Data comes from measurement

26 views
Skip to first unread message

Nadin, Mihai

unread,
Dec 29, 2022, 5:07:07 PM12/29/22
to Ontolog Forum

Dear and respected colleagues,

Data comes from measurement. Data is agnostic of theory. Referenced to meaning data becomes information. Nothing else. Neither the data nor operations on such data reflect understanding of the phenomena the data represent.

 

The recent developments (such as chat GPT or other LLM productions) illustrate the echo chamber effect of the data used to train neural networks. Of course, it helps that we understand how language functions but this does not mean that from how it functioned in the past we can infer to how it will function in new contexts.

 

Mihai Nadin

https://www.nadin.ws

https://www.anteinstitute.org

Google Scholar

 

Alex Shkotin

unread,
Dec 30, 2022, 4:49:01 AM12/30/22
to ontolo...@googlegroups.com
Dear Mihai,

Signal, data, information, knowledge, theory - exciting sequence of ideas.

It would be interesting to go from the other end. Everyone who has access to ChatGPT investigates what ChatGPT can do and how to use it for their own purposes. Somebody found it useful, and others found it a dangerous competitor (see NO AI artists).
For me, if it, in addition to generating program code, built proofs of theorems, for example, for code verification, its benefit would be significant, because we have universal algorithms for checking the correctness of the proof.
Consider this text [2] about "...the proof of the Boolean Pythagorean Triples Problem, a long-standing open problem in Ramsey Theory. This 200TB proof has been constructed completely automatically..."
Not only ANN are using computer power:-)

As for "understanding", I like the "Nagorny's thesis" [1]:
"Hilbert's idea  —   to turn axiomatized mathematics into a kind of "manipulation of formulas"  —  long before the onset of the "era of machine mathematics" anticipated the main, as it seems to us, ideological postulate of this era: namely, thought about the fact that something can become generally understandable, i.e. “understandable”, including by a computer, only when it does not require any understanding at all."
One can probably say in another way: if the system in its work keeps the laws of thinking, then it does not need to know anything about the laws of thinking.

The construction of a particularly formal theory sometimes looks like the construction of a coral reef: a lot of life, passions, hopes, in order to actually get the reef itself:-)

MC&HNY,

Alex

[1] (ru) https://iphras.ru/uplfile/logic/log10/Li_10_Nagorniy.pdf
[2] https://cacm.acm.org/magazines/2017/8/219606-the-science-of-brute-force/fulltext

пт, 30 дек. 2022 г. в 01:07, Nadin, Mihai <na...@utdallas.edu>:
--
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/DM8PR01MB6904E5A45E54BCF3B88D51F8DAF39%40DM8PR01MB6904.prod.exchangelabs.com.

Michael DeBellis

unread,
Jan 4, 2023, 2:12:25 PM1/4/23
to ontolog-forum
> Data is agnostic of theory.

I don't think that's completely true. Consider the information from a radio telescope. To someone like me it will just be a bunch of numbers but to an astronomer it will mean information about a red dwarf, black hole, etc. 

Or the core dumps we used to get in the bad old days of debugging on a mainframe. To most people just a bunch of numbers and letters. To the person who wrote the code it tells them where the program terminated, the value of various areas in memory, etc. 

Another example: data from an fMRI. What an fMRI measures is blood flow in the brain. But to a neuropsychologist it indicates increased or decreased brain activity because blood flow is positively correlated with neuron firing. What they are really interested in measuring is neuron firing but their theory enables them to interpret data about blood flow in the brain as data about neurons firing. 

Or consider what you see the Sun do every day. Does it move across the sky or does the Earth's rotation make the position of the sun change? In the past our ancestors data said the Sun moved but now our theory tells us that the real data is the Earth's rotation causes the Sun to change position over time. 

If we are talking about raw sense data such as "there is a big Yellow circle that changes position in the sky every 12 hours" then yes that is theory agnostic. But you seldom do science or engineering with that kind of data and as soon as you start using scientific terminology and talk about stars, planets, phenotypes, genotypes, etc. data is not completely agnostic of theory. 

Nadin, Mihai

unread,
Jan 4, 2023, 6:37:26 PM1/4/23
to ontolo...@googlegroups.com

Dear and respected Michael DeBellis:

The meaning of Data is agnostic of theory.

 

You refer to interpretation of data. Always dependent upon the theory.

My statement refers to the outcome of measurement. Even if you believe that the world is flat, the data from measurements is independent of your belief and of theories (right or wrong).

 

Mihai Nadin

--

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.

Alex Shkotin

unread,
Jan 5, 2023, 3:27:13 AM1/5/23
to ontolo...@googlegroups.com
Michael,

For me, the relationship between theory and data may be even more complex if we take in consideration the theory we use to create a device for measurement.

Whether the representations generated by our senses in our ideal should be called data is a moot point. Data is good because it is received and in some form exists or existed outside of us. With them there is something to do and what to study. Including how this specific data was created, is used and interpreted.
Therefore, I am always for the fact that if the data is given, then it can be "read loud" (J. Corcoran), verbalized, expressed in human language. That is, in some theory.
We know how to read 451°F. And this is not "four five upper circle capital letter F"

Your example "a bunch of numbers and letters" is exactly about an unknown language for us: we do not know how to read this data.
Maybe the maxima is that data is always written in some language.

Alex

ср, 4 янв. 2023 г. в 22:12, Michael DeBellis <mdebe...@gmail.com>:
--

dr.matt...@gmail.com

unread,
Jan 5, 2023, 4:15:54 AM1/5/23
to ontolo...@googlegroups.com

Dear Alex,

Yes, consider the humble thermocouple. It does not measure temperature at all, but the voltage generated by an electrical circuit of two dissimilar metals. It just happens that the voltage generated varies with temperature.

 

In my experience the area of measurement, physical quantities and measurement scales is one of the most challenging areas for ontological research. It is familiar, so we think we know what is going on, but when you start to poke at it you find it is much more of a mish mash than you would expect.

 

Regards

Matthew West

Neil McNaughton

unread,
Jan 5, 2023, 4:49:17 AM1/5/23
to ontolo...@googlegroups.com

In the context of data and measurement, you might like to read my 2017 editorial titled “'Digitalization,’ from Harry Nyquist to the ‘edge’ of the internet” https://oilit.com/2017+5+3 where I argue that “The problem [of interoperability] is not just about aligning data protocols, it is also, and more profoundly, about aligning models of reality. A harder, if not impossible task!”

 

Best regards,

Neil McNaughton

2022 SPE Regional Data Science and Engineering Analytics Award Winner

Editor Oil IT Journal – www.oilit.com

Recent readers’ testimonials

The Data Room SAS

7 Rue des Verrieres

92310 Sevres, France

Landline+33146239596

Cell+33672712642

Michael DeBellis

unread,
Jan 5, 2023, 10:41:15 AM1/5/23
to ontolo...@googlegroups.com
Matthew said

In my experience the area of measurement, physical quantities and measurement scales is one of the most challenging areas for ontological research. It is familiar, so we think we know what is going on, but when you start to poke at it you find it is much more of a mish mash than you would expect.

Exactly. That's the point I was trying to make. Not that there isn't a distinction between data and theory but that when you start to "poke at it" you realize the distinction isn't as clear as you would think. Our idealization of the scientific method is that we collect data then we do induction and create general laws based on the data. But in reality the theory often constrain and defines the data and the distinction between the two isn't nearly as unambiguous as we think. Also, agree with Alex's comment: data always implies some language. That's the difference between seeing a bunch of random numbers and letters on a printout vs reading Hexadecimal and using it to understand what state a computer was in when it crashed.  

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/1BkY97l7Rtg/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/019701d920e6%244dd80570%24e9881050%24%40gmail.com.

Azamat Abdoullaev

unread,
Jan 5, 2023, 3:43:41 PM1/5/23
to ontolo...@googlegroups.com
NM: “The problem [of interoperability] is not just about aligning data protocols, it is also, and more profoundly, about aligning models of reality. A harder, if not impossible task!”
That's a good point.
Broadly, there are three pillars for fundamental knowledge and intelligent technology:
Reality/Modeling/Mapping/Representation 
Data Universe (data model, data types, data sets, data units)
Intelligence Systems (models and algorithms, inferences and predictions)
Your models of reality is all and everything.
Say, you explore the causes of global warming, if it is caused by human activities or by natural causes.
There is data, a lot of data about its key environmental and demographic variables and parameters, from different data sources, from World Bank Data to NASA, like sampled below:
image.png
Your model of climate change is the key assumption on  how to process, classify and organize the CH data. 
It may be the true one, interactive causal model of CH or the wrong ones, as a statistical/correlative model of CH or causal inference model of CH, with the set of Python causal inference libraries, with the source code as described here: 
What we are still badly missing. It is World Data Ontology, with all the consequences.


Alex Shkotin

unread,
Jan 6, 2023, 4:54:31 AM1/6/23
to ontolo...@googlegroups.com
Dear Matthew,

Exactly. The direction of technological text formalization (aka formal ontology creating) is subtle. And the volume of texts is huge. 
The theoretical texts of engineers are special one as well as the way of thinking. Models in use is another great topic.

And any ontological value has a unit of measure. There are no just natural numbers in ontology but at least with one or another sort assigned (subsuming a UoM - item).
To express a unit of measure for physical value we have a fixed number of primary words (m s...) and, as far as I know, only two operations *, / to get terms (expressions) but not every expression is valuable. What value has sec*sec unit of measure?
Even entropy has UoM: kg⋅m2⋅s−2⋅K−1.

I had attempted to formalize the text of one governmental standard for quartz sands for the glass industry. It is not trivial at all.
And there is a great topic of material algorithms: some of them to produce, others to use.

Anyway the brute force direction is to take most important technical texts and formalize them one after another.

Regards,

Alex

чт, 5 янв. 2023 г. в 12:15, <dr.matt...@gmail.com>:

Alex Shkotin

unread,
Jan 6, 2023, 5:09:23 AM1/6/23
to ontolo...@googlegroups.com, nei...@oilit.com
Neil, 

Let me leave for a while the details of analog signal processing, including its correct conversion to digital, which should be justified in each specific case, including theoretically.
What I really miss in your quote is the term "theory". After all, the model is built by applying some theory. And the problem of "aligning theories of reality" has been solved one way or another since two theorists of the same part of reality met.

Is there any chance to talk not only about models but about theories?

Best regards,

Alex

чт, 5 янв. 2023 г. в 12:49, Neil McNaughton <nei...@oilit.com>:

Alex Shkotin

unread,
Jan 6, 2023, 5:48:00 AM1/6/23
to ontolo...@googlegroups.com
Michael,

The juxtaposition of data and theory is a bit odd. They seem to be from different levels of abstraction. They usually compare facts and theories. In order for the data to turn into facts, we verbalize them in one theory or another. And then it may turn out that the facts formulated in some theory contradict it.

Alex

чт, 5 янв. 2023 г. в 18:41, Michael DeBellis <mdebe...@gmail.com>:
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CALGFikcxQnHuBLqz6ML7Qr_4r4CxRz32Sdv7-389bE3C9b_c%3DQ%40mail.gmail.com.

dr.matt...@gmail.com

unread,
Jan 6, 2023, 11:55:35 AM1/6/23
to ontolo...@googlegroups.com

Dear Alex,

Unfortunately most of the texts that are supposed to be authoritative are actually flawed through making assumptions that were convenient for their purpose.

Regards

Matthew

John F Sowa

unread,
Jan 7, 2023, 12:55:25 AM1/7/23
to ontolo...@googlegroups.com
Neil,
 
I strongly recommend the last line of your editorial:  " The problem is not just about aligning data protocols, it is also, and more profoundly, about aligning models of reality. A harder, if not impossible task!"
 
I agree that it is immensely harder, and for different industries, with different goals, foundations, and requirements, impossible.  A bank, a poultry producer, an aircraft company, and an oil company will have very different models of reality.
 
A major reason why consistency is not a good thing:  it would stifle innovation.  It's impossible for people to think about new problems in new ways while remaining consistent with old ways of thinking, talking, and modeling.
 
John
 
 
From: "Neil McNaughton" <nei...@oilit.com>

??, 4 ???. 2023 ?. ? 22:12, Michael DeBellis <mdebe...@gmail.com>:

Alex Shkotin

unread,
Jan 7, 2023, 2:58:06 AM1/7/23
to ontolo...@googlegroups.com
Dear Matthew,

I am talking only about authoritative technological texts. Please, give me an example of an authoritative technological text "flawed through making assumptions that were convenient for their purpose."
For me we always should explicate a theoretical knowledge behind the text under formalization.

Regards,

Alex

пт, 6 янв. 2023 г. в 19:55, <dr.matt...@gmail.com>:

dr.matt...@gmail.com

unread,
Jan 7, 2023, 3:12:42 AM1/7/23
to ontolo...@googlegroups.com

Dear Alex,

Unfortunately that is the kind of text I am talking about. So I mean things like BIPM’s VIM.

Ravi Sharma

unread,
Jan 7, 2023, 4:10:34 AM1/7/23
to ontolo...@googlegroups.com
Mihai Nadin
The two bold parts in your sentence are confusing.
When you say referenced to meaning you are implying "in relation to some model or theory" and in the next sentence  you are saying "nor operations on the data". But to me the "operations on data" can be equivalent to reference to meaning.
Referenced to meaning data becomes information. Nothing else. Neither the data nor operations on such data reflect understanding of the phenomena the data represent.

Meaning and understanding are closely related!
Request your comments, it is also the inability of language to disambiguate different operations on data as in this case.
Thanks for your post that has evoked so much interest.
Regards.

Thanks.
Ravi
(Dr. Ravi Sharma, Ph.D. USA)
NASA Apollo Achievement Award
Chair, Ontology Summit 2022
Particle and Space Physics
Senior Enterprise Architect



Alex Shkotin

unread,
Jan 7, 2023, 5:17:48 AM1/7/23
to ontolo...@googlegroups.com
Dear Matthew,

I got it. And let me share the reference to the text [1] we had formalized (just a little) together with geologists. They did not mention any "flawed through making assumptions that were convenient for their purpose.". They just asked to formalize classification and vocabulary.
To say frankly if the BIPM asks to formalize VIM I do not see a problem at the beginning. Our task is to formalize text, not to criticize it. Well usually we found typos.
Anyway, formalization is a subtle work like cleaning definitions of terms, but for any vocabulary this is a unique task.

Thank you for your reference. "measurement science" [2] is a very important one to be formalized. Of course, when formalization is finished and we have axiomatic theory with established axioms, definitions and derivation rules; any of these parts may be criticized.

Regards,

Alex

[1] https://www.cambridge.org/gb/academic/subjects/earth-and-environmental-science/mineralogy-petrology-and-volcanology/igneous-rocks-classification-and-glossary-terms-recommendations-international-union-geological-sciences-subcommission-systematics-igneous-rocks-2nd-edition?format=AR&isbn=9780521619486
[2] https://iopscience.iop.org/article/10.1088/0026-1394/52/1/R1

сб, 7 янв. 2023 г. в 11:12, <dr.matt...@gmail.com>:

Azamat Abdoullaev

unread,
Jan 7, 2023, 8:11:40 AM1/7/23
to ontolo...@googlegroups.com
MN: 
"Data comes from measurement. Data is agnostic of theory. Referenced to meaning data becomes information. Nothing else. Neither the data nor operations on such data reflect understanding of the phenomena the data represent".

> Data is agnostic of theory.

 

MD: I don't think that's completely true. .. 

Or consider what you see the Sun do every day. Does it move across the sky or does the Earth's rotation make the position of the sun change? In the past our ancestors data said the Sun moved but now our theory tells us that the real data is the Earth's rotation causes the Sun to change position over time. 


I mostly agree with Michai, and his comments make sense, which somehow is missing the needful attention of our forum.

First, Data is a triple-faced thing, 

1. an aspect of things (quantity, quality, fact, relationship),  

2. a piece of information (the smallest units of factual information that can be used as a basis for calculation, reasoning, or discussion, analysis, presentation, visualization); organized in data hierarchy, a character (bit and byte), field, record, file and database 

3. some sorts of measurement, or statistics, levels of measurement

First of all, data is an ontological category, as facts or states of affairs. 

We might say "the world is the totality of data/facts, not things".

As such, Data is orthogonal, or independent of, not only to theories, but also to all human mentality/epistemology: knowledge, values, opinion, beliefs or theories.

We have the data universe of the whole world partitioned into the finite number of datasets (categories and classes, kinds or types) of an innumerable number of data items/elements/points, as instances, individuals, cases, or facts (tokens) 

For Data to represent the things in the world, the World Data Ontology was proposed to complete the data science and engineering, statistical AI and ML/DL.

Broadly speaking, all data falls into one or more of five categories: nominal, ordinal, interval, ratio, and number, going as the levels or scales of measurements. 

  • Nominal data, the simplest data type, classifying (or naming) data without suggesting any implied relationship between those data, as basic ontological categories, countries or species of animals.
  • Ordinal data, classifying data but it introduces the concept of ranking, as all hierarchical ontological categories or the Lichert scale or ‘slow’, ‘medium’, ‘fast’
  • Interval data, both classifying and ranking data (like ordinal data) but introduces continuous measurements, as the time of day or temperature measured on either the Celsius and Fahrenheit scale. 
  • Ratio data, it classifies and ranks data, and uses measured, continuous intervals, just like interval data. But, unlike interval data, ratio data has a true zero, an absolute, below which there are no meaningful values. All physical quantities, as mass, speed, age, or weight are examples  of the RD
  • Numbers, they count, measure, and label, number sets or number systems, N-natural numbers, Z-integers, Q-rationals, R-reals and C-complex numbers  image.png

The first four classes were introduced in 1946 by the psychologist Stanley Smith Stevens, widely used in sciences and engineering and data analytics. It is a hierarchical scale, each level builds on the one that comes before it.

It is crucial that for intelligent machines, the Data Universe Pyramid is replacing the DIKW pyramid, the DIKW hierarchy, wisdom hierarchy, knowledge hierarchy, information hierarchy, information pyramid, or data hierarchy, the Data, Information, Knowledge, Wisdom.

https://www.linkedin.com/pulse/world-data-ontology-science-ai-ml-deep-learning-graph-abdoullaev/


To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROTVejE9GNfpQdVP27f_FBsBVZsUrXrGZq7Y9RuHovd%3DTA%40mail.gmail.com.

Alex Shkotin

unread,
Jan 7, 2023, 11:52:51 AM1/7/23
to ontolo...@googlegroups.com
Azamat,

Very interesting. There is also a term "value" which for me is more "philosophical". 
And right now let me add interesting findings from my favorite m-w.com:
The first known use of data was in 1646 see
The first known use of information was in the 14th century see
value see
image.png
And about numbers, there are also quaternions and p-adic ones.

Alex

сб, 7 янв. 2023 г. в 16:11, Azamat Abdoullaev <ontop...@gmail.com>:
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAKK1bf_JmcPWBjJK-HUwUVbcUYgY2gN0XLEq31i9aBHzVHvUdA%40mail.gmail.com.

Azamat Abdoullaev

unread,
Jan 7, 2023, 1:50:20 PM1/7/23
to ontolo...@googlegroups.com
Alex,
You are talking about datum/observation, an individual value in a collection of data. Indeed, it is critically important.
In all, there is a data unit as one entity in the population,  a data item as a characteristic (or attribute or variable) of a data unit which is measured or counted, datum, an observation as an occurrence of a specific data item that is recorded about a data unit, a dataset as a complete collection of all observations or measurements, and the data universe as a total collection of all datasets.

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROQERM4hrG2eEZsPj%3DqQ3QH7T3prmzR495uR6QrY7ESEDA%40mail.gmail.com.

Ravi Sharma

unread,
Jan 7, 2023, 11:15:20 PM1/7/23
to ontolo...@googlegroups.com
Azamat
Your Image is very illustrative, It is sa
image.png
ying a lot in one diagram.
 
Thanks.
Ravi
(Dr. Ravi Sharma, Ph.D. USA)
NASA Apollo Achievement Award
Chair, Ontology Summit 2022
Particle and Space Physics
Senior Enterprise Architect


To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAKK1bf-um-m0g2G9SwAYF0fEgGDiQzTx-ms23s6b%3DAoUA4_18A%40mail.gmail.com.

Alex Shkotin

unread,
Jan 8, 2023, 2:54:03 AM1/8/23
to ontolo...@googlegroups.com
Azamat,

My point is just about the fact that the word "information" is known two centuries earlier than the word "data". 
Very interesting. And the word "data" first known use is in 1646. Very interesting #2.

Alex

сб, 7 янв. 2023 г. в 21:50, Azamat Abdoullaev <ontop...@gmail.com>:

Azamat Abdoullaev

unread,
Jan 8, 2023, 9:53:22 AM1/8/23
to ontolo...@googlegroups.com
Alex,
Dip deeper, down to the times of the ancient Egyptian astronomers or  Islamic Golden Age. It is when modern science was established (i.e. from the 8th century to the 14th century).
The history of data and information is the history of knowledge, starting from a simple counting. The first use of data goes back to 19,000 BC when our Palaeolithic ancestors used the bone tools to perform simple calculations.
You deal with data whenever collecting information such as facts or numbers by observations or measurements, research or analysis, to conclude about something or make decisions.

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROT9EKgjUYRW_oeHjtuQCfMh7UCDFq5zymjDaoUNosOhRg%40mail.gmail.com.

Alex Shkotin

unread,
Jan 8, 2023, 11:14:03 AM1/8/23
to ontolo...@googlegroups.com
Azamat,

Exactly. And Arabs got 1..9 from Sanskrit :-)


вс, 8 янв. 2023 г. в 17:53, Azamat Abdoullaev <ontop...@gmail.com>:
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAKK1bf-hXtixbTRDpRt49zfx5%2BTgZeBS8BPfFH00-z_X8Oq--w%40mail.gmail.com.

Nadin, Mihai

unread,
Jan 8, 2023, 11:29:47 AM1/8/23
to ontolo...@googlegroups.com

Data and calculations are two different things.

 

MN

Azamat Abdoullaev

unread,
Jan 8, 2023, 12:44:23 PM1/8/23
to ontolo...@googlegroups.com
They complement each other, as data and algorithms. 
The first one was data, then calculations, from arithmetical operations to electronic computations to complex logical reasoning to machine inferences..
It seems, the ancient people started from an elementary numbering, matching the number of marks with the number of their sheep. 
When and how they learnt the basic arithmetic operations, it is a big question. For it demanded a symbolic mind capable of thought experiments and language data communication. 
There is a historian, Harari, who suggested the idea of cognitive revolution due to genetic mutations or an alien intelligence intervention, some 70K-30K years ago.  

Nadin, Mihai

unread,
Jan 8, 2023, 4:44:42 PM1/8/23
to ontolo...@googlegroups.com

Dear and respected Ravi Sharma,

Dear and respected colleagues,

Data—outcome of measurement. As such, the data (stones, knots on a whip, scratches,  and eventually numbers, i.e. symbolic representations) is a representation. It constitutes the re-presentation, i.e. the substitute for whatever is measured. No representation is complete. You can operate on data: move around the stones (it is called calculus!—comes from literally moving stones), or operate on numbers.

The meaning—relates to the function. In other words, the what for of the measurement. The pragmatic. Usually this pragmatic level is reduced to semantics—what does it mean (such as in dictionaries).

Models come in place not when we measure, but when we want to use the data in order to predict.

Language is NOT the outcome of measurement. It is one form through which interactions take place.

Final note: those who quote Wikipedia—it is a natural “chatGPT”,  before the artificial neural network was conceived and trained on massive amounts of data. But so was Wikipedia. Often Wikipedia  “answers” in a convincing manner—when the data used for training is right. But it can, like the chat GPT produce non-sense.

Plus: there are editors—not on chat GPT, though. One of my students in Europe was an editor for Wikipedia. The individual was upset with his grade in my class…I let you smile at the rest…Jimmy Wales corrected by hand a less than pleasant characterization of some of my work. Lucky me, I was not yet participating in the ontolog-forum.

Best wishes.

 

Mihai Nadin

Ravi Sharma

unread,
Jan 9, 2023, 3:27:58 AM1/9/23
to ontolo...@googlegroups.com
Mihai Nadin and other colleagues
Many thanks for response
So now comes a question, if we do not have a model (or idea about) what we are measuring or collecting data, we can still collect items such as boulders in a riverbed, but we need some idea for example that we want to see erosion of land and a model etc.
Thus I just wanted to mention that there is meaning or model when data are related to information (of context or relevance or value!).
The second point was that if you understand the topic for which you relate to the data then you can express the value of information and may use languages also for expressing them.
I think after your explanation, our views are convergent or at least not divergent!
Regards,
Thanks.
Ravi
(Dr. Ravi Sharma, Ph.D. USA)
NASA Apollo Achievement Award
Chair, Ontology Summit 2022
Particle and Space Physics
Senior Enterprise Architect


Neil McNaughton

unread,
Jan 9, 2023, 4:11:24 AM1/9/23
to ontolo...@googlegroups.com

>> if we do not have a model (or idea about) what we are measuring or collecting data, we can still collect items such as boulders in a riverbed

 

I don’t think this is true. Collecting boulders in a riverbed needs a model. What minimum/maximum size constitutes a ‘boulder’? How and where should the survey be conducted? A model (braided stream, meander …) would imply a measurement strategy. You need a model first to test with measurement.

 

My 2cents

Neil   

Ravi Sharma

unread,
Jan 9, 2023, 4:47:55 AM1/9/23
to ontolo...@googlegroups.com
Neil
Yes that is the point, the boulders may or may not be there till we have a purpose or context to learn about then, there is still merit to the puzzle of how the boulders got there and then one needs the knowledge of the processes that make them.
Hence the data, information and model disambiguation is often difficult. 
Therefore hats off for the Vocabulary and Terms experts who manage domain overlaps etc.
John Sowa and others have used logic names for the intention, etc.
Thanks.
Ravi
(Dr. Ravi Sharma, Ph.D. USA)
NASA Apollo Achievement Award
Chair, Ontology Summit 2022
Particle and Space Physics
Senior Enterprise Architect


Nadin, Mihai

unread,
Jan 9, 2023, 7:10:27 PM1/9/23
to ontolo...@googlegroups.com

Dear and respected Neil McNaughton,

You seem to confuse model with metric. Your measuring stick is not a model. You measure in order to re-present something. Present again—for the purpose of describing in an effective manner change. The model is informed by measurement.

 

And on this note I shall leave the subject alone.

Best wishes.

 

Mihai Nadin

Michael DeBellis

unread,
Jan 9, 2023, 7:49:26 PM1/9/23
to ontolo...@googlegroups.com
You seem to confuse model with metric. Your measuring stick is not a model. 

I think you've managed to hit the core issue where we disagree. Because I think your measuring stick is indeed a model. It's not a very complex model but it is still a model. It uses the rational numbers with base 10. That in itself is a model. As modern humans we take it for granted but the invention of 0 and formats such as base 10 were a fundamental advance in ancient mathematics that enabled more complex math. It also uses some measurement model such as meters, centimeters, kilometers, etc. 

And that is for something as trivial as distance. Consider measuring velocity: distance per time or acceleration: distance per time squared. These are based on calculus and the fact that the derivative of velocity gives you acceleration. And when you get into measuring quantum events such as measuring the momentum or position of an electron traveling through a beam splitter there is quite a lot of model involved in the data. Or for something less complex and elusive than quantum theory consider just about any data in any modern scientific experiment: you are going to describe your results using terms like mean, standard deviation, p values, etc. Again, those are (statistical) models. Just to be clear: I'm not saying that there is no difference between data and theory. I'm just saying that the boundary isn't simple and there is some theory or model involved in just about any type of data. 

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/1BkY97l7Rtg/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/BL3PR01MB6897E352BA6C58C307EE1445DAFF9%40BL3PR01MB6897.prod.exchangelabs.com.

John F Sowa

unread,
Jan 9, 2023, 11:34:52 PM1/9/23
to ontolo...@googlegroups.com
Fundamental principle:  If you have a word or phrase for some kind of thing, you have a mental model for things of that kind..
 
If you know how to define a formal model in some way, you can map your mental model to that kind of formal model.
 
Re boulders:  If you don''t have a word for boulder, you are less likely to pay attention to them if you see them.
 
But if one of them impresses you for some reason ()it is pretty, unusually large, an obstacle on the road,...) you are likely to remember that encounter with a new kind of mental model.  Then you might invent a new phrase for it, such as "large round rock",  That  phrase indicates that you have formed a mental model, which will enable you to classify new things that may be similar,
 
John.
 
 

Ravi Sharma

unread,
Jan 10, 2023, 1:56:48 AM1/10/23
to ontolo...@googlegroups.com
John
Thanks for clarity, \
I would say any expressive notion or attribute even if not word for rock in a dialect.
Regards
Thanks.
Ravi
(Dr. Ravi Sharma, Ph.D. USA)
NASA Apollo Achievement Award
Chair, Ontology Summit 2022
Particle and Space Physics
Senior Enterprise Architect


--
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/cb67681cf7164590adb888f46a75edf1%40bestweb.net.

John F Sowa

unread,
Jan 10, 2023, 11:50:12 PM1/10/23
to ontolo...@googlegroups.com
Ravi,
 
Yes.  Anybody who can express  any comment about something has a mental model (or experience or feeling or whatever) about that thing and any other things that the person classifies as similar.
 
That model or experience may be vague.  Any expression or response about it is a sign that classifies it in some way.  The words of a natural language force it into pigeon holes that are supported by that language.  But  NLs allow quite a broad range of senses for their words.  Formal languages force the expressions into a narrower range but that narrow range, which may seem to be more precise might be precise in a way that is less accurate than the vague meaning in some natural language.
 
Summary:  I am in favor of using formal logics for precise reasoning.  But I agree with Whitehead and other logicians who admit that an attempt at formalizing something you don't really understand can be good or bad.  It's good if you are willing to throw it away when you recognize iits  limitations.  But it's bad if you don't admit its flaws.
 
John
From: "Ravi Sharma" <drravi...@gmail.com>

Ravi Sharma

unread,
Jan 11, 2023, 11:06:47 PM1/11/23
to ontolo...@googlegroups.com
John
Excellent, 
Yes we are continuously refining the concepts models in mind trying to match data-observations-experience that fit the models.
Logic is the most profound way and yes if not applied correctly to the model, it can give incorrect results.
Regards and appreciate the brief note.
Thanks.
Ravi
(Dr. Ravi Sharma, Ph.D. USA)
NASA Apollo Achievement Award
Chair, Ontology Summit 2022
Particle and Space Physics
Senior Enterprise Architect


--
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/ef82656792d44c30bf6390849ad660ae%40bestweb.net.
Reply all
Reply to author
Forward
0 new messages