Difference between revisions of "Evolution"
(→Complexity Catastrophe) |
(→Adjacent possible) |
||
| (37 intermediate revisions by the same user not shown) | |||
| Line 3: | Line 3: | ||
==Context== | ==Context== | ||
| + | * It would seem that one of the high goals of any [[Identity Ecosystem]] is stability; that is what computer programers seem to require. Yet the real-world is not stable over any long period of time, and so there is no reason to suspect that the digital world will be either. This wiki page is about handling the constant need to update [[Identifier]] systems; as needs evolve, so will solutions evolve. | ||
| + | |||
*[[Evolution]] and [[Disruption]] are the ying and yang of the process of change in any ecosystem. In the [[General Theory of Living Systems]] the case is made that an identity ecosystem has all of the characteristics of living ecosystems. | *[[Evolution]] and [[Disruption]] are the ying and yang of the process of change in any ecosystem. In the [[General Theory of Living Systems]] the case is made that an identity ecosystem has all of the characteristics of living ecosystems. | ||
| Line 15: | Line 17: | ||
* In 1931 Wright explained evolution as a hill-climbing optimization process. <ref> S. Wright, ''Evolution in Mendelian populations 1931 Genetics '''16'''-97</ref> He described a "fitness landscape" where each mutation was evaluated by nature to determine it's fitness within that landscape. Our thinking itself has evolved since that time. While they understood that the landscape had many hills which could result in different subpopulations, they did not consider the need for big break throughs brought about by substantial mutations. | * In 1931 Wright explained evolution as a hill-climbing optimization process. <ref> S. Wright, ''Evolution in Mendelian populations 1931 Genetics '''16'''-97</ref> He described a "fitness landscape" where each mutation was evaluated by nature to determine it's fitness within that landscape. Our thinking itself has evolved since that time. While they understood that the landscape had many hills which could result in different subpopulations, they did not consider the need for big break throughs brought about by substantial mutations. | ||
| + | * An appropriate success metric is needed in order to evaluate various identifier methods. The metric should correspond to the peak of one of the hills, but could be as simple as mere survival. By that simple metric, experience teaches that the vast majority of identifier methods will fail. Some methods will succeed in very narrow fitness landscapes and.only a very few will experience the adoption rate of [[OpenID Connect]]. | ||
===A Brief History of Information and Identity=== | ===A Brief History of Information and Identity=== | ||
| Line 29: | Line 32: | ||
Now we are at a place in society were no known rules can tell us what to do next. We need to try new things and evolve new technologies that will continue to the next, unknown, new level of information and identity. | Now we are at a place in society were no known rules can tell us what to do next. We need to try new things and evolve new technologies that will continue to the next, unknown, new level of information and identity. | ||
| + | |||
| + | ===Taxonomies=== | ||
| + | Charles Darwin basically invented the concept of evolution with his book, "The Origin of Species", where species was a term created by Linnaeus to build his taxonomy of living creatures. Darwin was soon called to task for the simple reason that he did not actually define what a "species" actually was. There have since been learned professors who have spent their professional lives trying to complete the process of defining species and the evolution of one species into another. It has not worked. An process of categorization of living entities into clear buckets has never been complete. Evolution is gradual and the boundary for when two populations of living creatures has diverged sufficiently to be called a different species will always be a source of contention. So to with artificial life in the digital space. We create taxonomies of digital entities so that we can discuss them, but the boundaries will never be clear. We will just have to live with that. | ||
==Problems== | ==Problems== | ||
While change has been discussed by philosophers, at least since Heraclitus 500 years BC, human history has been marked by the search for invariants. First with religion and later with physics humans have focused on trying to find the causes of the way of the world. The book of Ecclesiastes in the Hebrew bible teaches that "there is nothing new under the sun." But the final chapter of the book of causes will not be written until the last bit of free energy is expended forcing changes to continue to create confusion among the various purveys of prophecies. Only then it will be true that "there is nothing new under the sun." Darwin was the first to create a successful theory of change that has since been applied to most areas of knowledge. Ernst Mach in 1896 phased the part played by accident (aka mutation) in evolution of all areas of knowledge with this quote. <ref>Ernst Mach, ''In the part played by accident in Invention and Discovery''. '''Monist 6''' (1896) p 174 (also Radnitzky ibid p. 63)</ref> | While change has been discussed by philosophers, at least since Heraclitus 500 years BC, human history has been marked by the search for invariants. First with religion and later with physics humans have focused on trying to find the causes of the way of the world. The book of Ecclesiastes in the Hebrew bible teaches that "there is nothing new under the sun." But the final chapter of the book of causes will not be written until the last bit of free energy is expended forcing changes to continue to create confusion among the various purveys of prophecies. Only then it will be true that "there is nothing new under the sun." Darwin was the first to create a successful theory of change that has since been applied to most areas of knowledge. Ernst Mach in 1896 phased the part played by accident (aka mutation) in evolution of all areas of knowledge with this quote. <ref>Ernst Mach, ''In the part played by accident in Invention and Discovery''. '''Monist 6''' (1896) p 174 (also Radnitzky ibid p. 63)</ref> | ||
| − | + | <blockquote>After the repeated survey of a field has afforded opportunity for the interposition of advantageous accidents, has rendered all the traits that suit with the word or the dominant thought more vivid, and has gradually relegated to the background all things that are inappropriate, making their future appearance impossible; then, from the teeming, swelling host of fancies which a free and high-flown imagination calls forth, suddenly that particular form arises to the light which harmonized perfectly with the ruling idea, mode or design Then it is that that which has resulted slowly as the result of a gradual selection, appears as if it were the outcome of a deliberate act of creation. Thus are to be explained the statements of Newton, Mozart, Richard Wagner, and others, when they say that thoughts, melodies, and harmonies had poured in upon them, and that they had simply retained the right ones.</blockquote> | |
| − | |||
We all, human and corporate, do need to make our way given the ecosystem that we do inhabit. The most successful are those that detect imbalances developing in the ecosystem and are able to anticipate changes that will thrive given other changes that are already underway. It is clear to most of us that change in the identity ecosystem is afoot. Our problem is to discern the solution that will profit from that change. | We all, human and corporate, do need to make our way given the ecosystem that we do inhabit. The most successful are those that detect imbalances developing in the ecosystem and are able to anticipate changes that will thrive given other changes that are already underway. It is clear to most of us that change in the identity ecosystem is afoot. Our problem is to discern the solution that will profit from that change. | ||
| Line 41: | Line 46: | ||
# a school or work environment were both the user and the various relying parties are first registered with the school or employer before authentication is attempted. | # a school or work environment were both the user and the various relying parties are first registered with the school or employer before authentication is attempted. | ||
# a social setting where a user needs to authenticate to many different relying parties that are not known in advance. | # a social setting where a user needs to authenticate to many different relying parties that are not known in advance. | ||
| + | As we describe below, any fitness landscape can have its own population of solutions only if that landscape provides an incentive to create new solutions. | ||
| + | ===Adjacent possible=== | ||
| + | The concept of the “adjacent possible” was introduced by theoretical biologist Stuart Kauffman. It describes the idea that at any given moment, there are a set of possibilities that are just one step away from the current reality. These possibilities represent the next steps that can be taken based on the current state of knowledge, technology, and circumstances. Finding new uses for old artifacts (jury-rigging) is not a computation. "We cannot deduce what is 'in' the adjacent possible that the evolving biosphere creates and becomes. We do not even know what can happen. For 500 years we have thought the world is a machine. It is not."<ref>Stuart Kauffman, ''The "Adjacent Possible" – and How It Explains Human Innovation'' TED Talk https://www.youtube.com/watch?v=nEtATZePGmg</ref> | ||
===Complexity Catastrophe=== | ===Complexity Catastrophe=== | ||
| − | Stuart Kauffman describes the situation<ref>Stuart A. | + | Stuart Kauffman chapter 2 describes the situation<ref name=Kauffman>Stuart A. Kauffman, ''The Origins of Order'' (1993) Oxford ISBN 019507951-5</ref> where "the inexorable onset of a novel Complexity Catastrophe which limits selection. It it the consequence of attempting to optimize in situations of increasingly many conflicting constraints among components: Accessible optima become ever poorer, and fitness peaks dwindle in height." This is the situation we have seen in identifier standards over and over. As a standards effort attracts more participants with a wide variety of needs, the resulting solutions are not optimal for any one solution, but just a muddle of many inputs that satisfies no one. This was the fate of SAML and the first two attempts at OpenID. It was repeated starting in 2021 with self-sovereign identifiers. After the great effort failure to their own over ambitious goals, a simpler solution will eventually arise which just tries to solve one problem well. |
| + | |||
| + | Another way of saying the same thing is that the very idea of a level playing field where no new idea has any advantage over old ideas will effectively end the evolutionary advance of improving the [[Identity Ecosystem]] or even the Digital Ecosystem. | ||
| + | |||
| + | ===Catastrophic Change=== | ||
| + | What can happen when lots of small changes (as described in the above section) fail to find better performance, eventually a much large change comes along which makes a very significant change which sometimes will cause many of the older patterns to fail is a major way. | ||
| + | |||
| + | ===Homoplasy=== | ||
| + | A convergence of two or more otherwise divergent paths on the same solution or morphology. Generally this means that all these paths are climbing the same hill. Or that the paths are converging on the same solution. This is nearly the opposite of the complexity catastrophe. | ||
| + | |||
| + | Homoplastic characters can distort patterns of relationship indicated by homologous characters because they are often in conflict with them. Octopus eyes and human eyes are homoplastic; they develop very differently and have evolved independently in two remotely related groups of animals. | ||
==Solutions== | ==Solutions== | ||
| − | Given that we cannot know, in advance, which solutions will succeed in the identity ecosystem. We still can lay out some [[Framework]] where solutions might be expected. Solutions in identity ecosystems must be advantageous to three types of entity, where all interesting decisions are generated by the first two: | + | Given that we cannot know, in advance, which solutions will succeed in the identity ecosystem. We still can lay out some [[Framework]] where solutions might be expected. Solutions in identity ecosystems within the social fitness landscape must be advantageous to three types of entity, where all interesting decisions are generated by the first two: |
#Users or Consumers with a large bundle of desires for goods or service and fear of the unknown. | #Users or Consumers with a large bundle of desires for goods or service and fear of the unknown. | ||
#Resources of goods or services that are relying on the identity of those users for their survival. | #Resources of goods or services that are relying on the identity of those users for their survival. | ||
#Service providers to either users or relying parties who are evolving to meet the needs of both. | #Service providers to either users or relying parties who are evolving to meet the needs of both. | ||
| + | ===Long jump Adaptions=== | ||
| + | Stuart Kauffman chapter 3 <ref name=Kauffman /> describes an evolutionary process that can make larger changes than local hill climbing. The the wiki page on [[Resilience#Avoiding_Risk|Resilience]] there is a description about how Risk Taking, far from avoiding [[Resilience]] is likely required for any long term survival, | ||
| + | |||
| + | ===Models=== | ||
| + | At first Evolution scientists were drawing on models developed from the physical sciences, like hill climbing and simulated annealing. | ||
| + | |||
| + | through out the 1990's [[Genetic Programming]] started by taking models from evolution scientists, but then the two schools began to share models back and forth. | ||
==References== | ==References== | ||
Latest revision as of 12:43, 6 September 2024
Contents
Full Title or Meme
The gradual process of the changing morphology and capability of living entities in a natural ecosystem.
Context
- It would seem that one of the high goals of any Identity Ecosystem is stability; that is what computer programers seem to require. Yet the real-world is not stable over any long period of time, and so there is no reason to suspect that the digital world will be either. This wiki page is about handling the constant need to update Identifier systems; as needs evolve, so will solutions evolve.
- Evolution and Disruption are the ying and yang of the process of change in any ecosystem. In the General Theory of Living Systems the case is made that an identity ecosystem has all of the characteristics of living ecosystems.
- All evolutionary change occurs within a living ecosystem. Each change must improve the success of the species in the exiting ecosystem. Disruptions do occur, but even they must succeed in the existing ecosystem, even as they change it.
- Amara's law[1] tells us that we tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run. Since in the long run we are all dead (including all of the known enterprises), the short run is where we have the most opportunity to improve our odds of success, as well as the place where mistaken judgement will result in no impact on the evolving ecosystem.
- Unicorns are those special enterprises that are early to provide a solution for long term change, but they will need some means to find a way to survive until the big change comes. Their concerns are not addressed here.
- The field of Evolutionary Epistemology has developed in the past century[2] as an Framework of the manner in which Evolution applies to the theory of all knowledge, which, perforce, includes the identity of objects of our ecosystems.
- In 1931 Wright explained evolution as a hill-climbing optimization process. [3] He described a "fitness landscape" where each mutation was evaluated by nature to determine it's fitness within that landscape. Our thinking itself has evolved since that time. While they understood that the landscape had many hills which could result in different subpopulations, they did not consider the need for big break throughs brought about by substantial mutations.
- An appropriate success metric is needed in order to evaluate various identifier methods. The metric should correspond to the peak of one of the hills, but could be as simple as mere survival. By that simple metric, experience teaches that the vast majority of identifier methods will fail. Some methods will succeed in very narrow fitness landscapes and.only a very few will experience the adoption rate of OpenID Connect.
A Brief History of Information and Identity
This taxonomy of history is derived from one described by Popper.[4]
- Hydrogen and Helium fuse to form the heavier elements by rules that physicists claim to have mastered.
- The world is formed as an identifiable object by rules that physicists claim to have mastered.
- Liquids, crystals and other material objects are formed by rules that physicists claim to have mastered.
- Life forms and the ecosystem is altered in ways that no rules are known to predict. Although Chaitin claims to have made biology Mathematical, it is not predictive.[5]
- Animals become sentient and can identify one individual, and one species from all the others. Linneas creates a taxonomy of identities.
- Human consciences learns of the permanence of identity and its loss to the world at death.
- Human societies form and create languages and theories about identity and death.
- Human societies create a surplus which allows some individuals to specialize in the arts and sciences, including the meaning of identity.
- Human's invent thinking machines, which they claim are the result of intelligent design, but experience proves otherwise.[6]
Now we are at a place in society were no known rules can tell us what to do next. We need to try new things and evolve new technologies that will continue to the next, unknown, new level of information and identity.
Taxonomies
Charles Darwin basically invented the concept of evolution with his book, "The Origin of Species", where species was a term created by Linnaeus to build his taxonomy of living creatures. Darwin was soon called to task for the simple reason that he did not actually define what a "species" actually was. There have since been learned professors who have spent their professional lives trying to complete the process of defining species and the evolution of one species into another. It has not worked. An process of categorization of living entities into clear buckets has never been complete. Evolution is gradual and the boundary for when two populations of living creatures has diverged sufficiently to be called a different species will always be a source of contention. So to with artificial life in the digital space. We create taxonomies of digital entities so that we can discuss them, but the boundaries will never be clear. We will just have to live with that.
Problems
While change has been discussed by philosophers, at least since Heraclitus 500 years BC, human history has been marked by the search for invariants. First with religion and later with physics humans have focused on trying to find the causes of the way of the world. The book of Ecclesiastes in the Hebrew bible teaches that "there is nothing new under the sun." But the final chapter of the book of causes will not be written until the last bit of free energy is expended forcing changes to continue to create confusion among the various purveys of prophecies. Only then it will be true that "there is nothing new under the sun." Darwin was the first to create a successful theory of change that has since been applied to most areas of knowledge. Ernst Mach in 1896 phased the part played by accident (aka mutation) in evolution of all areas of knowledge with this quote. [7]
After the repeated survey of a field has afforded opportunity for the interposition of advantageous accidents, has rendered all the traits that suit with the word or the dominant thought more vivid, and has gradually relegated to the background all things that are inappropriate, making their future appearance impossible; then, from the teeming, swelling host of fancies which a free and high-flown imagination calls forth, suddenly that particular form arises to the light which harmonized perfectly with the ruling idea, mode or design Then it is that that which has resulted slowly as the result of a gradual selection, appears as if it were the outcome of a deliberate act of creation. Thus are to be explained the statements of Newton, Mozart, Richard Wagner, and others, when they say that thoughts, melodies, and harmonies had poured in upon them, and that they had simply retained the right ones.
We all, human and corporate, do need to make our way given the ecosystem that we do inhabit. The most successful are those that detect imbalances developing in the ecosystem and are able to anticipate changes that will thrive given other changes that are already underway. It is clear to most of us that change in the identity ecosystem is afoot. Our problem is to discern the solution that will profit from that change.
Fitness Landscape
There are many varied landscapes where an identifier is required to be authenticated. It is natural (like nature) to expect that different solutions will be optimized for each landscape. Some landscapes that need solutions might be:
- a school or work environment were both the user and the various relying parties are first registered with the school or employer before authentication is attempted.
- a social setting where a user needs to authenticate to many different relying parties that are not known in advance.
As we describe below, any fitness landscape can have its own population of solutions only if that landscape provides an incentive to create new solutions.
Adjacent possible
The concept of the “adjacent possible” was introduced by theoretical biologist Stuart Kauffman. It describes the idea that at any given moment, there are a set of possibilities that are just one step away from the current reality. These possibilities represent the next steps that can be taken based on the current state of knowledge, technology, and circumstances. Finding new uses for old artifacts (jury-rigging) is not a computation. "We cannot deduce what is 'in' the adjacent possible that the evolving biosphere creates and becomes. We do not even know what can happen. For 500 years we have thought the world is a machine. It is not."[8]
Complexity Catastrophe
Stuart Kauffman chapter 2 describes the situation[9] where "the inexorable onset of a novel Complexity Catastrophe which limits selection. It it the consequence of attempting to optimize in situations of increasingly many conflicting constraints among components: Accessible optima become ever poorer, and fitness peaks dwindle in height." This is the situation we have seen in identifier standards over and over. As a standards effort attracts more participants with a wide variety of needs, the resulting solutions are not optimal for any one solution, but just a muddle of many inputs that satisfies no one. This was the fate of SAML and the first two attempts at OpenID. It was repeated starting in 2021 with self-sovereign identifiers. After the great effort failure to their own over ambitious goals, a simpler solution will eventually arise which just tries to solve one problem well.
Another way of saying the same thing is that the very idea of a level playing field where no new idea has any advantage over old ideas will effectively end the evolutionary advance of improving the Identity Ecosystem or even the Digital Ecosystem.
Catastrophic Change
What can happen when lots of small changes (as described in the above section) fail to find better performance, eventually a much large change comes along which makes a very significant change which sometimes will cause many of the older patterns to fail is a major way.
Homoplasy
A convergence of two or more otherwise divergent paths on the same solution or morphology. Generally this means that all these paths are climbing the same hill. Or that the paths are converging on the same solution. This is nearly the opposite of the complexity catastrophe.
Homoplastic characters can distort patterns of relationship indicated by homologous characters because they are often in conflict with them. Octopus eyes and human eyes are homoplastic; they develop very differently and have evolved independently in two remotely related groups of animals.
Solutions
Given that we cannot know, in advance, which solutions will succeed in the identity ecosystem. We still can lay out some Framework where solutions might be expected. Solutions in identity ecosystems within the social fitness landscape must be advantageous to three types of entity, where all interesting decisions are generated by the first two:
- Users or Consumers with a large bundle of desires for goods or service and fear of the unknown.
- Resources of goods or services that are relying on the identity of those users for their survival.
- Service providers to either users or relying parties who are evolving to meet the needs of both.
Long jump Adaptions
Stuart Kauffman chapter 3 [9] describes an evolutionary process that can make larger changes than local hill climbing. The the wiki page on Resilience there is a description about how Risk Taking, far from avoiding Resilience is likely required for any long term survival,
Models
At first Evolution scientists were drawing on models developed from the physical sciences, like hill climbing and simulated annealing.
through out the 1990's Genetic Programming started by taking models from evolution scientists, but then the two schools began to share models back and forth.
References
- ↑ Roy Amara, Wikipedia 1972 https://en.wikipedia.org/wiki/Roy_Amara
- ↑ Gerard Radnitzky +1 editors, Evolutionary Epistemology (1987) ISBN 0-8126-9037-7
- ↑ S. Wright, Evolution in Mendelian populations 1931 Genetics 16-97
- ↑ Karl Popper +1, The Self and Its Brain. (1977) ISBN 0-415-05898-8
- ↑ Gerogory Chaitin, Proving Darwin: Making Biology Mathematical. (2012) ISBN 978-0-375-42314-7
- ↑ Edward E. Lee, Is Software the Result of Top-Down Intelligent Design or Evolution? (2018-09) CACM 61 No 9
- ↑ Ernst Mach, In the part played by accident in Invention and Discovery. Monist 6 (1896) p 174 (also Radnitzky ibid p. 63)
- ↑ Stuart Kauffman, The "Adjacent Possible" – and How It Explains Human Innovation TED Talk https://www.youtube.com/watch?v=nEtATZePGmg
- ↑ 9.0 9.1 Stuart A. Kauffman, The Origins of Order (1993) Oxford ISBN 019507951-5
