What is System Science?

38 views
Skip to first unread message

James Martin

unread,
Aug 22, 2018, 6:34:11 PM8/22/18
to SSWG

A week-long Conversation was held 8-13 April in Linz, Austria, sponsored by the IFSR to discuss the nature of system science. A short report on this event is attached.


This discussion was led by Gary Smith, co-led by Jennifer Makar, with participation from Gary Metcalf, George Mobus, Swaminathan Natarajan, and Hillary Sillitto. Below is a short report on this event. A more complete description will be provided in November in the IFSR Conversation Proceedings. For information about Conversations in general, see http://www.ifsr.org/index.php/ifsr-conversations/what-is-an-ifsr-conversation/.  


--
James
What is System Science.docx

Ken Lloyd

unread,
Aug 22, 2018, 7:32:22 PM8/22/18
to syss...@googlegroups.com

This presupposes two prior questions have been answered (correctly or not):

1.       What is science?

2.       What characterizes systems?

What normally happens is people start trying to precisely define these terms – by consensus - in a universally agreeable way. Historically, we have been precisely incorrect or at least incomplete. These are, and always have been, evolving terms representing increasingly more sophisticated concepts than most people are aware of. This caused Nobel Laureate and geophysicist Henry Pollack to pen, “People love science. The just don’t understand it”. I normally add to that statement, scientists are people, too.

 

Much of what we understand about science, today, comes from the physical sciences (i.e. physics, chemistry, biology). In many ways, “science” started to bifurcate from philosophy during the rejection of Scholasticism around the 17th Century.  I say started because this transition is still underway, today. The kid just won’t leave his parent’s basement.

 

In 1932, Max Planck identified an “inner” and “outer” aspect in science. The inner aspect he refined as “intrinsic” – the structure of all the perceptible stuff in our universe that occurs, naturally existing prior to mankind’s existence. The outer aspect he refined as “extrinsic” – the restructuring of the stuff of the universe by agency or intelligence, usually attributable to mankind, for whatever purposes mankind chose. Sometimes, there were unexpected and unintended consequences of these restructuring efforts.

 

Systems are an abstraction of those things -  matter, energy, information, entropy  known and unknown – that structures all the stuff of the universe. We can identify, describe and characterize a very tiny fraction of it. We attempt to learn about all those “things and stuff” by observing patterns of their existence and interaction. We humans make a lot of mistakes in these observations, and the conclusions we draw from them.

 

One of the patterns I’ve identified is that science, itself, is a complex-coupled system, which reflectively takes us back to my first sentence, and the process continues.

 

Ken Lloyd

--
The SysSciWG wiki is at https://sites.google.com/site/syssciwg/.
 
Contributions to the discussion are licensed by authors under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
---
You received this message because you are subscribed to the Google Groups "Sys Sci Discussion List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+u...@googlegroups.com.
Visit this group at https://groups.google.com/group/syssciwg.
For more options, visit https://groups.google.com/d/optout.


Virus-free. www.avg.com

Randall Russell

unread,
Aug 22, 2018, 7:58:15 PM8/22/18
to syss...@googlegroups.com
Excellent outcomes and discussions.

To respond directly, pragmatist that I am, I propose the following observations at a very simple level.

1. What is science? Knowledge.
2. What do we know about systems - interacting components?  

Systems Science knowledge is no different than any other science. We don’t have all the knowledge, yet. Biology, chemistry, physics, even mathematics continues to evolve and the body of knowledge grows. With respects to Planck, Occam has a point.

Plow forward on the knowledge of systems frontier. Understanding complexity in the simplest possible explanations is a worthy and noble endeavor.

Russell
Founder and President
Ex Nihilo Systems -- Sometimes all you have is a good idea!
 

From: 32421033000n behalf of
Sent: Wednesday, August 22, 2018 19:32
To: syss...@googlegroups.com
Subject: RE: [SysSciWG] What is System Science?
 

Smith, Gary [UK]

unread,
Aug 23, 2018, 6:23:59 AM8/23/18
to syss...@googlegroups.com

Thanks Ken. As you can see from the questions we raised on the very first day. The first one was indeed ‘what is science’.

We used the common definition to guide us.

Science is a systematic enterprise that builds and organizes knowledge in the form of testable explanations and predictions about the universe”

Our intent as described in the report, being to create a candidate structure for our system knowledge.

 

So yes, it was our starting point and as Len says, it is all about knowledgel.

 

I appreciate the view of Mark Planck that you referred to Ken, I hadn’t come across this. This fits nicely with the differentiation we made when structuring our systemic knowledge into systemic practice and systemic foundations.

 

BR Gary

 

From: syss...@googlegroups.com [mailto:syss...@googlegroups.com] On Behalf Of Ken Lloyd


Sent: 23 August 2018 00:32
To: syss...@googlegroups.com


--
Airbus Defence and Space has taken all possible efforts
to ensure this email is free from malicious code;
however, never open suspicious emails
or attachments or follow unknown hyperlinks.

--
The SysSciWG wiki is at https://sites.google.com/site/syssciwg/.
 
Contributions to the discussion are licensed by authors under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
---
You received this message because you are subscribed to the Google Groups "Sys Sci Discussion List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+u...@googlegroups.com.
Visit this group at https://groups.google.com/group/syssciwg.
For more options, visit https://groups.google.com/d/optout.

The information contained within this e-mail and any files attached to this e-mail is private and in addition may include commercially sensitive information. The contents of this e-mail are for the intended recipient only and therefore if you wish to disclose the information contained within this e-mail or attached files, please contact the sender prior to any such disclosure.

If you are not the intended recipient, any disclosure, copying or distribution is prohibited. Please also contact the sender and inform them of the error and delete the e-mail, including any attached files from your system.

Emails to Airbus Defence and Space Limited may be processed, recorded and monitored anywhere in the European Community.


Airbus Defence and Space Limited

Registered Office: Gunnels Wood Road, Stevenage, Hertfordshire, SG1 2AS.
Registered in England and Wales under company number 02449259.

Steven Krane

unread,
Aug 23, 2018, 9:19:18 AM8/23/18
to syss...@googlegroups.com
Science:  The development of truths of the form - If X happens, then Y will happen, where the happening of X and Y are determined by measurement. 

There is 1 System.  It, not all of it at once, is the subject of all science. 

It’s much more narrow than knowledge. I know Times Square is in NYC.

Randall Russell

unread,
Aug 23, 2018, 9:24:26 AM8/23/18
to syss...@googlegroups.com
I have to disagree, quite strongly here.

Location of a named place is a data point, which could lead to information, which may lead to knowledge.

Russell
Founder and President
Ex Nihilo Systems -- Sometimes all you have is a good idea!
 

From: 32571272440n behalf of
Sent: Thursday, August 23, 2018 09:19
To: syss...@googlegroups.com
Subject: Re: [SysSciWG] What is System Science?
 

Smith, Gary [UK]

unread,
Aug 23, 2018, 12:38:22 PM8/23/18
to James Martin, Jennifer Makar, Swaminathan Natarajan, George Mobus, Hillary Sillitto, Gary Metcalf
Hi James, could you be sure to add all of our Linz team to‎ the system science working group distributor. I'm keen to ensure that they pick up the feedback. 

Sent from my BlackBerry 10 smartphone.
From: James Martin
Sent: Thursday, 23 August 2018 00:34
To: SSWG
Subject: [SysSciWG] What is System Science?
--
The SysSciWG wiki is at https://sites.google.com/site/syssciwg/.
 
Contributions to the discussion are licensed by authors under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
---
You received this message because you are subscribed to the Google Groups "Sys Sci Discussion List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+u...@googlegroups.com.
Visit this group at https://groups.google.com/group/syssciwg.
For more options, visit https://groups.google.com/d/optout.

--
Airbus Defence and Space has taken all possible efforts
to ensure this email is free from malicious code;
however, never open suspicious emails
or attachments or follow unknown hyperlinks.

Steven Krane

unread,
Aug 23, 2018, 5:37:16 PM8/23/18
to syss...@googlegroups.com
Don’t like that example?  How about - I know my mother.  Is that knowledge?  Is it scientific?

Randall Russell

unread,
Aug 23, 2018, 8:32:54 PM8/23/18
to syss...@googlegroups.com
Can you confidently predict your mother’s behaviors under specific conditions, express the causes for those behaviors, and adjust inputs to influence the behaviors?  

The short answer is that not all science is first principles, mechanics, statistics, or even heuristics. Some knowledge and science is observation and empiricism. 

Long answer: Moms are complex, complicated systems, from at least one perspective. From at least one of those perspectives, Moms can be mysterious and consistent at the same time, and observations can be catalogued, correlated and interpreted, formally and informally. In general, Moms would prefer a call on Sunday, flowers on her birthday, cards and chocolate on Mothers Day. From a very different perspective, Moms would prefer to spend time with their kind and patient son, rather than be the subject of an academic analysis of epistemological concepts. — The universality of my particular depth of maternal knowledge has serious limitations, so choose your flowers and visits at your own risk. 

Not sure that answers your question. But it was an interesting excursion into the abstract wisdom of parental science. 

B/R

Russell
Founder and President
Ex Nihilo Systems -- Sometimes all you have is a good idea!
 

From: 32021323100n behalf of
Sent: Thursday, August 23, 2018 17:37

Steven Krane

unread,
Aug 23, 2018, 10:34:57 PM8/23/18
to syss...@googlegroups.com
Re: “Some knowledge and science is observation...”

Therein lies the (a) difference in what activities you and I would label “science”.

Jack Ring

unread,
Aug 23, 2018, 10:41:37 PM8/23/18
to Sys Sci
It is not science until you specify one or more fallibility experiments. 

James Martin

unread,
Aug 23, 2018, 10:46:14 PM8/23/18
to SSWG
Gary,

I added them. All but one of them were already members of the group. Don't know which one that was since it did not tell me...

James

James

Randall Russell

unread,
Aug 23, 2018, 11:35:15 PM8/23/18
to syss...@googlegroups.com
If you haven’t has a fallible moment with your maternal parent, well then boundary conditions are unknown. That would fall into the “it seemed like a good idea at the time” category.

Russell
Founder and President
Ex Nihilo Systems -- Sometimes all you have is a good idea!
 

From: 32051246160n behalf of
Sent: Thursday, August 23, 2018 22:41
To: Sys Sci

Randall Russell

unread,
Aug 23, 2018, 11:37:13 PM8/23/18
to syss...@googlegroups.com
Not my labels. WHO owns that one, with reasonable rationale.

Russell
Founder and President
Ex Nihilo Systems -- Sometimes all you have is a good idea!
 

From: 32041122620n behalf of
Sent: Thursday, August 23, 2018 22:35

To: syss...@googlegroups.com
Subject: Re: [SysSciWG] What is System Science?
 

Steven Krane

unread,
Aug 24, 2018, 12:33:25 AM8/24/18
to syss...@googlegroups.com
Oh.  Ok.  The “science” of health care seems bedeviled with malpractice.  Maybe that’s why.  Just my opinion based on first learning of leeching blood up to present day p-hacking

Steven Krane

unread,
Aug 24, 2018, 12:38:24 AM8/24/18
to syss...@googlegroups.com

Smith, Gary [UK]

unread,
Aug 24, 2018, 2:27:32 AM8/24/18
to James Martin
‎Thank you :)

Sent from my BlackBerry 10 smartphone.
From: James Martin
Sent: Friday, 24 August 2018 03:49
To: SSWG
Subject: Re: [SysSciWG] What is System Science?

Jack Ring

unread,
Aug 24, 2018, 6:35:32 AM8/24/18
to Sys Sci
Randall,
Can you do likewise for each thing you label “system”? 

Sometimes all you have is an interesting idea!. The rest of the system decides whether it is a good idea.
Jack

Steven Krane

unread,
Aug 26, 2018, 8:27:38 AM8/26/18
to syss...@googlegroups.com
A similar, perhaps more useful question than “what is System Science”, is “how can science be used in the study of systems, or can it”.  What science does/is doesn’t depend on any domain.  Much of the “sciences” are not science, they are Logos.  They lack the purification, the hallmark of science, that comes from belief in the fallibility of the greatest preceding teachers.

I think the question of how science can be used to correct bad thinking about systems is difficult and interesting.  I don’t know that SSWG has made much progress toward the answer.  I’m not sure the question “what is system science” even makes sense or what we would do with the answer if we had one, or more.  In a field where propositions (usually too vague to form hypotheses) far outnumber experiments we could expect a great number of faults and ambiguities.  In aircraft systems, we do science when we test what we said they would do.  It’s very imperfect with the answers being only good enough to fly, usually.

What did (insert your systems hero here) say that could be demonstrated to be wrong? If you can’t think of anything then what they said is unscientific and cannot be approached by that method.  If there are scientific statements made by systems “experts”, perhaps SSWG could demonstrate what system science does, rather than talk about what it is.

Hillary Sillitto

unread,
Aug 26, 2018, 8:58:53 AM8/26/18
to syss...@googlegroups.com
Steven

Couldn’t agree more.

It seems to me that to be a science, Systems Science needs to be able to make two kinds of provable/falsifiable statements. 

There might be others, but these would seem to be the acid test for whether a philosophy has yet made a successful transition into a science:

1. “The behaviour of these kinds of systems observed in nature can be represented by this kind of model <adequately/partially/in-some-defined-situations>";

2. "If a system is adequately represented by this model, then it will (within an understood level of variability) behave this way under these circumstances”.

This seems to imply that we need to be much clearer about what is in the physical universe and what is in the “model world” than most of us are at the moment.

Best regards

Hillary

Ken Lloyd

unread,
Aug 27, 2018, 7:57:47 AM8/27/18
to syss...@googlegroups.com

To add to what Hillary has contributed, here are two links that generally help characterize what is and isn’t science.

 

https://scienceornot.net/hallmarks-of-science/

 

https://scienceornot.net/science-red-flags/

 

My perhaps controversial conjecture is that scientific knowledge has bifurcated – become something completely different from – human belief (whether justified as true or not). While all potential knowledge is originally identified by insight, intelligence and intuition, it is always conditioned by prior beliefs in an ever-changing paradigm that must be validated and verified. The history of science shows this paradigm can be wrong, creating crisis ala Kuhn. Any paradigm of science enables one to see structural patterns in the paradigm (schemata). But that same paradigm can constraint one from seeing important existing patterns as well.

 

Science only moves toward knowledge development when there is perceived an inadequacy in either the depth and breadth of prior knowledge to accomplish the important and necessary things in life. In other words, the objectives of science can often be complicated by the fact it is conducted by scientists (believers).

 

Ken Lloyd

 

From: 'Hillary Sillitto' via Sys Sci Discussion List [mailto:syss...@googlegroups.com]
Sent: Sunday, August 26, 2018 6:59 AM
To: syss...@googlegroups.com
Subject: Re: [SysSciWG] What is System Science?

 

Steven


Virus-free. www.avg.com

Steven Krane

unread,
Aug 27, 2018, 1:05:57 PM8/27/18
to syss...@googlegroups.com
Thanks for the links. The red flags overlap a fair bit with general propaganda techniques. 

As Feynman says, the first step is to be honest. The second is to not fool yourself, and you are the easiest person to fool.

George Mobus

unread,
Aug 27, 2018, 1:41:01 PM8/27/18
to syss...@googlegroups.com

Steven wrote:

'A similar, perhaps more useful question than “what is System Science”, is “how can science be used in the study of systems, or can it.' 
There have been a spate of books and papers coming from a variety of inherently multidisciplinary fields, such as the 'origin of life' problem, in which systems concepts are being successfully used to explicate heretofore difficult (complex) problems. I very strongly recommend, from this genre, "The Origin and Nature of Life on Earth: The Emergence of the Fourth Geosphere" by Eric Smith and Harold J. Morowitz (https://www.amazon.com/Origin-Nature-Life-Earth-Emergence/dp/1107121884/ref=sr_1_1?s=books&ie=UTF8&qid=1535390848&sr=1-1&keywords=the+origin+and+nature+of+life+on+earth). It is an extensive review of a wide scope of research regarding the origin of life as a phase transition (or actually a series of micro transitions) in geochemistry (probably around thermal vents in deep ocean) as pre-Darwinian evolution/emergence.

The main point is that their language is explicitly systems as they describe the origin of metabolic cycles to the origin of multicellular life and ecosystems. They explicitly use Herbert Simon's views of modularity and hierarchical control, Ashby's requisite variety, and Shannon's information theory among other concepts that provide significant insights into this extremely important problem domain.

If we look at these multidisciplinary fields (e.g. socio-economics) we find a growing number of researchers turning to systems concepts to elucidate their subjects. They are doing this without the systems community having successfully defined what is systems science or systems engineering. I think this is because there are a number of basic meta-scientific principles in systems that become clearer once getting into such subjects.

Perhaps we should spend more time observing what is happening in these subjects and less time talking into an echo chamber!
My $0.02


---------------------------------------------------------------------------
George Mobus, PhD.
Associate Professor Emeritus, Institute of Technology
University of Washington Tacoma

Street mail: 1900 Commerce St. Tacoma, WA 98402-3100 Box 358426
Phone: 253.692.5894
E-mail: gmo...@uw.edu
Web site: http://faculty.washington.edu/gmobus


From: syss...@googlegroups.com <syss...@googlegroups.com> on behalf of Steven Krane <sk5...@gmail.com>
Sent: Sunday, August 26, 2018 5:27:34 AM

To: syss...@googlegroups.com
Subject: Re: [SysSciWG] What is System Science?

Steven Krane

unread,
Aug 27, 2018, 1:49:09 PM8/27/18
to syss...@googlegroups.com
That’s an interesting perspective and turn around of what I said. Applying system concepts/methods to scientific work vs applying science to the study of systems.  Science Systems and Systems Science. 

George Mobus

unread,
Aug 27, 2018, 4:04:22 PM8/27/18
to syss...@googlegroups.com

In the 2016 Linz Conversation (IFSR) the SS Research Team identified two scientific communities of interest as "needing" systems science. There are those who consider themselves systems scientist proper who do the science needed to define systemness proper and those who do trans/multidisciplinary science in the traditional sense of the natural sciences but need SS to guide their work. In the latter sense SS is not really a science as it is meta-science, as a guide to the overall structure and form of generalized knowledge. E.g. knowing that feedback loops are a key to understanding complex dynamics an ecologist looks for such loops in the ecosystem and does not think only in terms of trophic levels as one-way flows of energy.

What systems scientists proper do was left an open question. The 2018 Linz Conversation picked up this topic to some degree. It is still not clear how the exploration of systemness can be pursued in a strictly scientific way. If, as I have asserted, SS is really meta-science (akin to our notion of metaphysics) then it might be a category mistake to think it is or should be science. If SS were a science of universal patterns then one hypothesis might be: every thing that we call a system is a Simonian hierarchy of modular subsystems and components. The empirical problem would then be to survey all things we call systems and see if this hypothesis bears out, clearly an impossible task except possibly by some kind of inductive reasoning (an approach I think is evident in many authors' writings).


I am growing content with the idea that SS is really meta-science and as such can best be approached as making ontological commitments based on our best grasp of phenomena observed developing an epistemology based on that ontology and then applying those concepts through the natural and social sciences to grasp reality as an integration of seemingly separate phenomena. That, at least, is the approach I am currently working on, started independently, but aided now by commenters in this group and from ISSS and IFSR. Book in progress.


---------------------------------------------------------------------------
George Mobus, PhD.
Associate Professor Emeritus, Institute of Technology
University of Washington Tacoma

Sent: Monday, August 27, 2018 10:49:05 AM

Curt McNamara

unread,
Aug 27, 2018, 4:20:02 PM8/27/18
to Sys Sci
If I were to work on creating a toolbox for systems workers, the following seems useful:

-- analyze the work of exemplary practitioners. Here is my short list, who would you add? Or: who is on your list?

Howard Odum

Peter Csermely (for an overview of the field)

Buckminster Fuller

-- summarize best practice in mapping and modeling

         Curt

To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+unsubscribe@googlegroups.com.

--
The SysSciWG wiki is at https://sites.google.com/site/syssciwg/.
 
Contributions to the discussion are licensed by authors under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
---
You received this message because you are subscribed to the Google Groups "Sys Sci Discussion List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+unsubscribe@googlegroups.com.

--
The SysSciWG wiki is at https://sites.google.com/site/syssciwg/.
 
Contributions to the discussion are licensed by authors under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
---
You received this message because you are subscribed to the Google Groups "Sys Sci Discussion List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+unsubscribe@googlegroups.com.

--
The SysSciWG wiki is at https://sites.google.com/site/syssciwg/.
 
Contributions to the discussion are licensed by authors under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
---
You received this message because you are subscribed to the Google Groups "Sys Sci Discussion List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+unsubscribe@googlegroups.com.

--
The SysSciWG wiki is at https://sites.google.com/site/syssciwg/.
 
Contributions to the discussion are licensed by authors under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
---
You received this message because you are subscribed to the Google Groups "Sys Sci Discussion List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+unsubscribe@googlegroups.com.

Jack Ring

unread,
Aug 27, 2018, 4:46:33 PM8/27/18
to Sys Sci
A Wayne Wymore and John N. Warfielf because each conducted more than 20 projects to determine whether their systemics were fit for user purpose.

To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+u...@googlegroups.com.

George Mobus

unread,
Aug 27, 2018, 8:27:10 PM8/27/18
to syss...@googlegroups.com

Fuller and Odum are definitely examples of scientist who used systems thinking to advance their disciplinary approaches. Odum was the “father” of systems ecology (my colleague in biophysical economics, Charlie Hall, was his PhD student) and Fuller, while advancing a wide swath of physical science, such as tensegrity and the geodesic dome, was first and foremost a systems thinker. (warning – shameless name dropping on my part) I had the good fortune to spend some small amount of time with him back in the early 80s just before his passing. Both Fuller and Odum discovered attributes of systemness from their pursuits of their respective multidisciplinary sciences. They did not start with systems science and advance their disciplines from it. But their example of how systems thinking can influence the pathways of discovery in a traditional science-based discipline and give rise to recognition of systemness as a general approach to understanding phenomena is instructive. In other words, people who think in systems naturally will end up advancing their disciplines because the concepts of systemness provide significant guidance to understanding those phenomena.

 

The question is (ala von Bertalanffy) can a theory of systems (systemness) be broadly applied in the sciences such that it provides guidance to new and more meaningful insights (understanding)? Evidence from the works of people like Odum and Fuller, as well as Lee and Morowitz (I also had the honor of conversing with Morowitz in the early 80s), suggests that it can. SS and a theory of systemness, rightly understood, can provide such guidance if understood to work this way. A theory of systemness can provide guidance to the sciences, which is why it is worth pursuing as a field of understanding on its own.

 

The implications for this approach to systems science for systems engineering is that the latter is not “mere” engineering, but must be meta-engineering – the engineering of engineering. So the language of mere engineering may not be adequate to tackle the field. The Linz Conversation touched on this as well.

 

George

To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+u...@googlegroups.com.

--
The SysSciWG wiki is at https://sites.google.com/site/syssciwg/.
 
Contributions to the discussion are licensed by authors under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
---
You received this message because you are subscribed to the Google Groups "Sys Sci Discussion List" group.

To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+u...@googlegroups.com.

--
The SysSciWG wiki is at https://sites.google.com/site/syssciwg/.
 
Contributions to the discussion are licensed by authors under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
---
You received this message because you are subscribed to the Google Groups "Sys Sci Discussion List" group.

To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+u...@googlegroups.com.

--
The SysSciWG wiki is at https://sites.google.com/site/syssciwg/.
 
Contributions to the discussion are licensed by authors under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
---
You received this message because you are subscribed to the Google Groups "Sys Sci Discussion List" group.

To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+u...@googlegroups.com.

--
The SysSciWG wiki is at https://sites.google.com/site/syssciwg/.
 
Contributions to the discussion are licensed by authors under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
---
You received this message because you are subscribed to the Google Groups "Sys Sci Discussion List" group.

To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+u...@googlegroups.com.

--
The SysSciWG wiki is at https://sites.google.com/site/syssciwg/.
 
Contributions to the discussion are licensed by authors under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
---
You received this message because you are subscribed to the Google Groups "Sys Sci Discussion List" group.

To unsubscribe from this group and stop receiving emails from it, send an email to syssciwg+u...@googlegroups.com.

Ken Lloyd

unread,
Aug 27, 2018, 9:22:34 PM8/27/18
to syss...@googlegroups.com

In re: a science for systems, it seems there are two aspects – an inner or intensive aspect (where a direct science is applicable), and an outer or extensive aspect (what may be considered by some as a meta-science, of which there may be many levels). This was coarsely identified by Max Planck in 1936. Like trying to understand an elephant by cutting it into smaller and smaller slices – so-called “elephant carpaccio” (A. Cockburn) - one cannot adequately understand any system exclusively from within the context of that system (i.e. Godelian incompleteness).

 

Many of the explanations of systems done in the 1960s – 1980s suffer from insufficiency in knowledge, gained subsequently beyond their original paradigms in which they were formed. This is clearly a Bayesian phenomenon – the effects of new information on the prior. The problem with the “laws” of science is they need to be changed and upgraded from time to time primarily because they were incongruent, incomplete or paradoxical – and in some cases, just plain wrong.

 

D’Abro wrote in “The Evolution of Scientific Thought from Newton to Einstein”, “Although in the course of the last three centuries scientific theories have been subject to all manner of vicissitude and change, the governing motive that has inspired scientists has been ever the same --- a search for unity in diversity, a desire to bring harmony and order into what might at first sight appear to be a hopeless chaos of experimental facts.”

 

Ken Lloyd


Virus-free. www.avg.com

Hillary Sillitto

unread,
Aug 28, 2018, 3:25:55 AM8/28/18
to syss...@googlegroups.com
This article on “impossible systems” from the current edition of “Physics World” is very interesting and relevant to this discussion.

Ken Lloyd

unread,
Aug 28, 2018, 6:24:24 AM8/28/18
to syss...@googlegroups.com

Re: “impossible systems”, just because we lack knowledge about systems – or they seem not to make sense to us give our current paradigm – does not, in and of itself, mean these systems are impossible. At one time, heavier than air human flight was considered impossible. Things are only impossible until they aren’t. If you watch the Science Channel, or read many scientific reports, there is a lot of “science” that isn’t science – merely a lot of conjecture and belief statements (which, because we are human, may be unavoidable).

 

On a different tack through the known/unknown problem space, one of the experiments we conducted (actually over 10,000 of them) trying to falsify a theorem by Kenneth Stanley and Joel Lehman popularized for a general audience in their book “Myth of the Objective”, we found there was some potential validity to their hypothesis. Their conclusion was there is often “deception”, both in the prior conceptualization in path development to, and within the objective, itself, in searching a space for solutions. The interesting question became, what are the sources of that deception (an observable phenomena such as falling into a local minima, or strangeness in the attractor)? For example, the prior accepted kludge or patch to this phenomenon was artificial simulated annealing. High potential, new knowledge emerged from these experiments.

 

Further examination and study of this “interesting” system phenomenon required we employ scientific methods. Merely thinking about the problem was futile. (One tricky issue, try to define or characterize what is “interesting” about a problem space).

 

We may be at a point of bifurcation in the study of systems. Some will continue on exclusively relying on mental and philosophical methods and thought in the existing paradigm. Some will diverge using competitive co-evolution between human, mathematical and computer evolved models in myriad contexts. Which will prove better? Stay tuned, news at 11:00 ….

H

unread,
Aug 28, 2018, 6:40:49 AM8/28/18
to syss...@googlegroups.com
Ken
Look before you leap, my friend!
You need to read the article to understand what the authors mean by ‘impossible systems’. What they actually mean is ‘systems it is impossible for us to get access to’, a key example being black holes. The trick is to identify analogue systems, that follow the same maths, that we can get access to - so isomorphism. This is absolutely about real science and nothing at all to do with philosophy!

Sent from my iPhone

Ken Lloyd

unread,
Aug 28, 2018, 8:24:51 AM8/28/18
to syss...@googlegroups.com

I did look. It may be inaccessible to us at this point in time. That does not imply it will remain inaccessible in the future (which I doubt).

 

There was a time when it would have been inconceivable to us that black holes even existed. Very large mass in an almost infinitesimal space? Neutron degeneracy? Wow! These were originally a manifestation from the mathematics (and while Einstein was a brilliant physicist, he was only above average as a mathematician. He wasn’t always spot on correct in either. His philosophy and the paradigm of the time got in his way).

 

Here’s a problem I have (I admit it is MY problem), the “real world” exists at some distance from equilibrium (Prigogine, Bishop, Brussels-Austin Group) – not equilibrium. Historically, mathematics has been considered exclusively at equilibrium – i.e. allowing (mandating?) isomorphism. The mathematics I use allows non-equilibrium morphisms (homomorphisms, endomorphisms, etc.) that work in statistical thermodynamics and other “real world” phenomena (like evolution). The equilibrium conceptualization of math IS a vestige from its philosophical roots. There are similar vestiges in mathematical logic (save us, please, from Boolean logical thinking). For example, see Abramsky’s work in  logic https://arxiv.org/abs/1604.02603 or Bennett et al. https://link.springer.com/article/10.1023/A:1020083231504 .

 

Most science starts from priors, and these are often of a philosophical nature and often quite old. That does not mean they are true and correct!

 

Ken Lloyd

Aleksandar Malečić

unread,
Aug 28, 2018, 6:50:10 PM8/28/18
to syss...@googlegroups.com
I think that discussions such as this one show how and why systems science will be stuck "for a while". Someone might just as well out of the blue claim for instance that artificial self-awareness is inaccessible, but that doesn't mean that it will be inaccessible in the future. Even though that's mathematically proven by Robert Rosen. Or that the Heisenberg Uncertainty Principle is about something uncertain to people "at the moment". What if information (some of it maybe inherently unknowable and/or inaccessible) is at the bottom of reality? I think that David Deutsch with his fabric of reality (the four strands rather than Kent Palmer's ten or so schemas) and constructor theory (possible and impossible (And Unknowable?) tasks) is on to something

Aleksandar

Ken Lloyd

unread,
Aug 28, 2018, 8:23:07 PM8/28/18
to syss...@googlegroups.com

The operative words possibly indicative of dysfunction are: “I claim”, “I think”, “I believe” or “that’s been mathematically proven (true)”.

 

It seems there are infinitely many more ways to misinterpret any statement than to get somewhat close to the mark. Great sport if you enjoy that game.

Janet Singer

unread,
Aug 29, 2018, 7:52:42 AM8/29/18
to syss...@googlegroups.com
Ken, I still do not understand the bright-line distinction you make between ‘philosophizing’ and ‘doing science’. 

It seems you are disparaging your own theory, which you ground in selecting four speculative ‘worlds’ posed by philosophers. 

How should those of us who find your four-world model useful deal with your cutting the legs out from under it?

Ken Lloyd

unread,
Aug 29, 2018, 10:05:40 AM8/29/18
to syss...@googlegroups.com

Janet,

 

It has become increasingly apparent that many in the ISSS community do not understand the “bright line distinction” between philosophy and science. Most scientific knowledge starts as belief – it just can’t end there. The philosophical sub-system (between the physical, mental and conceptual worlds) starts the processes. There is no speculation between the worlds. The concepts embodied in each of the worlds, acting as a meta-model to the others, can be used to cross-validate each other, using mutually couple mathematical and empirical subsystems in addition to philosophy. There is a transformation between the objects, their structure, behavior and morphism in each categorical world that, when validated and limited to the resulting “domains of validity” – or Domain.

 

For example, in the natural or real numbers, there are areas excluded from the domain of validity where certain mathematical operations in arithmetic yield non-valid results. We say that arithmetic is “almost everywhere” valid. Arithmetic is still useful and a valuable tool.

 

In this way, one might approach truth asymptotically over time, but validity is a necessary precursor to arriving at truths. There is an inherent problem with using truth as a direct objective in traversing the unknown into the known, and that is a form of deception in both the objective and the path to the objective. This deception was demonstrated, empirically, originally by Stanley and Lehman, and confirmed as “not-demonstrably-false” by many other scientists (including myself).

 

At this point, I see it futile to illuminate alternative views to the prevailing “orthodoxy” (which may not be “ortho”) dominated by philosophy. I see it as a game, more akin to a religion or cult because any challenge to the hermeneutics is met with immediate opposition rather than an attempt to see the logic underlying the reasons. I choose not to play that game. History will sort it out, eventually.

 

Play on.

Hillary Sillitto

unread,
Aug 29, 2018, 10:24:22 AM8/29/18
to syss...@googlegroups.com
Ken

Indeed, physics got a lot more interesting in the last 20-30 years since it started a) looking seriously, ) developing the experimental equipment and techniques to investigate, and c) co-developing with mathematicians the mathematical tools to explore quantitatively, non-equilibrium systems. 

There is a nice story about a nature photographer back in the 60s, who was told he must have faked a very-short-exposure photo of a bird flying through a narrow gap, because one of its wings was folded while the other was extended. Some aviation expert claimed it could not possibly fly like that. He was of course proved wrong...

Cheers

Hillary


(Mr) Hillary Sillitto, Edinburgh, UK

Janet Singer

unread,
Aug 29, 2018, 1:50:09 PM8/29/18
to syss...@googlegroups.com
Ken – 

1) Where can I see an example of ANYONE suggesting that science should END with philosophy rather than engage in an ongoing process of guessing, testing, evaluating, and subjecting to public criticism? This is not a novel insight. This is not controversial.

2) As I said, you undercut your own work by disparaging what you take to be philosophy and glorifying what you take to be science. Consistently applying the same criteria to your own speculations that you use to judge those of others would moderate and strengthen your position from “My views are scientific; those who don’t agree with me are philosophizing”. (As you know, a position of certainty is not capable of updating and learning.)

3) It appears your four worlds are objectively real to you, which is why you don’t see yourself engaging in philosophizing. Believing the products of one’s own imaginings have special ontic status over those of other people is, again, a religious or metaphysical position rather than a scientific one.

4) Efforts to find a mechanical rule for distinguishing ‘science’ from ‘non-science’ are futile (think Hilbert’s Entscheidungsproblem). Science is ultimately only characterized by the values and commitments of the people engaged in the scientific enterprise. It is what one obtains when a drive to improve understanding is combined with an acceptance of fallibilism, leading to ongoing testing and correction from reality plus willingness to subject work to critique by colleagues. You are to be commended for how much you have taught yourself about science (and math and philosophy), but failure to understand the core values of a community is a hazard of being an autodidact. If one is aware of that hazard one can take steps to address it.

Janet

Ken Lloyd

unread,
Aug 29, 2018, 7:45:35 PM8/29/18
to syss...@googlegroups.com

Hillary,

 

There has historically been a tension between physicists and philosophers. This was exacerbated when the physical sciences started using the formalism of mathematics along with empirical methods to develop their knowledge. It may not be obvious to many how deeper understandings emerge from “mathematized” encodings of concepts. This caused Eugene Wigner to refer to the “unreasonable effectiveness of mathematics in natural sciences”. There are times when scientific “truths” appear unreasonable.

 

To the uninformed, formal methods looks like hokus-pokus or magic. That, unfortunately, is a human condition I seem unable to either change or accept. I am resigned to accepting a bifurcation – a separate path – and letting history be the judge of which path was correct. With the maturing of AI technologies, and the concomitant advances in computing technologies (i.e. the integration of HPC and quantum, where HPC clusters avail themselves of CPUs and QPUs), “objects in the future are closer than they appear”. While it may take decades for the general public to become aware of this paradigm shift, that shift has already started – and well underway.

 

I frequently recall Eric Hoffer’s comment: “In times of dramatic change, the learners will inherit the earth while the learned will find themselves beautifully equipped to deal with a world that no longer exists.” I say, buckle up, it’s going to be a bumpy ride.

 

Ken Lloyd

Lenard Troncale

unread,
Aug 29, 2018, 9:11:22 PM8/29/18
to syss...@googlegroups.com
Dear SysSci Google Group: I have hesitated to engage in this lineage of comments on “What is Systems Science”, (i) despite two books appearing fairly recently from past leaders (Klir and Warfield) that had the titles of “Introduction to Systems Science” but neither were IMHO, (ii) because of systems thinkers claiming they are systems scientists but are not, and (iii) because of its and my own history of being slammed or even excommunicated for past comments and persistent critiques. But here goes:
  • I agree with Ken’s characterizing a “bright line” demarking the worldviews and techniques and tools and findings of science versus philosophy as well as his characterizing that the ISSS community does not understand the distinction (see below comments). (Janet, recall that his theory and the four worlds he described are still within the domain of philosophy and not between philosophers and science; so his comments on the distinction are possibly beyond that theory which is still useful to you).
  • Still notice that he maintains that a lot of science may begin as philosophical intuition and debate but must move beyond that to experiments and seeking evidence from the world outside human comprehension or neural nets. (Note also the importance here of the worldviews exposed in INCOSE & ISSS by Hillary Sillitto in his questionnaires on What is a System - they expose major differences between those who believe there is a world of objects beyond humans; and those who refuse to think there is anything real beyond humans) which positions deeply effect their attitudes to science, theory, models, and applications.
  • I lamented when decades ago ISSS became dominated by philosophers and so-called systems thinkers or application-dominant people, and most all scientists left it; an indication that the rigor of scientific testing and validity left it also (compared to endeavors like SFI, NECSI, and the ICCS series). But this allows and enables their “cult” of belief alone, as described next.
  • After a young student complained in the last ISSS Council meeting (ISSS’18 Corvallis, Oregon) that their graduate professors advised against the student even attending ISSS because it had lost its rigor and reputation, I was saddened to hear several past, present, and future leaders speak against the student by saying that that is exactly what they treasured about ISSS (its openness to all kinds of new ideas, whether valid or not). So rigor did not slip through our fingers, it was excluded consciously & vehemently.
  • In a paper at that very conference (poorly attended) I presented draft lists of criteria for science, criteria for a good theory, criteria for systems science, and criteria for a general systems theory (what we need are criteria debated to consensus to allow selection between theories and to enable distinctions between philosophy-based systems thinking and systems science) or there will not be the gradual improvement in consensus that Ken anticipates. And hasn’t been for the 50 years I have witnessed.
  • I also have delivered papers to both INCOSE CSER and ISSS on the differences between systems thinking and science to no avail.
  • I agree that both GST and systems work should first be based on solid science evidence; but also agree that such systems science is not exactly conventional science as much as beyond science (altho’ I tend to avoid the term meta-science).
  • I hope to agree with Rousseau’s claim that we are now past the “constructionist” anti-science philosophers and their widespread influence, but still detect a lot of that in both the ISSS and INCOSE in the knee-jerk, non-reductionist arguments of systems types. It is not EITHER/OR that is the essence. Either reductionist or not. It is BOTH-AND that is needed. Back and forth without losing one’s place.
  • So I still claim that systems science has a great deal to learn from the conventional sciences in order to become more of a systems science, and by condemning or excommunicating them is impoverishing both ISSS and INCOSE, and that submitting its claims to evidence and linking them to the vast data from experiments will increase the rigor and standing of systems science until it has a series of exemplar applications that clearly demonstrate its utility. Just the belief that systems approaches are the complete answer is insufficient and easily attacked as ultimately disappointing.
  • So what is systems science? Study and debate the criteria for determining judgement, and only then select alternatives (while still encouraging alternatives) for the best to our current understanding.
Len Troncale

Duane Hybertson

unread,
Aug 29, 2018, 10:42:32 PM8/29/18
to syss...@googlegroups.com
From: Hybertson, Duane W. <dhyb...@mitre.org>

Subject: RE: [SysSciWG] What is System Science?

 

Ken,

 

I did not quite understand your explanation of the bright line distinction, but I will just comment on your second sentence: “Most scientific knowledge starts as belief – it just can’t end there.” If scientific knowledge cannot end as belief, what does it end as? Certainty? Proof? Mathematical expression? I would characterize it differently, something like this: Scientific knowledge may begin in a variety of forms—idea, guess, conjecture, hypothesis, model, theory, supported by a degree of evidence ranging from none to some. Then through means that include observation, experimentation, reasoning, and debate, the initial form becomes either generally accepted knowledge or not. So to oversimplify a bit, it starts as weakly supported belief and ends as strongly supported belief. But even accepted knowledge is never certain or proven or final. It is accepted because a significant majority of the members of the given scientific field (and especially recognized leaders in the field) accept it as the best available explanation at some point, based on accumulation of evidence such as observation, experimentation, reasoning, debate, beauty, simplicity, consistency, and agreement with theory. Because of this uncertainty, scientific knowledge does not end; it is always susceptible to being replaced or significantly reinterpreted by a different framework, as Kuhn points out in his Scientific Revolutions. Nevertheless, even though scientific knowledge is never certain, and probably wrong to some degree, it is often useful—in terms of both making the world more coherent and supporting enterprises such as engineering and medicine.

 

Thanks,

Duane

Jack Ring

unread,
Aug 30, 2018, 5:24:44 AM8/30/18
to Sys Sci, Duane W. Hybertson
Say something about tense. Seems to me knowledge is limited to past tense while the application of the scientific method results in belief(s) about future tense. Justified true belief. Predictions are not expressions of knowledge. 
Make sense?
Jack

LAPLUME Yannick (SAFRAN)

unread,
Aug 30, 2018, 5:36:58 AM8/30/18
to syss...@googlegroups.com, Duane W. Hybertson

Hi everyone,

 

I never really thought about this aspect, but I think I would agree on the knowledge being linked to past tense : knowledge gathers all we already know. Beyond that, it is not knowledge itself, but the use of knowledge, for instance with scientific methods.

 

But for science, I think it is a bit broader than connection to future, as prediction of future. I think it is connected :

- to the past, because scientific models are validated against knowledge (past)

- to the past and present, to explain what was and what is (and I guess that this statement would relate to the discussion about what is science)

- to the future to estimate what will be (predictions as you say)

 

All three aspects (validation, understanding / explaining, and predicting) are I think equally important as a part of the definition of science.

 

Best regards,

 

Yannick LAPLUME
Complex Systems Engineering

P +33 (0)
1 61 31 84 01  • M +33 (0)6 76 61 00 16

Safran Tech / Modelisation & Simulation
Rue des Jeunes Bois, Chateaufort, CS80112
78772 Magny Les Hameaux Cedex
www.safran-group.com

logo_safran

SUBSCRIBE TOSAFRAN NEWS FEED

Twitter

Facebook

LinkedIn

 

 

De : syss...@googlegroups.com [mailto:syss...@googlegroups.com] De la part de Jack Ring
Envoyé : jeudi 30 août 2018 04:53
À : Sys Sci
Cc : Duane W. Hybertson
Objet : Re: [SysSciWG] What is System Science?

#
" Ce courriel et les documents qui lui sont joints peuvent contenir des informations confidentielles, être soumis aux règlementations relatives au contrôle des exportations ou ayant un caractère privé. S'ils ne vous sont pas destinés, nous vous signalons qu'il est strictement interdit de les divulguer, de les reproduire ou d'en utiliser de quelque manière que ce soit le contenu. Toute exportation ou réexportation non autorisée est interdite Si ce message vous a été transmis par erreur, merci d'en informer l'expéditeur et de supprimer immédiatement de votre système informatique ce courriel ainsi que tous les documents qui y sont attachés."
******
" This e-mail and any attached documents may contain confidential or proprietary information and may be subject to export control laws and regulations. If you are not the intended recipient, you are notified that any dissemination, copying of this e-mail and any attachments thereto or use of their contents by any means whatsoever is strictly prohibited. Unauthorized export or re-export is prohibited. If you have received this e-mail in error, please advise the sender immediately and delete this e-mail and all attached documents from your computer system."
#

Ken Lloyd

unread,
Aug 30, 2018, 6:52:29 AM8/30/18
to syss...@googlegroups.com

Duane,

 

It seems the lack of understanding – and the failure to see “the bright line” – is the distinction between belief (i.e. believing justified true belief is knowledge) and a completely different phenomenon, knowledge independent of belief, WRT science. This seems to be related to differences in the paradigms of “science”, only part of which might be attributed to philosophy. I contend that the paradigms people hold enable them to see part of the picture of reality, but constrain them from seeing the entirety of the picture. People won’t see what they cannot see. See the history of relativity theory or phlogiston theory for examples of this phenomenon.

 

One of the incorrect assumptions you have correctly identified is that there exists certainty in science. This rests on a fundamentally shaky concept from philosophy – that one can correctly ascertain “truth” through belief or reason. Our only hope in reaching truth and true knowledge is asymptotically approaching that state or condition over time. Usually, the precursors are: WFF --> validity (actually a system, see Inverse Theory), then over time if the domain of validity hasn’t proven false … maybe, then, a theorem or theory is true. Lynton Caldwell stated that the “role of science is to separate the demonstrably false from the probably true”. We can test that conjecture – a belief statement -- not accept it as true. Recall the Sokal Hoax.

 

You mention experimentation and facts. Let’s be honest with each other. There is very little actual experimentation, merely a lot of conjecture, anecdote and pontification in the systems sciences. Most so-called experiments are exercises in confirmation bias, framing and anchoring errors.  Any relationship (correlation) to facts and measurement should be very carefully considered. Most are superficial – lacking both breadth and depth into what constitutes actual (valid, but not necessarily true) knowledge.

 

I have contended that knowledge emerges from a complex system that often starts as insight, intuition or belief (from agency or intelligence). Here, look at the corpus of bifurcation theory. Just getting a group of experts together that agree with that belief – prima facie – isn’t science.  Richard Feynman noted “science is the belief in the ignorance of experts”, yet studies have shown experts notice “chunks” of meaningful patterns of information that others don’t see (NAP- How People Learn). That initially seems a paradox in a static system, but it often resolves in a complex system not-at-equilibrium.

 

This is contrary to the reductionist approach to science (and complexity).

 

So before we can adequately characterize “what is science?”, it seems we must first characterize what science isn’t, and why. Most of that work may have already been done.  I don’t see any honest attempt to accomplish that distinction. Maybe that’s due to my paradigm. Maybe not.

 

Ken Lloyd

Ken Lloyd

unread,
Aug 30, 2018, 7:06:19 AM8/30/18
to syss...@googlegroups.com

Rather than look upon knowledge as a state in the past tense, it might be useful to see knowledge as emergent (bifurcation) from a systemic process of a forward model (predictive, future) and inverse (refinement of the model from observation), under the umbrella term Inverse Theory (ala Tarantola, et al.). This seems to fit your description rather well, but it’s slightly more complex than that – there are a complex-coupled systems of inverse theoretic operations that mutually affect each other (which can be described in n-Category Theory).

 

I think Inverse Theory unfortunately named, because the system is not an arithmetic inversion, but represents a duality of opposite adjunctions (adjoint functors).

 

Ken Lloyd

image001.png
image002.jpg
image003.jpg
image004.jpg
image005.jpg
image006.jpg

Ken Lloyd

unread,
Aug 30, 2018, 7:16:49 AM8/30/18
to syss...@googlegroups.com

Belief (i.e. JTB) is usually the prior, not knowledge. The forward predictive model is part of the test. There is additional information (often in the error) from observation, that can be seen as new information on the priors, that can help refine the model (looking for convergence, which may point to a knowledge attractor ala Newton-Raphson or Runge-Kutta). If divergent, maybe look elsewhere.

 

Make sense?

 

Ken Lloyd

 

 

 

From: syss...@googlegroups.com [mailto:syss...@googlegroups.com] On Behalf Of Jack Ring


Sent: Wednesday, August 29, 2018 8:53 PM
To: Sys Sci <syss...@googlegroups.com>
Cc: Duane W. Hybertson <dhyb...@mitre.org>

Janet Singer

unread,
Aug 30, 2018, 8:16:11 AM8/30/18
to syss...@googlegroups.com
Can you clarify what physics you are referring to?

Hillary Sillitto

unread,
Aug 30, 2018, 11:42:53 AM8/30/18