdEQ1De%2BEW_o2JEN_08624%2Bj5rb-m1UqA%40mail.gmail.com.

[Alex Shkotin]

Hi Kingsley!

Nice presentation of your approach.

Let me just mention that nowadays we have so many kinds of data so intensively processed that we have Data Science [1]. The situation of 40 years ago when we had maybe 10 data models and 5 DBMS looks so simple now.

For data in a broad sense we have this [2]. Where for me the key ideas are "value" and "structure".

Alex

[1] https://en.wikipedia.org/wiki/Data_science

"Data science is an interdisciplinary academic field^[1] that uses statistics, scientific computing, scientific methods, processes, scientific visualization, algorithms and systems to extract or extrapolate knowledge and insights from potentially noisy, structured, or unstructured data.^[2]

Data science also integrates domain knowledge from the underlying application domain (e.g., natural sciences, information technology, and medicine).^[3] Data science is multifaceted and can be described as a science, a research paradigm, a research method, a discipline, a workflow, and a profession.^[4]

Data science is "a concept to unify statistics, data analysis, informatics, and their related methods" to "understand and analyze actual phenomena" with data.^[5] It uses techniques and theories drawn from many fields within the context of mathematics, statistics, computer science, information science, and domain knowledge.^[6] However, data science is different from computer science and information science. Turing Award winner Jim Gray imagined data science as a "fourth paradigm" of science (empirical, theoretical, computational, and now data-driven) and asserted that "everything about science is changing because of the impact of information technology" and the data deluge.^[7][8]

A data scientist is a professional who creates programming code and combines it with statistical knowledge to create insights from data.^[9]"

[2] https://en.wikipedia.org/wiki/Data 

"In common usage, data (/ˈdeɪtə/, also US: /ˈdætə/; ) are a collection of discrete or continuous values that convey information, describing the quantity, quality, fact, statistics, other basic units of meaning, or simply sequences of symbols that may be further interpreted formally. A datum is an individual value in a collection of data. Data are usually organized into structures such as tables that provide additional context and meaning, and which may themselves be used as data in larger structures."

--
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 on the web visit https://groups.google.com/d/msgid/ontolog-forum/54c1e5f4-8a4d-4438-ae36-cb0823464190%40openlinksw.com.

[Kingsley Idehen]

Hi Alex,

On 6/18/24 5:30 AM, Alex Shkotin wrote:

Hi Kingsley!

Nice presentation of your approach.

Let me just mention that nowadays we have so many kinds of data so intensively processed that we have Data Science [1]. The situation of 40 years ago when we had maybe 10 data models and 5 DBMS looks so simple now.

For data in a broad sense we have this [2]. Where for me the key ideas are "value" and "structure".

Alex

[1] https://en.wikipedia.org/wiki/Data_science

"Data science is an interdisciplinary academic field^[1] that uses statistics, scientific computing, scientific methods, processes, scientific visualization, algorithms and systems to extract or extrapolate knowledge and insights from potentially noisy, structured, or unstructured data.^[2]

Data science also integrates domain knowledge from the underlying application domain (e.g., natural sciences, information technology, and medicine).^[3] Data science is multifaceted and can be described as a science, a research paradigm, a research method, a discipline, a workflow, and a profession.^[4]

Data science is "a concept to unify statistics, data analysis, informatics, and their related methods" to "understand and analyze actual phenomena" with data.^[5] It uses techniques and theories drawn from many fields within the context of mathematics, statistics, computer science, information science, and domain knowledge.^[6] However, data science is different from computer science and information science. Turing Award winner Jim Gray imagined data science as a "fourth paradigm" of science (empirical, theoretical, computational, and now data-driven) and asserted that "everything about science is changing because of the impact of information technology" and the data deluge.^[7][8]

A data scientist is a professional who creates programming code and combines it with statistical knowledge to create insights from data.^[9]"

[2] https://en.wikipedia.org/wiki/Data 

"In common usage, data (/ˈdeɪtə/, also US: /ˈdætə/; ) are a collection of discrete or continuous values that convey information, describing the quantity, quality, fact, statistics, other basic units of meaning, or simply sequences of symbols that may be further interpreted formally. A datum is an individual value in a collection of data. Data are usually organized into structures such as tables that provide additional context and meaning, and which may themselves be used as data in larger structures."

Yes, but please note that I am very deliberate in the generic definition: Data is observation expressed in reusable form.

"Form" is where notations (and their underlying syntax) and serialization formats come into play. “Data Science” still fits into the aforementioned generic description.

We observe entity relationship types (relations) in reusable form via notations used for expression and formats for serialization (regarding persistence). All of this is informed by logic as the organizing schema, meaning: 

Everything is related to something else in a variety of observed or observable ways.

Kingsley

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROSem2eraEdKgQm6j-5tEVzUWf_c80MjWV8svFXJOYMa_g%40mail.gmail.com.

[Nadin, Mihai]

Excellent:

Everything is related to something else in a variety of observed or observable ways.

I wish I can convince you all to read my new book:

https://link.springer.com/book/10.1007/978-3-031-43957-5

It argues for this idea.

Mihai Nadin

 

From: 'Kingsley Idehen' via ontolog-forum <ontolo...@googlegroups.com>

Sent: Tuesday, June 18, 2024 12:06 PM
To: ontolo...@googlegroups.com

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/629d1e1a-c340-4b3e-be25-d5279b637430%40openlinksw.com.

[Alex Shkotin]

Hi Kingsley,

It would be nice to discuss a little your definition of one of the meanings of the term "data".

So you: "Data is observation expressed in reusable form."

For me observation is a process. So, is it possible to say this way:

"Data is a result of observation expressed in reusable form."?

I like to work with definitions. Precise definitions are a gem of knowledge.

Alex

вт, 18 июн. 2024 г. в 20:05, 'Kingsley Idehen' via ontolog-forum <ontolo...@googlegroups.com>:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/629d1e1a-c340-4b3e-be25-d5279b637430%40openlinksw.com.

[Alican Tüzün]

I am just adding my part,

even though the observation is a process, the observer can "observe" only discretely. The expressed result of this discrete "observation" is the data. However, I did not understand the reusable form part.

Best,
Alican 

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROSMmYY-EBNJqWM4DTVaLUjujHU%2BnWCbg1ZzvNM0%3DXHYNg%40mail.gmail.com.

[Alex Shkotin]

Alican,

Observation is quite a complex process, as we have now a lot of different observing systems. 

And your "observer" sounds human centric. Is this necessary?

About discrete portions of data - a very important topic. As we now have technologies to process data flow on the fly.

Best,

Alex

ср, 19 июн. 2024 г. в 11:07, Alican Tüzün <tuzun...@gmail.com>:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAKTnhHx8UjpvaU_qaEuVPO-hqxvthspGQAKbmw-REDao-hqbng%40mail.gmail.com.

[Alican Tüzün]

Alex,

> And your "observer" sounds human-centric. Is this necessary?

Not really; even if we consider any artificial sensor, digital or analog, still holds. 

 I am heavily influenced by a controversial book by a Nobel Prize winner. In my opinion, it explains the problem quite well. 

https://www.goodreads.com/book/show/5090358-the-logic-of-modern-physics?from_search=true&from_srp=true&qid=6zs3MC48H6&rank=1

Best,
Alican

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROTt9dherEPRvH55MTqZsk84vgs4Wp2UkvdDi3YsgzzrDw%40mail.gmail.com.

[Alex Shkotin]

What problem? And how does it explain?

For a particular problem it should be a particular page?

If we can formulate a problem in one or 10 sentences, how many sentences do we need to write an explanation?

You know, it is like you read this book and now you can help us to solve a problem. And a point "read this book and you will solve the problem" does not help very much especially during the conversation :-)

Best,

Alex

ср, 19 июн. 2024 г. в 14:03, Alican Tüzün <tuzun...@gmail.com>:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAKTnhHyQh%3DW9ni455t2V5ri1TgzsUN%2B7cFTyjs2ncF%3DF5SKKGQ%40mail.gmail.com.

[Alican Tüzün]

Alex,

You are totally right. I'm sorry for the implicit information. I WIll try to answer more thoroughly when I get home.

Best,
Alican

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROTgNbXyaUATugrOCmQ0C5njgeGvgiCEn8HeqXf10GsP%2Bw%40mail.gmail.com.

[Kingsley Idehen]

Hi Alex,

On 6/19/24 3:47 AM, Alex Shkotin wrote:

Hi Kingsley,

It would be nice to discuss a little your definition of one of the meanings of the term "data".

So you: "Data is observation expressed in reusable form."

For me observation is a process. So, is it possible to say this way:

"Data is a result of observation expressed in reusable form."?

I like to work with definitions. Precise definitions are a gem of knowledge.

Alex

As John Sowa typically suggests, it’s always best to start from a good natural language dictionary. For instance, here’s a link that denotes the definition of observation, in English, from Merriam Webster.

https://www.merriam-webster.com/dictionary/observation#this

As you can see, the definition includes:

...

a: an act of recognizing and noting a fact or occurrence often involving measurement with instruments e.g., weather observations

b: a record or description so obtained e.g. Scientific observations were sent to the committee....

...

Kingsley

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROSMmYY-EBNJqWM4DTVaLUjujHU%2BnWCbg1ZzvNM0%3DXHYNg%40mail.gmail.com.

​

[Kingsley Idehen]

Hi Alican,

On 6/19/24 4:06 AM, Alican Tüzün wrote:

I am just adding my part,

even though the observation is a process, the observer can "observe" only discretely. The expressed result of this discrete "observation" is the data. However, I did not understand the reusable form part.

Best,
Alican

"Reusable form" is about expression and representation. Without these characteristics, there's no communication mechanism; it's just in our heads (i.e., a mark), so to speak :)

Typically, we have a mark (a thought in our head) that's tokenized (in some shareable way) as a critical part of exporting observations from our heads (minds) for consumption by others.

Kingsley

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAKTnhHx8UjpvaU_qaEuVPO-hqxvthspGQAKbmw-REDao-hqbng%40mail.gmail.com.

[Alex Shkotin]

Exactly! This is why I asked for more words in your definition for "data" to eliminate ambiguity of "observation" meanings:

"Data is a result of observation expressed in reusable form."

Sometimes we may use this notation: 

"Data is a result of observation[m-w.a] expressed in reusable form."

"Data is observation[m-w.b] expressed in reusable form."

This is my observation of your definition :-)

Alex

ср, 19 июн. 2024 г. в 18:11, 'Kingsley Idehen' via ontolog-forum <ontolo...@googlegroups.com>:

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/0aa40a69-4944-428a-acd0-41b5cf510306%40openlinksw.com.