nice free talk

79 views
Skip to first unread message

Alex Shkotin

unread,
Mar 21, 2019, 10:46:34 AM3/21/19
to ontolog-forum
Hi All,

we have got it occasionally, but maybe it's a good idea to have free talk for ex. one per month. 
It's a pity we do not have a record.

Thank you, David, for your attention! You mentioned Wirth+Pascal and my response was that we may look at any type (record...) in a language like Pascal as a hierarchical schema from DB point of view. And we may look at the DB schema as RDFS one and transform it into DL isomorphically. But usually, this is not a good idea, as we need to transform it logically.

There were other interesting topics touched: category theory, "ages", CL, program verification...

I lowered my hand because I have forgotten what I'd like to say: axioms, definitions, and finite models for any kind of science and technology.

And the last but not least, when we have the math model of part of Reality, some attributes have very special meaning - these are input parameters for material algorithms to find out this particular mostly unique object. Without this kind of reference to reality, we have just statistics.

Alex

David Whitten

unread,
Mar 21, 2019, 4:16:44 PM3/21/19
to ontolog-forum
I really appreciate your insight as well Alex.
As I said on the call, I feel I gain every time I read one of your e-mails.
I may not always understand it, but I have some ideas for more research.

For those who were not there,
I was talking about Dr. Wirth and his work on the Pascal programming language.
When I studied it, there were some discussions about how do you compare
two variables which may or may not be equal.  Do we define the equality 
of a value based on the NAME of the datatype for each, or the STRUCTURE
of the datatype for each.

So if we have a variable P of type PCOUNTER and another variable K of type KCOUNTER
and we say names matter for equality
and we see code IF (P == K) 
then we can stop and say they are not equal
since we just see that PCOUNTER is a different name than KCOUNTER,

But if say structure matters,
and there is a definition that has PCOUNTER as INTEGER
and there is a definition that has KCOUNTER as INTEGER
then we can't just stop and say they are not equal
because we need to look at the variables P and K and see if they both
represent the same INTEGER value.

We didn't even get into where PCOUNTER is a subtype of KCOUNTER,
though Pascal as I recall had the ability to define a type as a range of values
like FROM 1 TO 100  (though I don't recall the syntax of how to define those types)

In ontologies, we have two ways of describing Categories/Classes

First, define by list where we have a list of the instances of the Categories/Classes
or a list of the subCategories/subClasses.

Second, define by rule where we have some statement we can use in a logical way
to test and determine if a particular potential instance is actually an instance of that
Category/Class.

I was asking if there were other ways where we used the Name of the Category/Class
and or the structure of instances of the Category/Class to determine  if they are equal.

Does this make sense?
Dave Whitten


--
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/CAFxxRORkOP%2B%3DuzH-tcnasA2dpLMOLXmbuMGE15vRPEdYnZJZnQ%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Alex Shkotin

unread,
Mar 22, 2019, 5:34:37 AM3/22/19
to ontolog-forum
Thank you, Dave. 

I am ready to clarify any misunderstanding and the best way for me is by email as I need to translate carefully Eng2Ru, Ru2Eng:-)

About your question: if two structures are equal we may use one of them to model another. In ontologic the identity is important. Consider we look at the object directly and at the same time through the mirror. When we merge two ontologies equality of structures is just precondition for identity. For me, "equal" is a logical and math term, and "identity" - ontological. But maybe you just need examples of "if there were other ways"?

Alex

чт, 21 мар. 2019 г. в 23:16, David Whitten <whi...@worldvista.org>:
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAH8N84y_yjHU7gZOGWteYftzPjNyj33R5_cp6_8L4Mk6ZfKqtw%40mail.gmail.com.

Matthew West

unread,
Mar 22, 2019, 6:48:44 AM3/22/19
to ontolo...@googlegroups.com

Dear Alex and Dave,

You need to be very careful about using structure for identity. ISO 10303 (STEP) developed a set of data structures (integrated resources) and then interpreted them in different ways for different purposes. Some people thought this would enable them to integrate the data, but of course the interpretation was as important as the structure, and you could not, at least not on the basis of the common structures.

Regards

Matthew West

Alex Shkotin

unread,
Mar 22, 2019, 7:31:43 AM3/22/19
to ontolog-forum
Exactly Metthew,

and for me, identity and identification are more about fingerprint, i.e. very special data structures.

Alex

пт, 22 мар. 2019 г. в 13:48, Matthew West <dr.matt...@gmail.com>:
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/000901d4e09c%24d0234780%247069d680%24%40gmail.com.

Alex Titov

unread,
Mar 22, 2019, 7:43:09 AM3/22/19
to ontolo...@googlegroups.com
In my personal opinion, an identification is a subset of classification relations, so that we have a restriction on the class - its cardinality is to be equal to one. Some consequences - there may not be an ‘absolute’ identification - only ‘relative’ - in some external context - and what can be treated as a thing identification in one context - might not be identification in the other context (where a singular cardinality property is violated)… 

Kind regards,
Alex

Alex Shkotin

unread,
Mar 22, 2019, 8:24:16 AM3/22/19
to ontolog-forum
Alex,

there are generally accepted meanings of this word and I like this one "an act of identifying" see https://www.merriam-webster.com/dictionary/identification

Alex

пт, 22 мар. 2019 г. в 14:43, 'Alex Titov' via ontolog-forum <ontolo...@googlegroups.com>:
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/fc9c6064-8086-4ecc-be95-fd8f627f98b2%40Spark.

Alex Titov

unread,
Mar 22, 2019, 9:22:20 AM3/22/19
to ontolo...@googlegroups.com
I appreciate that in some spatio-temporal extents, there may be some agreement between some things that probably can ‘agree’ on something… I also understand that sometimes I am forced to share those ‘agreements' (at least publicly), otherwise either me or my family may get some negative consequences…

but personally I think that a reference to a ‘general accepted meanings’ a bit strange… in the context of ontology discusions. A ‘general accepted meaning’ as a class of classes?

Kind regards,
Alex

Alex Shkotin

unread,
Mar 22, 2019, 9:52:49 AM3/22/19
to ontolog-forum
Alex,

time from the time I ask people here if ontology is an axiomatic theory plus finite model of it connected with a part of reality to model it. 
What do you think?

Alex

пт, 22 мар. 2019 г. в 16:22, 'Alex Titov' via ontolog-forum <ontolo...@googlegroups.com>:
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/d57c0656-d9fb-47ff-ace5-2d744e98d8c0%40Spark.

Matthew West

unread,
Mar 22, 2019, 12:43:32 PM3/22/19
to ontolo...@googlegroups.com

Dear Alex,

As it happens I am currently reading “Varieties of Reference” by Gareth Evans. What that suggests is that there is more than one way of referring to things. So far, I’m getting the kind of description you are saying and proper names where a particular name has been introduced as referring to a particular thing.

The book seems to be quite good, so I recommend it if you want to discuss/understand the related issues.

Regards

Matthew West

Alex Shkotin

unread,
Mar 22, 2019, 1:35:26 PM3/22/19
to ontolog-forum
Dear Matthew,

thank you very much for interesting reference:-) Maybe I begin with https://www.academia.edu/7162073/Review_of_The_Varieties_of_Reference_by_Gareth_Evans
And let me say that a way of reference defers from science to science and from technology to technology. It's just important to know this kind of attributes in an ontology as well as in DB.
For example, for rock samples, we have coordinates of excavation at first, but also where are they now and also reference to the expedition details.

Thank you!

Alex



пт, 22 мар. 2019 г. в 19:43, Matthew West <dr.matt...@gmail.com>:
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/006c01d4e0ce%245fe47700%241fad6500%24%40gmail.com.

Matthew West

unread,
Mar 22, 2019, 4:20:38 PM3/22/19
to ontolo...@googlegroups.com

Dear Alex,

From a cursory glance based on what I have read, that is a good summary. Good spot.

Regards

Matthew

Mike Bennett

unread,
Mar 22, 2019, 4:58:29 PM3/22/19
to ontolo...@googlegroups.com

We were talking about some of these issues on the fringes of the Object Management Group meetings this week, around the concept of a 'Digital Twin' and the issues of identifiers for things.

As I recall, Leibniz's Law stipulates that if two things have exactly the same characteristics / properties then they are the same thing. I have always been suspicious of this. It seems to me that there is something to identity qua identity, regardless of the characteristics one might use to detect it. The 'soul' of the thing, as it were. But I don't know any good theories or references that unpack that idea.

Mike

PS I'm glad you were able to have such a valuable discussion in the absence of my having been able to organize something on the day. It sounds like we should do that more often.


For more options, visit https://groups.google.com/d/optout.
-- 
Mike Bennett
Hypercube Limited
89 Worship Street, London EC2A 2BF
Tel 020 7917 9522  Mob. 07721 420 730
Twitter: @MikeHypercube
www.hypercube.co.uk

Paul Tyson

unread,
Mar 22, 2019, 9:28:15 PM3/22/19
to ontolo...@googlegroups.com
On Fri, 2019-03-22 at 16:58 -0400, Mike Bennett wrote:
> We were talking about some of these issues on the fringes of the
> Object Management Group meetings this week, around the concept of a
> 'Digital Twin' and the issues of identifiers for things.
>
>
> As I recall, Leibniz's Law stipulates that if two things have exactly
> the same characteristics / properties then they are the same thing. I
> have always been suspicious of this. It seems to me that there is
> something to identity qua identity, regardless of the characteristics
> one might use to detect it. The 'soul' of the thing, as it were. But I
> don't know any good theories or references that unpack that idea.

Most anything of a Thomist strain will work. Maritain, "Degrees of
Knowledge" is a good start. Veatch, "Intentional Logic", is a lengthy
critique of mathematical logic from a realist perspective, based largely
on a concept of identity that is foreign to western philosophical
idealism.

Regards,
--Paul
> > <drmatth...@gmail.com>:
> > email to ontolog-forum
> > +unsub...@googlegroups.com.
> > To view this discussion on
> > the web visit
> > https://groups.google.com/d/msgid/ontolog-forum/CAFxxRORkOP%2B%3DuzH-tcnasA2dpLMOLXmbuMGE15vRPEdYnZJZnQ%40mail.gmail.com.
> > For more options, visit
> > https://groups.google.com/d/optout.
> >
> > --
> > 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-forum
> > +unsub...@googlegroups.com.
> > emails from it, send an email to ontolog-forum
> > +unsub...@googlegroups.com.
> > To view this discussion on the web visit
> > https://groups.google.com/d/msgid/ontolog-forum/CAFxxRORZ%2BFJFaYn13PMzZKtMCrCVfGysDBMA2XEidH7RgRqGsA%40mail.gmail.com.
> > For more options, visit
> > https://groups.google.com/d/optout.
> >
> > --
> > 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-forum
> > +unsub...@googlegroups.com.
> > To view this discussion on the web visit
> > https://groups.google.com/d/msgid/ontolog-forum/fc9c6064-8086-4ecc-be95-fd8f627f98b2%40Spark.
> > For more options, visit https://groups.google.com/d/optout.
> >
> >
> > --
> > 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-forum
> > +unsub...@googlegroups.com.
> > To view this discussion on the web visit
> > https://groups.google.com/d/msgid/ontolog-forum/006c01d4e0ce
> > %245fe47700%241fad6500%24%40gmail.com.
> > For more options, visit https://groups.google.com/d/optout.
> >
> > --
> > 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-forum+unsubscribe@googlegroupscom.
> > To view this discussion on the web visit
> > https://groups.google.com/d/msgid/ontolog-forum/CAFxxROSj%
> > 2BmNg9gpCxQWQ8hY3GC5HSi%2B%2BR90X%2BDbB-z%2B8Es6w0A%
> > 40mail.gmail.com.
> > For more options, visit https://groups.google.com/d/optout
> >
> >
> > --
> > 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/004201d4e0ec%
> > 24b438ec00%241caac400%24%40gmail.com.
> > For more options, visit https://groups.google.com/d/optout.
> --
> Mike Bennett
> Hypercube Limited
> 89 Worship Street, London EC2A 2BF
> Tel 020 7917 9522 Mob. 07721 420 730
> Twitter: @MikeHypercube
> www.hypercube.co.uk
>
> --
> 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/5c78d8f6-08a7-2e4d-1eff-3e7b0ec42aa7%40hypercube.co.uk.

Matthew West

unread,
Mar 23, 2019, 4:50:08 AM3/23/19
to ontolo...@googlegroups.com

Dear Mike,

This is one of the things that 4D can simplify: If two things have the same spatio-temporal extent, they are the same thing. The main challenge is that requires maintaining some history of where something is/has been to maintain the confidence that what you are looking at is a state of the whole concerned.

Regards

Matthew

Alex Titov

unread,
Mar 23, 2019, 5:08:21 AM3/23/19
to ontolo...@googlegroups.com
> This is one of the things that 4D can simplify: If two things have the same spatio-temporal extent, they are the same thing.

Let’s say that I agree and share that postulate. But there may be plenty of other observers, who don’t agree…

The issue I don’t know how to solve in that situation - how one knows that two (or more) patio-temporal extents are the same? That has to be observed/measured somehow. Suppose I can make such observations/measurements. What happens if another observer gets another results? In addition, my observation is spatio-temporal related as well and observations from one point and time moment may be different from another.

That makes me thinking that 4D is not enough. Somehow the observers provide influence on observations and are to be included into the model. But I don’t feel that I have enough knowledge, skills, experience to work on this.

Kind regards,
Alex Titov
https://www.linkedin.com/in/alextitov

Alex Shkotin

unread,
Mar 23, 2019, 6:43:00 AM3/23/19
to ontolog-forum
Dear All,

an act of identification varies from science to science and from technology to technology. Have a look https://en.wikipedia.org/wiki/Identification
IMHO It's interesting to formalize each separately.
And it is based on another subtle process - measurement.

By the way, I have only asked Dave if he had written about equality or identity:-)

Alex


сб, 23 мар. 2019 г. в 11:50, Matthew West <dr.matt...@gmail.com>:
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/002c01d4e155%24692a8110%243b7f8330%24%40gmail.com.

Matthew West

unread,
Mar 23, 2019, 6:53:46 AM3/23/19
to ontolo...@googlegroups.com

Dear Alex,

I’m only saying it helps to know what you are trying to establish. If you do not even have that, then you have no more than opinions. If you do have that, then people who are seeking the truth can usually work it out and agree. We do not usually have a problem finding our car in the car park.

Regards

Matthew

Alex Shkotin

unread,
Mar 23, 2019, 7:59:38 AM3/23/19
to ontolog-forum
Dear Matthew,

please identify which one Alex do you refer to:-)
or give the identification procedure.

AlexS

сб, 23 мар. 2019 г. в 13:53, Matthew West <dr.matt...@gmail.com>:
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/002001d4e166%24afc69d50%240f53d7f0%24%40gmail.com.

David Whitten

unread,
Mar 23, 2019, 9:53:12 AM3/23/19
to ontolog-forum
I"m enjoying this discussion.

If I say that I have a 4-D spatial temporal entity.
To me, this means that I can take a "double location" pair (spacial location, temporal location) 
and see if that pair coincides with the entity.

If I give this a pair a name, does this make that "double location" into a possible property value 
of that named property for the entity?

Since I don't have infinite capacity in storage for my equality comparison operation to use,
nor do I have infinite capacity to describe the entity, I would assume that for any particular application,
or for any particular granularity,  I would limit the number of such "double location" pairs used by a comparison
in some way.  

Perhaps that might even be a good way of defining "granularity" as I struggle with finding a good definition.

 As referenced by Mike Bennett, Leibniz's Law stipulates that if two things have exactly the same 
characteristics / properties then they are the same thing.  
Do you think Leibniz was thinking about a definition of property similar to what I mentioned here?

As to identity versus identification, I'm not clear enough of the distinction you are making to be
able to have a strong opinion.

Dave Whitten

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/4204c596-573d-4f91-b531-42fb8a62fcab%40Spark.

Matthew West

unread,
Mar 23, 2019, 11:12:34 AM3/23/19
to ontolo...@googlegroups.com

Dear Alex,

The one whose email I replied to.

Matthew West

unread,
Mar 24, 2019, 4:50:07 AM3/24/19
to ontolo...@googlegroups.com

Dear David,

Let me try to illustrate the difference between identity and identification with a story (apocryphal).

I was house siting for a friend whilst he was abroad for a few months and he lent me his car with registration no ABC1234. Unfortunately, I had an accident and wrote the car off. I was desperate to avoid the embarrassment of having to own up to what I had done, so I scoured the second hand car ads for one the same. Fortunately, it was a common model, spec and mileage, so I was able to find one just the same. However, my friend was going to notice it did not have the right registration no. Fortunately, that is not too hard, you can re-register a car without too much difficulty, and with a little paper work I was able to do that. So, when my friend came home everything appeared as it should be.

Did my friend have the same car that he left me with?

Regards

Matthew West

Alex Shkotin

unread,
Mar 24, 2019, 7:31:10 AM3/24/19
to ontolog-forum
It's as old as Theseus ship :-)
https://en.wikipedia.org/wiki/Ship_of_Theseus

вс, 24 мар. 2019 г. в 11:50, Matthew West <dr.matt...@gmail.com>:
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/005801d4e21e%2492e38eb0%24b8aaac10%24%40gmail.com.

Matthew West

unread,
Mar 24, 2019, 11:27:14 AM3/24/19
to ontolog-forum
The ship of Theseus is a more difficult case. Here I was just trying to distinguish between identity (there are clearly two cars not one) and identification (a "name" that is given to something.
Regards 
Matthew West 

To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/CAFxxROQGpgzA5dYuboe50%2B-KOht9hEWynffJ2-Pn6y9E6jxbAw%40mail.gmail.com.

John Bottoms

unread,
Mar 24, 2019, 12:44:35 PM3/24/19
to ontolo...@googlegroups.com
"The Ship of Theseus" is not too difficult.

There are 3 cases:

1. Theseus had the ship built and the captain named the ship. That was The Ship of Theseus.

2. The carpenter replaced a board and the ship was still called The Ship of Theseus. Unfortunately, the captain had done the repair on the island of Amatoria where the children of the school name each new ship. The captain was captured and executed for violating the law.

3. The ship's carpenter collected all boards from the original ship and built a new The Ship of Theseus. Just before the launch he was captured and executed. Everyone knows the Harbormaster names ships. This is true regardless of whether it is the Queen, or a person of note who does the christening.

So, in these cases, the answer is: "Find out who is the naming authority", then follow the rules or regulations for naming. YMMV. {This question was put to me by the U.S. Patent office.]

-John Bottoms
 FirstStar Systems
 Concord, MA USA
 T: 978-505-9878

Matthew West

unread,
Mar 24, 2019, 4:09:20 PM3/24/19
to ontolo...@googlegroups.com

Dear John,

That is not the usual way the story is told. Mostly they are similar to the Wikipedia version:

“First, suppose that the famous ship sailed by the hero Theseus in a great battle has been kept in a harbour as a museum piece. As the years go by some of the wooden parts begin to rot and are replaced by new ones. After a century or so, all of the parts have been replaced. Is the "restored" ship still the same object as the original?

Second, suppose that each of the removed pieces were stored in a warehouse, and after the century, technology develops to cure their rotting and enable them to be put back together to make a ship. Is this "reconstructed" ship the original ship? And if so, is the restored ship in the harbour still the original ship too?[1]”

The issue is to explore identity (and not compare identity with identification). The question is about what are the limits of change for something still to be the same thing. If a cat looses its tail, we don’t think of it as a different cat, and generally, as long as something does not undergo replacement of all or most of its parts at once we consider it the same thing (there are exceptions to this rule though).

Regards

Matthew West

Nadin, Mihai

unread,
Mar 24, 2019, 4:21:30 PM3/24/19
to ontolo...@googlegroups.com

Dear and respected colleagues,

Identity makes sense only at the pragmatic level. What you—those who brought up examples-- described were syntactic aspects and at most some semantic identifiers. The identity is related to the purpose, to the actions through which whatever entity is identified and recognized as such.

My stubborn suggestion that foundations (in this case of semiotics) should not be ignored comes from the realization that best intended examples do not substitute for a more comprehensive understanding of what we want to define.

 

Best wishes.

 

Mihai Nadin

David Whitten

unread,
Mar 24, 2019, 7:37:56 PM3/24/19
to ontolog-forum
Mihai,
your emphasis on the semiotic foundations sounds like you have an opinion about this.

I think you are equating the various parts of these various ships we are discussing as purely of a syntactic
nature.  You also have mentioned some identifiers that are mostly semantic.

First, what are examples of identifiers that are something other than mostly semantic ?
Secondly, are there examples of identifiers that are completely syntactic ?
Thirdly, what are semiotic identifiers ? Do they exist ? If so, how do I recognize them?

I assume we are using a Car or a Ship as our topic of discussion because we can hold to the idea that
there is a pile of parts, perhaps sorted by internal material then shape and then by size
Next, there is the combination of that pile of parts that constitutes a working example of some category, like a ship or a car.
Are we in agreement that these two are different?  That the configuration of the parts is just as much a part
of the object as the parts that make it up ?
I think mereology is involved in this somehow.

I think the Car of MatthewWest is based on the idea that if you have a working example, you can have
two working examples. Especially in a digital world, where there aren't things like individual parts having different
wear patterns or internal inconsistencies, etc.

The Car of MatthewWest example also implies that a car is recognizable by the function it performs, (identity)
and the name given to it (identification).  
I think Matthew may be saying that when some object provides all the same functionality with parts that provide
all the same capabilities, and the object has the same identification, then 

John Bottoms seems to bring up several interactions between identity and identification.

The original assemblage of parts configured a particular way can be given a name such as 
JohnBottomsObjectVersion1

Alternately, the name is attached to the configuration and the exact parts which constitute that object, so if
any of the parts (which have their own identity) are replaced with another part, then there must be a new name 
perhaps JohnBottomsObjectVersion2

Finally, we might have JohnBottomsObjectVersion0 made of all of the same parts and which may be configured 
together in the same way as the object but with a different history'provenance.
John Bottoms seems to say that if the naming operation itself is performed by a different person at a different time, 
and without an official mark by the Harbormaster then the name still isn't the same, 
as the naming authority also defines some part of a name.

There also is Museum-Ship-of-MatthewWest.  When placed in the Museum, all the parts worked together, performing
a particular function.  Over time, the parts are replaced and in the context of the Ship in the Museum, they appear
to still provide the same function.  Perhaps the Anchor is made of Tin instead of Iron, but it looks the same, and
in the museum, the difference of weight and of material isn't important as far as its active context.

Don't they do that with Dinosaur skeletons in museums that have the right shapes and configuration but rarely
are constituted of the bones that were dug up in the original discovery site ?

The Re-Constituted-Warehouse-Ship-of-MatthewWest seems to be more "real" than the the 
Museum-Ship-of-MatthewWest but perhaps it actually is worse because it doesn't function in the context of the Museum
as well as the Museum-Ship-of-MatthewWest does.

Have I captured all the varieties of our conversation up to now ?

David Whitten


To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/90df69c0b3c941efb2df88164841eb59%40utdallas.edu.

Nadin, Mihai

unread,
Mar 24, 2019, 7:45:08 PM3/24/19
to ontolo...@googlegroups.com

Just to keep it simple: what about a digital reproduction? VR—you can walk through, etc.

Pragmatics means: what for? Car, ship, etc. The first thing manufacturers do is to reverse engineer a competing product. In software the same holds true.

Short of understanding that identity is pragmatically defined and tested we will not make much progress. The meaning of whatever identity you are focused on is disclosed in its functioning, doing what it is supposed to do.

 

I hope my short answer does not convey some impatience, but rather the desire to keep things simple. If you want a more detailed reaction, I can write to you.

 

Mihai Nadin

Alex Shkotin

unread,
Mar 25, 2019, 7:04:03 AM3/25/19
to ontolog-forum
John,

my 5 pence to your point: we need a definition - what does it mean to be ShoTh. And in this particular case, it's just an agreement.
We create definitions for concepts (unary predicates) and relations (binary ones) and in practice, these are agreements or recommendations as IUGS's rock types.

Alex

вс, 24 мар. 2019 г. в 19:44, John Bottoms <jo...@firststarsystems.com>:
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/f83dac30-9cc4-0b4c-af99-b92db70bc7b9%40firststarsystems.com.

John Bottoms

unread,
Mar 25, 2019, 8:23:46 AM3/25/19
to ontolo...@googlegroups.com
Alex,

Yes, it is an agreement, but of what type, and what is the framework for that agreement. It is also a model and we know what George Box said about models; "All models are wrong but some are useful." He goes on to imply that further elaboration is not helpful, presumably because starting from a "wrong" model leads one further astray.

i feel there is a shortcoming in how we define "ontology". Our working definition is, according to Gruber, "An ontology is a [shared] specification of a conceptualization." He goes on to include a "shared vocabulary".

in the case of the identification process we need to take a wider view. When I visited the CIA headquarters I was informed of their identification process. It involves 1.) Something you have, 2. Something you know and 3.) Something you are, or something you can do. This is from an internally published administrative security manual, and the procedures are specified within and are required by security personnel.

So the system to be modeled in this case includes several components:

  o A Security Manual identifying procedures and identification elements such as keys or image facsimile such as photo, fingerprint, or bio-metric information. This entails the production of those items.

  o A Security Administrator who insures that each user's identification elements reflect the requirements of the Security Manual.

  o The person or actor to be identified as a person with an acceptable risk as identified in the Security Manual.

  o A Security Agent that has been educated in the practice of using those presented identification elements to approve the access for the person with security access approval.

  o We might want to include the hypothesized mental models of the people or agents involved, but that depends on who is using this model information.

  o Finally, back somewhere in the administration process we entail the salaries and the business practices that require these to be applied as honestly as possible within common business practices.  In a company or corporation these are referred to as "Principles Documents", and are part of the employment contract.

So, the part of this ontology definition that I take issue with is that the meta-definition of the model does not enforce the need for inherited properties of the parent or entailed contextual entities.

More to your point, Theseus' (himself) authority to name the ship, as Captain of the ship, derives from his control of the purse-strings of the enterprise; he has responsibility and authority. If you do not adhere to his definition of "the ship", you are free to withdraw, and sign up on another ship. He may also draw on the other resources of the crew and the threat of the brig, but these can be viewed as components of a commonly accepted business context.

Theseus' authority extends to others who recognize him as the captain of the ship. A term that is being used for his influence derives from Teilhard de Chardin's view of spheres of influence, though the current term is "capsule". Capsules contain the mixture of responsibility and authority. They arose from an incident in which a female player in a game was surrounded by disembodied knives above her head. She was physically terrified and she left the game, taking with her several other ]paying] friends who she had convinced to joint the game. Current capsules allow penetration of swords for participants who have a formal agreement to duel.

My students and I have been developing the infrastructures for Smart-Capsules. This is needed to allow elements, such as a Bitcoin, to pass through the capsule so commerce can be transacted.


-John Bottoms
 FirstStar Systems
 Concord, MA


John F Sowa

unread,
Mar 26, 2019, 10:39:16 AM3/26/19
to ontolo...@googlegroups.com
On 3/25/2019 7:03 AM, Alex Shkotin wrote:
> So, in these cases, the answer is: "Find out who is the naming
> authority", then follow the rules or regulations for naming.
> (This question was put to me by the U.S. Patent office.]

I agree. Names are arbitrary choices, but there are useful
guidelines for minimizing confusion.

For example, confusion in drug names could cause death if a
pharmacist misreads a doctor's handwriting or a nurse mishears
a request. Therefore, the FDA has very strict rules for
choosing names, especially for drugs that could be dangerous
if carelessly taken (or not taken when required).

As another example, the US Army allowed a soldier to request
a replacement part if some part of a rifle was broken. But
they discovered that some soldiers would keep asking for enough
new parts to assemble a complete rifle.

Their solution: They put a serial number on the stock and
made it the only part that could not be replaced. If it was
broken, the soldier had to request an entirely new rifle.

John

Alex Shkotin

unread,
Mar 26, 2019, 11:06:20 AM3/26/19
to ontolog-forum
for JFS from Alex, 

actually, you have cited John Bottoms letter. Here is a full text of him:

The Ship of Theseus" is not too difficult.

There are 3 cases:

1. Theseus had the ship built and the captain named the ship. That was The Ship of Theseus.

2. The carpenter replaced a board and the ship was still called The Ship of Theseus. Unfortunately, the captain had done the repair on the island of Amatoria where the children of the school name each new ship. The captain was captured and executed for violating the law.

3. The ship's carpenter collected all boards from the original ship and built a new The Ship of Theseus. Just before the launch he was captured and executed. Everyone knows the Harbormaster names ships. This is true regardless of whether it is the Queen, or a person of note who does the christening.

So, in these cases, the answer is: "Find out who is the naming authority", then follow the rules or regulations for naming. YMMV. {This question was put to me by the U.S. Patent office.]


-John Bottoms
 FirstStar Systems
 Concord, MA USA
 T: 978-505-9878"

вт, 26 мар. 2019 г. в 17:39, John F Sowa <so...@bestweb.net>:
--
All contributions to this forum are covered by an open-source license.
For information about the wiki, the license, and how to subscribe or
unsubscribe to the forum, see http://ontologforum.org/info/
---
You received this message because you are subscribed to the Google Groups "ontolog-forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ontolog-foru...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ontolog-forum/2bdcdcab-1981-c2a7-0e04-39416209ceda%40bestweb.net.

Alex Titov

unread,
Mar 26, 2019, 12:07:41 PM3/26/19
to ontolo...@googlegroups.com
Sorry for delay in comments… I thought a bit about the difference between an ‘identity’ and ‘identification’… Well, and I don’t see (or feel) any difference (or any necessity in identity as a separate concept/idea). 

First of all - at the present moment I don’t believe (or can observe) any ’soul’ or any ‘eternal identity’ of any thing. I really mean _any_ in space and time. I cannot proof or disprove that, so that becomes an arbitrary ‘axiom’ for me.

And based on that ‘axiom’ I understand the meaning of ‘identity’ as a short name for an informational  manifold of observable characteristics of a thing, that describes a classification of that thing in a such a way, that the given thing is considered as a member of a class with cardinality of one.

So, I don’t know about any ’natural’ identity of any thing. Such thing does not exist in my personal ontology.

I would guess that a subsequent conversation about identity may lead to 'On What There Is’ by Quine… so I should reread it before continue...

Coming to examples - car and ship - I don’t see any issue with identification in both of those examples. 

In my paradigm - 2 things are 1 (or the same) if an observer has information (or beliefs or orders) which (in the observer’s context, space and time) allows him (or her, or it) to make a conclusion (or to make a reasonable assumption) that the spatio-temporal extent of both things is the same.

So, if in the observer's (information) context, the only one characteristic of a car is its registration number, and the identification is build only from it - it does not matter if all parts, components, molecules, atoms, of information is changed. The next time the registration number is observed and the value is not changed - the conclusion - it is the same car.

If all boards of the ship are changed or the name is changed - it does not matter. The identification depends only on how the ’singleton’ classification is arranged. For one observer the ’singleton’ classification may be build only from the name characteristic. As soon as the observed characteristic is different - the ship is different. For another observer - some specific (or all) boards. If the value of characteristic is different - the whole ship is different. For the third observer it may be that both characteristic are irrelevant in her/his  ’singleton’ classification, and her/his identification is based on something else… and son on.

Kind regards,
Alex

Matthew West

unread,
Mar 26, 2019, 12:39:37 PM3/26/19
to ontolo...@googlegroups.com

Dear Alex,

Identity is about whether what might be two things are actually the same. The considerations are philosophical. For example, when the cat Tibbs looses its tail, does he cease to exist altogether because he is not the same as he was and there is now some new object TibbsMinusTail? We usually say that this is still Tibbs, he has just lost a part that was not critical to his IDENTITY. As a result, we can continue to use the same name (Tibbs) for him IDENTIFICATION.

Regards

Matthew

 

From: 'Alex Titov' via ontolog-forum <ontolo...@googlegroups.com>
Sent: 26 March 2019 16:07
To: ontolo...@googlegroups.com
Subject: Re: [ontolog-forum] nice free talk

 

Sorry for delay in comments… I thought a bit about the difference between an ‘identity’ and ‘identification’… Well, and I don’t see (or feel) any difference (or any necessity in identity as a separate concept/idea). 

 

Kind regards,
Alex

 

--

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

Alex Titov

unread,
Mar 26, 2019, 1:10:56 PM3/26/19
to ontolo...@googlegroups.com
I appreciate your beliefs.  But probably don’t share.

From my point of view, the cat may lose all molecules it consist of (or die and all its atoms being reused in building some house, or completely disassembled into smaller particles), but still be the same cat (the thing can cease to exist in the 4D world of information only when the information is erased or the observer has no energy to maintain information) - that depends on the observer (and available information, and context) only, rather than the mythical (from my point of view) ‘identity’.

One observer may consider TibbsMinusTail as a different thing (from Tibbs) than another observer. That depends on the scope, context, purpose, available information, etc. Not necessarily those observers are to agree. But they may find an interception of (observable) characteristics and their values in such a way, that it gives them an opportunity to define/agree whether the cat is the same or not.

I see the ’name’ as just another observable characteristic, which can be used (by some observers in some scopes and contexts) for identification purposes. I don’t feel/see the name as something special. Just a sign, a label, a piece of information used by some observers. 

Kind regards,
Alex

Matthew West

unread,
Mar 26, 2019, 1:25:55 PM3/26/19
to ontolo...@googlegroups.com

Dear Alex,

It is not about beliefs about whether you think Tibbs does or does not cease to exist when he loses his tail. It is talk about about it that is talk about identity. I quite agree you could make some different decisions. But talk about it is talk about identity. On the other hand talk about what to call him is talk about identification.

Regards

Matthew

 

From: 'Alex Titov' via ontolog-forum <ontolo...@googlegroups.com>
Sent: 26 March 2019 17:11
To: ontolo...@googlegroups.com
Subject: RE: [ontolog-forum] nice free talk

 

I appreciate your beliefs.  But probably don’t share.

Alex Titov

unread,
Mar 26, 2019, 4:49:04 PM3/26/19
to ontolo...@googlegroups.com
Hi Matthew, 

From you point of view we might speak about identity of a thing. But I don’t share this point of view.

We may know (or agree between us on) a few characteristics (i.e. name, cat, absence or a tail, etc.), and a combination of those values is enough to have a discussion. You may think that there is plenty of other characteristics and use them to build some kind of ‘identity’ in your mind (some image, or may be that is a cat at your home, or anything else). And you might think that we are speaking about the particular thing (you have in mind). 

On the other hand, I don’t need such illusions. From my point of view we speak about any thing which is identified by a set of characteristics with the defined values. And that’s all. I can imagine (if I need) plenty of things with such characteristics and their values  and we continue our discussion. 

I reckon that should be enough for us to agree (means to identify the subject of our conversation) that we are speaking about one thing.


Kind regards,
Alex
Reply all
Reply to author
Forward
0 new messages