Tom - that's exactly my point.
The bank's 'intelligence system' will have a system in place that
will enable it to deal with 'random input' - such as that method,
just as you describe, where the bank's system will receive the
'rogue' input AND then, have a valid method [3-2] to physically
contact you to see if that new site of input, is valid. If so- then,
it will 'learn' and add this new site-of-input to its knowledge base.
That's called, in my view, learning, i.e., the capacity to absorb
new data and incorporate it within its existent data base.
It's not that much different from biological learning!
Edwina
On Tue 18/05/21 12:14 PM , "Tom Tinsley" [email protected]
sent:
Edwina – I understand your analysis in respect to Peirce, but
todays banking systems do not learn.
Financial systems transmit through secure network systems where the
allowable parties are known and banks pay for the service. Adding a
new party is the responsibility of the people in the organization
providing the network system. This involves contracts that define the
services offered and their cost.
A potential example of what you described might be using an app on a
personal device to access your bank information. If you use a device
that the bank has never seen before, the bank will send you an email
or message to confirm that you are the one using the device. If you
confirm that you are you, the bank will then add the device to its
list of authorized devices to access your account.
Tom
From: [email protected]
On Behalf Of Edwina Taborsky
Sent: Tuesday, May 18, 2021 11:12 AM
To: [email protected]; [email protected]
Cc: [email protected]
Subject: Re: [PEIRCE-L] Peircean Categories
Tom - another comment using your banking machine example.
The Peircean categorical framework enables complexity, in that the
categories are not just 'pure' Firstness, Secondness, and Thirdness.
They are nuanced.
So, Secondness can be a pure crisp data bit. But, there is also a
so-called 'degenerate Secondness' which is called
Secondness-as-Firstness, [which just means it's not pure but has
added the qualitative aspects of Firstness [chance, randomness,
fuzziness] to the crisp 'data bit'. So, I use my bank card at a
totally new site. A banking machine must be able to 'read' this
random input. It comes from a different location - never used before.
So- we've got a valid bank card but a seemingly invalid nodal site.
Then, how to deal with this? Thirdness, or the Mind, so to speak,
which operates as the Rules of the Process - is also nuanced. There
is pure Thirdness [3-3], which could be understood as the basic
rules. But there are also two 'degenerate modes' which enable more
complex mind processes.
There is 3-2, or Thirdness [habits] operating by direct indexical
contact with other sites. This would enable the bank to read and
accept that random input from a different location, because it can
'judge' that the other location is existentially real.
And there is 3-1, or Thirdness [habits] operating by mimetic
continuity. The bank Mind would learn - and accept that all input
from these many existential sites are valid. It might refuse my card
if the nodal site was in a different country - but- presumably, it
could 'learn' to deal with this.
In this sense, the Bank Mind is capable of learning and adding
knowledge to its Thirdness/Mind/habits.
Edwina
On Mon 17/05/21 3:51 PM , "Tom Tinsley" [email protected]
[1] sent:
Edwina,
I appreciate your description of the two basic triads. Here is my
understanding from an IT perspective.
a] Since computers began to replace human processes with automated
processes the model has been: Input -> Process -> Output.
Input occurs as the result of an event. As you know, events is a
huge subject. Input requires metadata and is often described as
having a signature. The signature to me is a formal definition of
signs. The actual data then defines the specific quantitative
attributes of an object. The process then interprets the input and
takes action. Output, like input, also has a signature that defines
the output signs and the actual object. A process may be a program, a
subroutine, a method of a class, a service of a component, etc.
Consider this simple example: A bank receives a transaction. The
bank recognizes it as a demand deposit debit transaction due to its
format or sign. The specific object is then processed by verifying
the existence of the checking account with a balance able to accept
the debit amount. If accepted, the account is updated. Output is
success or failure of the transaction with the data that documents
the outcome. This process is consistent for all debit transactions.
You said, “Thirdness is the realm of common, continuous habits of
organization. ..and is operative in the MEDIATIVE part of the triadic
process.” I agree and believe that is exactly how current digital
systems operate.
There is nothing about the current digital systems that make them
Complex Adaptive Systems. The mediation is the logic provided by
programs written by developers. The signs are defined by formal
technical definitions prepared by domain experts. The objects
involved are either transient messages or elements stored in
persistent data under the control of database administration. The
term used for changing systems is “maintenance”. It is done by
humans with the assistance of computer-based tools.
As for integration, there are some very good standards. In banking,
ACH is a big success. In healthcare it is not so good. There are many
standards, but each organization interprets them differently. As a
result, the systems that use this information must hire programmers
to figure out the translations.
b] In this triad, you noted the term Mode of Being. As I understand
your comments, this is when computing systems are able to write their
own code.
The current Semantic Web standards do not provide a means to reach
this mode. They do however provide a means to uncover inconsistency.
OWL-DL provides a means to define classes, data properties, and
object properties within a domain. These definitions determine what
is possible as well as what is not possible. They are organized as
the TBox of an OWL ontology. OWL also provides the means to define
objects in the ontology ABox. There are reasoners available that can
uncover inconsistencies in both the Tbox and ABox.
Currently, inconsistency identified by a reasoner must be resolved
by a human. Imagine if all digital systems were constructed so
complex inconsistencies could be identified before system failures
occur. I believe this would be a big step towards identifying
problems in a standard manner so there could be an AI mediator to
resolve inconsistency issues.
To fully address AI, a usual requirement for the being is it must
have a goal. Is there anything in Peirce that addresses goals? I am
working with a group that is planning to provide a digital twin of a
person to reach the highest goals in the Maslow hierarchy of needs.
Initially, the twin will only offer opportunities to satisfy
physiological needs based upon the person’s circumstance. Their
greater intent is for the twin to mentor the person through life on a
path to the higher levels where each person has their own unique
digital twin that learns and adapts to the person.
Your insight has only given me greater enthusiasm for the Peirce
framework and its applicability to digital system. As I follow up
with the contacts you have provided, I will post the results.
Respectfully,
Tom T. - OtterServer.com [2]
From: Edwina Taborsky
Sent: Sunday, May 16, 2021 1:33 PM
To: [email protected] [3]
Subject: RE: [PEIRCE-L] Peircean Categories
Tom
I think the key thing[s] to focus on is that there are, in the
Peircean framework, two basic triads.
a] The 'process' triad, which is the movement of input raw data -via
mediation- to information or interpretation . Peirce's terms are:
Object ->Representamen or Sign --> Interpretant. But the everyday
more common terms of input-mediation-output mean the same thing.
The thing to note is that our systems, in interaction with the
'outside-of-us' environment, don't input raw data directly but via a
mediating system. This can be understood as: the knowledge base. It
acts to screen raw input, decide whether it's noise or value, whether
its poison or food;, whether it's friend or foe - and if of value,
then, use it. It's not simply the brain that does this. All living
systems do it - even without the separate brain. Call it the
knowledge base of the cell. Or the immune system...
The mediating system functions, not so much with a data-base of
'specific data units' - but - it GENERALIZES input and transforms it
into long term 'general typologies'. So- for an organic system, all
input bacteria of Type ##, are accepted, while all input bacteria of
Type ** get the immune system working to reject. it. This can result
in both good and bad ...where doctors have to make the body accept
what it doesn't naturally want! A social system will take raw data
[how we behave] and transform it into a set of 'normative behaviour
in our society'. So- we will purchase goods and services rather than
steal them.
The fact that the mediation system is one that generalizes means
that it provides continuity. The system doesn't have to decide every
time a new input comes along - if its OK or Reject. Its general
Rules....enable it to make a rapid decision. The other thing about a
General Mediating system - as you note - is the INTEGRATION of
multiple input data from different sites. That capacity to analyze
discrete bits of data from multiple sites and generalize within the
common habits of the community - enables the expansive spread of the
functions of this system.
The other thing to note about this process triad - is that in the
higher living systems - the knowledge base is OPEN. That is, it can
learn. This enables adaptation and evolution. The mediative system
is usually in a Mode-of-Thirdness.
Essentially, this means that living systems are Complex Adaptive
Systems. They Self-Organize this mediating knowledge system - which
means that the mediating system doesn't need any External Ruler - to
change its Rules and adapt. A population will itself decide to change
its rules of behaviour.
Now, machines and computers lack this internal Mediating system -
and that's what people in AI and IT etc, are trying to solve - how to
make what is a Dyadic System [input/output]….become a triadic system
[input/mediation/output].
b] The other triad to note is one that refers not to the
Process-of-information development, but to the Mode of Being. The
terms here are: Firstness, Secondness and Thirdness. These are NOT
ordinal terms. They refer instead to the mode, the way that the data
or information organizes itself and functions.
Firstness data is pure open sensation: qualitative, subjectively
measured, relativist, highly dissipative and individualistic. The
pure unanalyzed feeling of heat..It is useful to bring something to
our attention.
Secondness data is discrete, separate units. This
-as-differentiated-from-that. This cell, with its cell walls, is
separate from that cell with those cell walls. This word is distinct
from that word. Quantitative measurements.
Thirdness is abstract generals of Type. A Type focuses on
commonalities rather than differences [which is the realm of
Secondness]. A Type can be defined because all follow the same RULES;
;they all breathe oxygen, they all have four legs; they all ...etc.
Here - we can only MODEL the generalities; ie, that this Type always
follows a particular pattern of development.
Thirdness is the realm of common, continuous habits of organization.
..and is operative in the MEDIATIVE part of the triadic process.
------------------------------------------
I think the key values of Peirce to IT would be the triadic process
- with that insertion of a mediating process of Thirdness.
-----------------
Just a spelling note. It's Chris Laundauer. Not Christ. And it's
Mark Burke.
I think key contacts could be Ricardo Gudwin and Daniel Dubois.
Regards
Edwina Taborsky
On Sun 16/05/21 11:35 AM , "Tom Tinsley" [email protected]
[4] sent:
Edwina,
I am overwhelmed with your list of contacts. I will reach out to
each of them.
The term you used, triadic infrastructure, hits home for me. Having
managed hundreds of systems in banking, government, retail, and media
the biggest issue is integration of systems both internal to an
organization and external to other organizations. I believe a common
implementation of the triadic infrastructure would solve this issue.
Also, I have seen multiple presentations of OWL-based systems and
none have been constructed with a common infrastructure. Each
utilizes diverse technical standards and implementations. Although
each system meets a unique need, each is a silo of knowledge. As a
result it is complex for these systems to share knowledge.
Bright spots like linked data provide sharing. Not because they are
all created and accessed using products from the same vendor, but
because they all apply the same sematic standards. How much more
could be accomplished if all digital systems followed an integrated
set of common standards in implementing the triadic infrastructure?
I know it can be done. I have a prototype of one approach. What I
expect to learn from the sources you provided is that others have
better solutions for implementing the triadic infrastructure.
Thank you for taking the time and providing so many contacts.
Sincerely,
Tom
From: Edwina Taborsky
Sent: Saturday, May 15, 2021 5:36 PM
To: [email protected] [5]; [email protected] [6]
Subject: Re: [PEIRCE-L] Peircean Categories
Tom, list
Yes, you are right; many others in the field of information
technology have recognized the analogies with Peircean semiosis. they
don't use the same terms [such as Firstness, Secondness and Thirdness]
for the cognitive process -but - they certainly recognize both the
triadic infrastructure and the realities of possible, actual,
continuity/necessary.
The field is huge as I'm sure you know - and I couldn't give you
just one name, but would suggest that you contact:
Daniel Dubois Institute of Mathematics, University of Liege, Belgium
[email protected] [7]
He has set up the remarkable conference system, "CASYS: or Computing
Anticipatory Systems, with scholars working in AI, IT, biology,
chemistry, physics, mathematics - to examine essentially cognitive
processes in these areas. The people who have participated in these
conferences are 'the top'.
He can provide you with names and contacts.
Again, it is important to be able to see the common conceptual
infrastructure that these different disciplines are using and accept
that they are not using Peircean terms - and frankly - there is no
need to do so. It's the conceptual infrastructure that is the same...
I'd also suggest some others who have participated in the Dubois
conferences;
Marke Burke: [email protected] [8] [Dept of Mathematics, U of
Limerick, deals with anticipatory systems
Christ Laundauer and Kirstie Bellman [Aerospace Integration Science
Center; [email protected] [9], [email protected] [10]
And Leonid Perlovsky, editor of Physics of Life [Elsevier] - which
also deals in artificial life, robotics, AI.
[email protected] [11]
He also knows plenty of people in the AI field - And there are more
in the field of AI and robotics - who do understand the analogy with
the Peircean framework.
For instance, the book: 'Semiotics and Intelligent Systems
Development,, 2007, edited by Ricardo Gudwin and Joao Queiroz. Idea
Group Publ.
Both Ricardo and Joao would know of more contacts - and that
includes with IEEE symposiums and are interested in computational
semiotics.
[email protected] [12]
Hope these contacts help.
Edwina Taborsky
On Sat 15/05/21 2:35 PM , "Tom Tinsley" [email protected]
[13] sent:
I have just recently joined this group. My background is in
Information Technology. My interest is in others’ understanding of
the Peircean categories of Possibility, Actuality, and Necessity as
the foundation of engineering, specifically software engineering.
Here is a little more about my interest.
The engineering of something new adheres to a standard methodology.
First there must be a requirement or concept, followed by
specifications, and then construction. In software engineering the
same steps are followed: requirements, design, and programming.
Traditionally, these steps are done in the order listed (The
Waterfall methodology). Currently, Agile methodologies are being
applied that recognize the benefit of small iterations of these
steps. Once, I even reorganized the IT software development group
into three separate organizations (Requirements, Design, Programming)
and followed a then well accepted project methodology.
I wrote a book on enterprise architecture back in 2009. During my
research for the book, I was disappointed that I did not find
standards in the development of conceptual models to describe an
enterprise. The best available products for conceptual design were
provided by database providers. My research took me into the Semantic
Web standards, where I became convinced that description logic as
specified in the OWL-DL language could be used as a standard for
conceptual models in enterprise architecture. And I was surprised to
learn that consistency could be proven by reasoners across all three
software development steps.
To learn more, I joined the ontology-forum, and also began
developing a prototype to prove that consistency could be maintained
across all development steps. I presented a paper on the prototype at
KEOD 2011 where I met Professor Jan Dietz, a leader in this area.
Then, after observing discussions on the ontology-forum, I realized
that the Peirce categories matched the IT categories. Possibility is
equivalent to design. Actuality is equivalent to database, file, and
message specifications. And Necessity is equivalent to programming.
Today’s efforts in AI are to teach computers to be equivalent to
human cognition. Since Peirce addresses human cognition, it seems
reasonable that the Peircean categories should be the foundation for
AI.
I am confident that many others have recognized these same
equivalences, but I am not finding publications. So, I am interested
in others’ evaluation of this conjecture, and wondering if there
are ideas, papers, books, videos, etc. that could provide a greater
understanding of how Peirce’s publications could be applicable to
the categories of software development.
I look forward to dialog!
Respectfully,
Tom Tinsley
OtterServer.com [14]
Links:
------
[1]
http://webmail.primus.ca/javascript:top.opencompose(\'[email protected]\',\'\',\'\',\'\')
[2] http://webmail.primus.ca/OtterServer.com
[3]
http://webmail.primus.ca/javascript:top.opencompose(\'[email protected]\',\'\',\'\',\'\')
[4] http://webmail.primus.ca/javascript:top.opencompose(
[5] http://webmail.primus.ca/javascript:top.opencompose(
[6] http://webmail.primus.ca/javascript:top.opencompose(
[7] http://webmail.primus.ca/javascript:top.opencompose(
[8] http://webmail.primus.ca/javascript:top.opencompose(
[9] http://webmail.primus.ca/javascript:top.opencompose(
[10] http://webmail.primus.ca/javascript:top.opencompose(
[11] http://webmail.primus.ca/javascript:top.opencompose(
[12] http://webmail.primus.ca/javascript:top.opencompose(
[13] http://webmail.primus.ca/javascript:top.opencompose(
[14] http://webmail.primus.ca/OtterServer.com
_ _ _ _ _ _ _ _ _ _
► PEIRCE-L subscribers: Click on "Reply List" or "Reply All" to REPLY ON
PEIRCE-L to this message. PEIRCE-L posts should go to [email protected] .
► To UNSUBSCRIBE, send a message NOT to PEIRCE-L but to [email protected]
with UNSUBSCRIBE PEIRCE-L in the SUBJECT LINE of the message and nothing in the
body. More at https://list.iupui.edu/sympa/help/user-signoff.html .
► PEIRCE-L is owned by THE PEIRCE GROUP; moderated by Gary Richmond; and
co-managed by him and Ben Udell.
