|
Author
|
Topic: Tinkering Design
|
Pim van Meurs
Member
Member # 541
|
posted 07. December 2003 15:45
Evan raises a good point, if tinkering if something human then the plausible hypotheses to explain tinkering in the genome would appear to be limited to 'regularity, chance or humans'.
Does that mean any other intelligent designer is ruled out?
IP: Logged
|
|
Rex Kerr
Member
Member # 632
|
posted 08. December 2003 00:54
Given an omnipotent deity who doesn't need to tinker, it's not hard to imagine that said deity nonetheless does so for non-obvious reasons.
A Bayesian would nonetheless regard the deity-designer hypothesis as less probable if tinkering was observed.
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 08. December 2003 08:06
The trick is to distinguish what is scientifically avaliable in theory (from an ID perspective) from what is theology. An omni-everything deity could do anything, including having created the world last week and made it appear to have a past. These types of speculations are not relevant here.
But the question of the nature of the designer(s) as is revealed by the evidence for design is relevant. Of course, there is no one "theory of design." For those that believe it was front-loaded (even as early as the Big Bang), or that the basic cell was designed, deposited on earth, and left to evolve as circumstances dictated, the designer(s) were not tinkerers.
But for those that believe that design has happened often (such as for the creation of many biological systems and for the creation of new types of organisms, including perhaps all speciation), then the designer(s) appear most definitely to be tinkerers.
To argue that a designer might have reasons for tinkering even though they are not constrained to do so, as opposed to doing so because that tinkering reflects their nature, is not a scientifically valid approach. We have to look at and argue from the data we see, not from speculations about why the data might be deceptive due to metaphysical reasons.
IP: Logged
|
|
Micah Sparacio
Member
Member # 6
|
posted 08. December 2003 08:27
Pim,
Tinkering is clearly not limited to human beings. Rather, it seems to be a function of the following two conditions:
1. beings that act in time (e.g. ants, cats, humans)
2. beings with purposes/goals
[ 08. December 2003, 10:41: Message edited by: Micah Sparacio ]
IP: Logged
|
|
Steve Petermann
Member
Member # 884
|
posted 08. December 2003 10:19
It might be helpful at this point to unpack the metaphysical baggage that is, in some cases, tagging along with the notion of tinkering. For some, tinkering implies some sort of intervention in the regular operations of the design. In other words, the tinkerer lets things run along unattended until which time some change needs to be made. In the case of biological systems this seems to imply an external interjection of intelligence at various points. If this is the view, then there is an underlying metaphysics that, at least, should be made explicit which also may be difficult to support scientifically.
Ever since pre-Socratic times there has been a prevailing notion(particularly in science) that the entities within reality have intrinsic properties. This intrinsicality represents an ontological independence where the phenomenon that can be detected are attributed to the inherent properties of the entities. These intrinsic properties then give rise to the regularities that we see in nature. Nature becomes this ontolologically independent system with no real intentionality. Intelligent design in this type of system must occur through either a supernatural violation of these regularities or through an outside interface with the mechanisms of regularity to interject intelligence.
Today, however, this metaphysical view is being forcefully challenged. Perhaps starting with Bohr the idea of intrinsic properties has been brought into question. According to Bohr, "In the absence of measurement, physical entities have no intrinsic properties." This position is also being supported both by other physicists like Stapp: quote:
"There is no support in current fundamental physics for the notion that the physical world is made out of, or even contains, what Isaac Newton (1721) called “solid, massy, hard, impenetrable movable particles,” or that our conscious thoughts are patterns of motions of such entities." http://www-physics.lbl.gov/~stapp/BBSU1.pdf
and philosophers of the mind like Chalmers where he suggests that phenomenon are not the result of intrinsic properties but due to some(as yet unknown) consciously generated proto-phenomenon. http://www.u.arizona.edu/~chalmers/papers/nature.html section F.
Both of these positions shift from a materialistic position to some form of idealism. If these positions have merit, then the regularities and chance that we experience are, in fact, intentional, the product of the designer's "mind". Tinkering is no longer seen as a periodic "jumping in" but rather as part of a persistent mental process in the creation of both regularity and novelty. By definition then, Darwinian processes would be part of the intelligent design process. Tinkering might not be an appropriate descriptive term for this type of design process. It would also mean that the burden of proof would shift to those who reject intelligent design.
IP: Logged
|
|
Pim van Meurs
Member
Member # 541
|
posted 08. December 2003 12:36
Micah:
quote:
Pim,
Tinkering is clearly not limited to human beings. Rather, it seems to be a function of the following two conditions:
1. beings that act in time (e.g. ants, cats, humans)
2. beings with purposes/goals
Which also seems to show that it need to be limited to beings either. I was responding to Nobody's correction to your claim that
"All design implies tinkering" with "All human design implies tinkering."
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 08. December 2003 14:16
Steve Peterman provides an interesting version of reductionism. I have some more or less random remarks on both his posting and on the Schwartz, et al, paper that Peterman references, of which Stapp is second author.
Steve Peterman wrote quote:
Ever since pre-Socratic times there has been a prevailing notion(particularly in science) that the entities within reality have intrinsic properties. This intrinsicality represents an ontological independence where the phenomenon that can be detected are attributed to the inherent properties of the entities. These intrinsic properties then give rise to the regularities that we see in nature. Nature becomes this ontolologically independent system with no real intentionality.
In my view, as science became differentiated into disciplines - physics, chemistry, biochemistry, biology - the conception of "entities with reality" (and their associated ontological properties) also differentiated. Each level of analysis has its own set of fundamental entities with their own "intrinsic properties". The 'fundamental' entities of chemistry, say atoms, are composites of subatomic particles, and those subatomic particles in turn are composites of quarks. Now we hear that there may be a level of entities below that, superstrings. At each level, the "intrinsic" properties of the fundamental entities of that level are taken as a given. In the version of reductionism implicit in Steve Peterman's posting the intrinsic properties of fundamental entities at one level are explained by theories at the next lower level. The intrinsic properties of atoms are taken as a given in chemistry; but the chemically "intrinsic" properties of atoms that are relevant to chemistry are explained by theories in physics.
This analysis leads then to lower and lower levels, with 'intrinsic' properties of entities at each level explained by theories at a lower level. At the lowest level, we are told Chalmers concludes, there is "some (as yet unknown) consciously generated proto-phenomenon." (I have not yet read the Chalmers paper referenced in Peterman's posting.)
Note carefully that it is at the basal level, somewhere in the neighborhood of QM, where some sort of designing consciousness is invoked. Since the intrinsic properties of entities at levels higher than the basal level are explained by theories at lower levels, it is not until one reaches some basal level that something like "consciously generated proto-phenomen[a]" need be invoked. That puts design, if design there be, only at the very base of the intellectual and conceptual pyramid.
Two questions follow from this analysis. The first has to do with the nature of scientific explanation. Scientific explanations provide causal explanations of phenomena. A causal explanation identifies the variables of which a phenomenon is a function, and provides an account of why that phenomenon rather than some other phenomenon is observed. That is, a scientific explanation both explains what happens and why something else does not happen. The first question, then, is what explanatory leverage is gained by positing an "(as yet unknown) consciously generated proto-phenomenon" at the basal level of analysis? What does it explain, as opposed to merely name?
The second has to do with the burden of proof. That brings me to the Schwartz, et al, paper. First note that Schwartz, et al, do not rule out material entities, but only massy Newtonian particles, at the basal level. But that, of course, is trivial. Instead, they argue for a quantum account of the interaction between consciousness and material (brain) events. quote:
Does it make scientific good sense to try to understand the process of self-directed neuroplasticity solely in terms of brain mechanisms?
For at least one quite straightforward reason it seems clear that it does not. That reason is that it is intrinsically impossible to explain and describe to real people the techniques they must learn to perform and strategies required to initiate and sustain self-directed neuroplastic changes without using language that contains instructions about what to do with your mind, i.e., without using terms referring to mental experience, words like: feeling, effort, observation, awareness, mindfulness and so forth. (p. 15)
This is equivalent to the observation that it does not make good sense to "explain and describe" the burning of hydrogen in the presence of oxygen solely in terms of quarks. The level of explanation associated with quarks has no terms for oxygen, hydrogen, nor does it have processes that map directly onto oxidation. Hence one does not explain oxidation by appealing to quarks. That does not render oxidation inexplicable in material terms; it says the level leap from quarks to oxidation is too long.
This whole section of Schwartz, et al, is not about explanation, but about the language with which communications between humans occurs - in order for some concepts to be communicated to humans, one must employ human natural language with its references to intentionality, will, self, and so on. Schwartz, et al, implicitly take this to mean that those 'things' - intentionality, etc. - have some sort of objective reality independent of the language: That language refers to them somehow confers ontological status on them.
Schwartz, et al, interpret this to mean that the effects of that language in the recipient cannot in principle be explained by appeal solely to neural phenomena and that some other kind of explanation is required. That explanation is couched by Schwartz, et al, in terms of a consciousness, an 'observer' that affects the quantum probabilities associated with trillions of calcium channels at synapses.
Would a Geiger counter click if no human were watching it?
The Schwartz, et al, argument depends upon a combination of quantum theory and chaos theory. At base, it suggests that the macrostate into which the brain 'settles' can on occasion be determined by the 'choice' made at one calcium channel at one synapse, that single channel's 'choice' (to transmit or not transmit the calcium ion) propagating to a change in macrostate via a chaotic system's sensitive dependence on initial conditions. Consciousness can affect that 'choice' by playing the role of a human making an observation of a quantum system, thereby converting a quantum smear into a single outcome.
To illustrate the effect of an observer on a quantum system Schwartz, et al, describe the Quantum Zeno effect: quote:
Quantum Zeno has been verified experimentally many times. One of the neatest confirmations came in a 1990 study at the National Institute of Standards and Technology. There, researchers measured the probability that beryllium ions would decay from a high-energy to a low-energy state. As the number of measurements per unit time increased, the probability of that energy transition fell off; the beryllium atoms stayed in their initial, high-energy state because scientists kept asking them, "So, have you decayed yet?" The watched pot never boiled. As Sudarshan and Rothman concluded, "One really can stop an atomic transition by repeatedly looking at it." (pp. 50-51)
This is a question seeking information (I'm not a physicist): What would have happened to the probability of changing state of the beryllium atoms with repeated measurements if the measuring apparatus repeatedly interrogated the system, but human observers consulted the output of the measuring apparatus only at the start of the experiment and then not again until after some long interval, with the intervening series of measurements being repeatedly overwritten by the next measurement in the series? That is, were those beryllium atoms maintained in their high energy state because the scientists kept "asking" them their state, or because the measuring apparatus repeatedly (automatically) interrogated the system and irrecoverably threw away each measurement immediately after it was made? Is it the measuring apparatus/operation or the human knowledge of the outcome of the measurement operation that maintains the high-energy state? Schwartz, et al, interpret it to be the former; is that a necessary interpretation? (By the way, I myself think that Schrodinger's cat is either definitely alive or definitely dead, and we merely do not know which. I don't think human ignorance implies quantum superposition of states of macro-objects. But then, I'm a naive realist at heart. ![[Smile]](smile.gif) )
Understatement of the year: quote:
There is work to be done: hypotheses that fill in the missing details need to be formulated and tested. (p. 47)
Indeed!
More may come as and if I have time later.
RBH
IP: Logged
|
|
Steve Petermann
Member
Member # 884
|
posted 08. December 2003 16:19
RBH: quote:
Since the intrinsic properties of entities at levels higher than the basal level are explained by theories at lower levels, it is not until one reaches some basal level that something like "consciously generated proto-phenomen[a]" need be invoked. That puts design, if design there be, only at the very base of the intellectual and conceptual pyramid.
True but a base level causation does not preclude a top down source.
RBH:
quote:
In the version of reductionism implicit in Steve Peterman's posting the intrinsic properties of fundamental entities at one level are explained by theories at the next lower level. The intrinsic properties of atoms are taken as a given in chemistry; but the chemically "intrinsic" properties of atoms that are relevant to chemistry are explained by theories in physics.
Clearly some reductionist strategies have borne tremendous fruit in understanding higher levels. There are however, as many would affirm, points of diminishing return if reductive explanations are pushed too far.
RBH:
quote:
Two questions follow from this analysis. The first has to do with the nature of scientific explanation. Scientific explanations provide causal explanations of phenomena. A causal explanation identifies the variables of which a phenomenon is a function, and provides an account of why that phenomenon rather than some other phenomenon is observed. That is, a scientific explanation both explains what happens and why something else does not happen. The first question, then, is what explanatory leverage is gained by positing an "(as yet unknown) consciously generated proto-phenomenon" at the basal level of analysis? What does it explain, as opposed to merely name.
I suppose the same thing could have been said about the initial speculations of quantum theory or string theory. What they attempt to discover is the most logical and empirically sound explanation of the ultimate causation of phenomenon. Presumably the next step for these non-materialistic frameworks is to formulate a scientific strategy to provide a compelling case. Chalmers, for one, is doing just that.
RBH: quote:
This is equivalent to the observation that it does not make good sense to "explain and describe" the burning of hydrogen in the presence of oxygen solely in terms of quarks. The level of explanation associated with quarks has no terms for oxygen, hydrogen, nor does it have processes that map directly onto oxidation. Hence one does not explain oxidation by appealing to quarks. That does not render oxidation inexplicable in material terms; it says the level leap from quarks to oxidation is too long.
Granted the explanatory level of oxidation from quarks is minimal but surely an understanding of quarks offers some explanation of physical phenomenon. I don't think the explanatory power of reduction can be determined before hand. As an example, the discovery that problems with reuptake of neurotransmitters causes depression may not explain all the causal factors of depression but it does have some explanatory power.
RBH: quote:
This is a question seeking information (I'm not a physicist): What would have happened to the probability of changing state of the beryllium atoms with repeated measurements if the measuring apparatus repeatedly interrogated the system, but human observers consulted the output of the measuring apparatus only at the start of the experiment and then not again until after some long interval, with the intervening series of measurements being repeatedly overwritten by the next measurement in the series? That is, were those beryllium atoms maintained in their high energy state because the scientists kept "asking" them their state, or because the measuring apparatus repeatedly (automatically) interrogated the system and irrecoverably threw away each measurement immediately after it was made?
This is where I agree with your skepticism. Idealistic systems that appeal to human consciousness alone have difficulty dealing with these questions. However, there are idealistic systems dating back at least to pre-Socratic philosophy that may offer a more systematically compelling answer. While Anaximanes postulated the immutable quality of matter, he also claimed that the motion of all matter was driven by Nous which is Reason, Mind, or Thought. Ever since then there have been various versions of this where the "observation" as stated by quantum theory would not be just the human mind but a universal mind. Hegel called this "Absolute Mind". To quote a good summary from the web: quote:
According to Hegel, reality is Absolute Mind, Reason, or Spirit, which manifests itself in both natural and human history. This Mind is universal and therefore cannot be identified with the mind of any particular person. Rather, each particular mind is an aspect of this World Mind (Weltgeist), and the consciousness and rational activity of each person is a phase of the Absolute itself.
Presumably an "Absolute Mind" is not constrained by the presence of individual conscious minds. Instead, while not essential for the emergence of reality, individual consciousnesses that participate in the universal mind are, none the less, causal factors in that emergence.
While much of this is based on metaphysical speculations, the question is whether they provide a more compelling philosophical and scientific view. As with any reasonable speculation, if it shapes research then perhaps it is a positive thing.
IP: Logged
|
|
Rex Kerr
Member
Member # 632
|
posted 08. December 2003 17:38
Beings who act in time need not tinker. For example, they could find an optimal solution analytically, or they could employ approximate yet single-step methods such as associative neural networks that directly produce a solution.
Also, if I'm just fooling around with stuff for no particular reason, why isn't the outcome of what I am doing deserving of being called "tinkering", even if I had no purpose or goal--or at least not any which had any relevance to the functionality (or lack thereof) of what I produced?
I think I shall avoid the discussion of whether consciousness undergirds the most primitive subatomic particles, since despite being interesting to think about, it is highly speculative and apparently somewhat off topic.
Edited to change a period to a question mark. It really was a question, despite being so long that I had forgotten by the time I reached the end!
[ 09. December 2003, 00:57: Message edited by: Rex Kerr ]
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 08. December 2003 20:30
Continuing my somewhat disconnected remarks, Schwartz, et al, say that quote:
Process 1, which is connected to conscious awareness and to what actually happens, has itself two modes. The first involves mere passive consent, and temporally isolated events; the second involves mental effort, and a rapid sequence of Process 1 events that bring importantly into play the Quantum Zeno Effect. The first mode involves passive processing, and can exploit the massively parallel processing capacities of Process 2, whereas the second mode involves an effortfully sustained rapid linear sequence of Process 1 events.
Active Process 1 intervention has, according to the quantum model described here, a distinctive form. It consists of a sequence of intentional actions, the rapidity of which can be increased with effort. Such an increase in Attention Density, defined as an increase in the number of observations per unit time, can bring into play the Quantum Zeno Effect, which tends to hold in place both those aspects of the state of the brain that are fixed by the sequence of intentional actions, and also the felt intentional focus of these actions. Attention Density is not controlled by any physical rule of orthodox contemporary quantum theory, but is taken both in orthodox theory and in our model to be subject to subjective volitional control. This concordance of atomic physics and neurodynamics is the core of our model.
This notion of "conscious awareness" as a causal determinant of the state into which the (quantum state of the) physical material brain 'settles' is undermined by research on the reporting of intentions. On page 3 of the "Philosophy of Mind" thread I referred to a study where conscious mental intentions are not reported as precursors to actions in a decision task performed by patients with parietal lobe damage. According to Carl Zimmer's summary (I haven't read the original paper yet) quote:
[Recently, a European research group] ...ran Libet's experiment again, but some of the people they chose as their subjects had damage to certain parts of the brain. As they report in Nature Neuroscience, some kinds of brain damage make no difference to people's performance. But something fascinating happened to people who suffered damage to the parietal cortex, located at the back of the head. Like the healthy controls, they could nail the moment they actually pressed the button, to within a few milliseconds. But they also noted that they intended to press the button just around the time they actually did press the button. In other words, they were completely unconscious of their action until the action was already taking place.
That contrasts with undamaged persons, who reliably report the intention to act some time before it is performed, but some time (half a second) after EEG changes that presage the action occur. It's a little hard to see how some sort of Attentional Density, as Schwartz, et al, term the conscious effort that resolves the neural quantum state, can play a part in making a decision when it cannot be reported by persons with that damage, though their behavior in most respects is indistinguishable from undamaged persons. In Schwartz, et al's terms, those patients cannot report "Active Process 1" intervention, which is the conscious mental causal process that is supposed to influence the neural structures to 'settle' into some chosen state, yet they perform the task appropriately.
Recall that Schwartz, et al, invoke a conscious attentional effort by persons in explicit analogy with the Quantum Zeno effect described above. They even invoke an increase in the number of (mental, conscious) observations per unit time as the meaning of Attentional Density, exactly as in the beryllium experiment mentioned above. Yet without reportable consciousness of intention, parietal-damaged patients perform a task involving choice and intention.
Again, I caution that I haven't read the original paper yet, so I'm depending on Zimmer's summary and the abstract. Yet it seems clear that quantum neural states (if such there are) can get resolved and decisions can be made and executed without the involvement of reportable conscious attention (or intention).
RBH
IP: Logged
|
|
Micah Sparacio
Member
Member # 6
|
posted 09. December 2003 10:07
quote:
Beings who act in time need not tinker. For example, they could find an optimal solution analytically, or they could employ approximate yet single-step methods such as associative neural networks that directly produce a solution.
Also, if I'm just fooling around with stuff for no particular reason, why isn't the outcome of what I am doing deserving of being called "tinkering", even if I had no purpose or goal--or at least not any which had any relevance to the functionality (or lack thereof) of what I produced?
Hey Rex. Sure, beings in time need not tinker. However, to tinker *at all* requires time.
Second, maybe we should qualify the "tinkering" that we are talking about as "purposeful tinkering." The type of tinkering that takes place *when goals are in place*
Tinkering - physical changes that an agent causes over time in the service of achieving some target set of end product
IP: Logged
|
|
Rex Kerr
Member
Member # 632
|
posted 09. December 2003 12:52
Agreed--tinkering requires time. Luckily, we live in a temporal world, so this shouldn't ever be a constraint!
I am happy to accept goal-directed tinkering as the object under study, but I would caution that goal-directed and undirected tinkering may have many characteristics that are very similar. For example, if I fiddle around on my lab bench and produce a bunch of little widgets for fun, most of which don't do anything at all, and someone else comes through and picks out one that accomplishes a task they have but which I wasn't thinking of (e.g. my widget overheats, and they need a miniature heater), it's very hard to distinguish this from intentional tinkering based on observing the object (miniature heater) alone. I can't think of any good way to tell the two cases apart aside from observing my messy lab bench and realizing that I was just fooling around--and that the goal-directed intentionality was not particularly on the side of doing the work of tinkering, but in the selection of the undirectedly-tinkered object that had the desired properties.
Fixed a typo.
[ 11. December 2003, 03:36: Message edited by: Rex Kerr ]
IP: Logged
|
|
Stephen Wright
Member
Member # 195
|
posted 12. December 2003 17:29
quote:
Micah Sparacio wrote,
“Second, maybe we should qualify the "tinkering" that we are talking about as "purposeful tinkering." The type of tinkering that takes place *when goals are in place*”
Tinkering is a great word. Micah, I suggest that it can be rendered formal by defining it as a “search of design space.” In this way physical processes, not driven by a goal, can still tinker by realizing certain configurations. From the viewpoint of an intelligent agent who can discern patterns, the spiral nebula can be derived by the random formations of galaxies and seen in many versions and variations on a theme. Natural forces have explored the space available and created designs, without intent.
My point of comment would address the issue at hand from the viewpoint that information is real and capable of being altered by consciousness. The crux of this could be described by the term -Transformational Mode. Defined as: following the transformations, in a series of related events, noting the balanced exchanges making up the process. Modern physical science does exactly this. We understand all discrete events as balanced, in energy exchange and mass, when viewed before and after each process step.
However, information is only recently seen in this way. In this proposed conjecture regarding balanced transformation, the behavior of an intelligent agent has an actual effect, when focusing on events. Information would be found to be in four states in its relation to intelligent agents.
· Information can be strictly physical. There would be only random events and no control or guidance.
· Information can be potential. This happens when an intelligent agent makes an internal representation of a circumstance. This interaction changes probabilities of future events happening with control and guidance. It doesn’t change anything physically, in the environment.
· Information can become active with intent. An intelligent agent now combines this external information with the information of its local orientation and considers guidance and control. Here in this step, the search takes place at a mental level.
· Information active with intent can be physically realized. In this last state the external circumstance is bonded with the internal purpose of a local agent and control or guidance is integrated in an event. When is happens the potential information is fully reduced to a fact. Energy is spent in the choice and implementation of the intentional event. The result is an artifact where, to borrow from M. Polanyi, the intent indwells with the product of the event. Here the search is physical tinkering.
These additional states may relate to the problem where designs are tried but are disassociated from a potential successful use, not discovered until a future time. As designers or inventors search the design space where potential solutions are to be found, all four information states unfold themselves in discrete order. The Transformational Mode would enable the interaction of agent and it’s object of tinkering to categorize the interactions, in terms of process flow and overall equilibrium, just as we would a chemical reaction or machine operation.
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 12. December 2003 19:55
Stephen Wright wrote quote:
· Information can be potential. This happens when an intelligent agent makes an internal representation of a circumstance. This interaction changes probabilities of future events happening with control and guidance. It doesn?t change anything physically, in the environment. (Emphasis added)
and quote:
· Information active with intent can be physically realized. In this last state the external circumstance is bonded with the internal purpose of a local agent and control or guidance is integrated in an event. When is happens the potential information is fully reduced to a fact. [b]Energy is spent[/b[ in the choice and implementation of the intentional event. The result is an artifact where, to borrow from M. Polanyi, the intent indwells with the product of the event. Here the search is physical tinkering. (Emphasis added)
In the first case, is that internal representation (and the intelligent agent that bears it) unembodied, not instantiated in matter and energy? Unless it is unembodied, the "information" in that internal representation is actual, physical, not merely "potential." The step in the second quotation is then a transfer from one matter-and-energy instantiation to another, with the second being (perhaps) an amplified mapping of the first. (The amplification is provided by the involvement of external stuff in the mapping process, and the SLoT bill is mostly paid there, too.) What is it that "spends" the energy?
That is, this is either an account of the transformation of "information" from one physical instantiation to another, or it invokes a causal interaction between an unembodied representation and a physical representation with no notion of how non-physical unembodied "information" can push around the matter and energy necessary to create the physical realization. Which interpretation does Stephen prefer?
RBH
IP: Logged
|
|
gedanken
Member
Member # 594
|
posted 12. December 2003 23:39
Tinkering, as by a "designer", would imply a goal of some sort.
The problem in abstract concepts like that presented are with respect to variability of goals. Goals are a temporary property of an individual cohesive "designer". Multiple designers (remember MDT?) have multiple goals, and even single desingers have multiple goals over time or as their thinking processes change.
"Information" has been mentioned. But there is no absolute information, only relative to a particular measure system. (Related somewhat to the same aspect as "goal", e.g. "information" can only be measured in terms of some goal.) As interests (a kind of goal) change, then "information" of the exact same structure changes.
(Yes, there are "absolute" information structures by choosing some apparently absolute basis like thermodynamic entropy--but then "information" is equivalent to or transformable to entropy and obeys its physical constraints! And even that is a particular choice of measure system.)
So now one of the main reasons that designs are "tinkered" with is changing goals, not just a matter of the optimization process. This is why design shares such commonality with evolutionary processes.
In fact, (if one accepts an overall evolutionary processes), then "design" is simply a manifestation of evolutionary process--one that is complex in that the brain state of the desinger is an aspect of the larger evolutionary state.
Human design can be likened to a search process. So can "evolution", though the design is often considered to have a predefined "goal", while the evolution only has environmentally determined fitness. If one takes a designer's "goal" and makes the system act such that the goal has higher fitness, then we have an interesting comparison. (And that can be done by defining "goal" as the hiher fitness, or by modifying or controlling the system so that the goal has higher fitness--there is no distinction because goal==fitness in either case.)
In this case evolution and design become nearly indistinguishable. For one, "information" is thus measured in terms of the "goal" (suitably defined), and either system optimizes toward that "goal".
Another comparison, however, is when we consider the thinking process as a simulation process. The thinker (or thinking process) can pre-simulate the physical consequences of a design, by forming the design as a "specification" in some state system and operating on that state system in a way so as to make predicitons that can be related to the "goal" (once again suitably defined). Thinking can thus be likened to a simulated environment in which mutation and selection take place in simulation, prior to implemention in the physical world. While constraints on the thinking process will change the search pattern, the essential chacteristics of search "reachable states" is not changed by the precomputation or simulation capacity of thought, only probability distributions. Thus these are only distinguishable by the distributions--and distinguishing these require probability knowledge of the thinker or "designer" goals. That gets us back to "motive"!!! One can't escap "motive" in this issue.
When motive and thinking processes are unmodeled, then tinkering cannot be distinguished from evolutionary process.
[ 12. December 2003, 23:47: Message edited by: gedanken ]
IP: Logged
|
|
|
Printer-friendly view of this topic