On 9/30/2012 2:51 PM, meekerdb wrote:
On 9/30/2012 6:54 AM, Alberto G. Corona wrote:
Whoever said that does not know what he says:
"There are great differences between evolutionary designs and
rational design, rational designs are, well, rational, but
evolutionary designs are idiotic. Mother Nature (Evolution) is a slow
and stupid tinkerer, it had over 3 billion years to work on the
problem but it couldn't even come up with a macroscopic part that
could rotate in 360 degrees! Rational designers had less difficulty
coming up with the wheel. The only advantage Evolution had is that
until it managed to invent brains it was the only way complex objects
could get built."
First of all, 360 degrees rotation is present in the flagela of
the bacteria, invented about 3800 million years ago under intense
comet bombardement.
Did you miss the word "macroscopic"?
Hi Brent,
Could it be that "legs" are more efficient in generating motion
that can be directed than wheels? For example, how do you propose that
Nature would implement way to get blood and nerve signals across the gap
in the bearings that is necessary in some form of "wheel". It is hard
enough to get signals reliably across the boundary of the moving part of
a wheel and the axle in the steering wheel of a car. How do you do
implement an interface for liquids?
The main unstated assumption in this conversation is that organism
have to be mutually compatible to some degree with each other in order
for "living" to occur. Evolution that does not "jump" gaps. /*Natura non
facit saltus.*/
Sometimes your remarks demonstrate a remarkable lack of imagination.
Try to do it yourself in the same conditions ;). If there is no weel
in natural evolution is because legs are far superior.
And if you think that weels are superior, NS invented it, because the
invertor of the weel was a product of natural selection. Even your
feeling of superiority of the weel and the very feeling of
superiority of reason is a product of natural selection. The claim
of superiority of reason over nature is the last vestige of
unjustified antropocentrism in its most dangerous form: Pride and
self worship.
But NS couldn't 'invent' it for macroscopic size animals traveling on
hard smooth surfaces, because it had already 'invented' legs and there
was no evolutionary path from legs to wheels.
And second, with more relaxed mood, I have to say, as I said many
times here, that evolution works simultaneously with infinite
variables and problems at the same time: log term and short term.
Therefore we NEVER are sure of knowing in FULL the reasons
You mean the random events and the selection events - there are no
*reasons*.
That you can imagine, sure.
behind an evolutionary design and therefor we can not understand FULY
an evolutionary design. That gives evolutionary design an appearance
of mess poor design and so on. This is NOT the case. If evolution and
reason collide, the prudent is to consider that the reason don´t know
enough.
That is because Reason work to solve a single problem, Cognitive
scientist say that can handle no more than seven variables at the
same time for a single problem.
THAT is the reason WHY the human designs are made of modules with
discrete interfaces. No matter if we talk about architecture,
computer science or social engineering, Each rational design module
solves a single problem and comunicate with other modules in discrete
ways. This is what is considered "good designs" ,.
Not in general. Some engineers consider it good design to make
multiple uses of the same structure, e.g. there was a German
motorcycle that used the gear shift lever to also serve as the
kickstarter, more recently Ducati designed their racing motorcycles to
be 'frameless', using the engine as the structural member.
So what? You are assuming an intelligent entity in your argument
and complaining that Nature is stupid because it does not seem to do
things as nicely as some tiny cherry-picked selection of human
engineered designs. What is your purpose in this conversation? It is
certainly not to increase the understanding of the members of this list!
BUT THESE RULES OF GOOD DESIGN ARE A CONSEQUIENCE OF THE LIMITATIONS
OF REASON. Reason does not produce optimal solutions. it produce the
optimal solution that he can handle without breaking.
Not limitations, but the recognition that a modular design can be
easily changed to solve problems, usually by changing one module. But
with a non-modular design you may have to start over from scratch;
which is why evolution gets stuck on local maxima like legs and no
wheels. Incidentally Ducati's frameless design was a maintenance
nightmare, didn't handle well, and couldn't be modified. They went
back to a conventional frame.
Organism on Earth seem to have a basic structure that is modular,
in that each cell has a power supply, guidance system and other
components that, IMHO, match up well with Von Neumann universal
constructor
<http://en.wikipedia.org/wiki/Von_Neumann_universal_constructor>.
It helps to come to the conversation with an attitude that is not
happy to make perfection the enemy of the possible.
Natural selection takes the whole problem and produce the optimal
solution without modular limitations. Starting from scratch,
evolutionary algoritms have designed electronic circuits with a half
or a third of components, that are more fast that the equivalent
rational designs. As Koza, for example has done:
http://www.genetic-programming.com/johnkoza.html
These circuits designs are impossible to understand rationally. why?
because they are not modular. There is no division of the problem in
smaller problems. a transistor may be connected to more than one
input or output and so on. But they are better, ligther, faster. it
seems a "Bad" design but this is a subjective perception, as a
consequience of our rational inherent limitations.
It is not a casual that genetic algoritms are used whenever 1) it is
or very difficult to break a problem in parts 2) is easy to measure
how good a solution is.
I have used genetic-evolutionary algoritms for deducing the location
of extinction resources in a simulated firing. The algoritm deduced
the optimal location every time. the only problem is that we did not
know WHY this was the optimal solution.
But in genetic-evolution algorithms you have the luxury of adjusting
the randomness and of starting over from different initial values.
Sure, and you don't have to worry about being eaten as you adjust
parameters and reset simulations.
In the same way, an human organ can perform 3 4 5 functionalities at
the same time. the capillar tubes in a tree act as pumps, conducts,
architectural sustaining foundation and may be many more that still
we don´t know.
In the same way societies are subjects of evolution. A natural
socio-biological institution, like the family has many functions, far
more than the social engineers think. Its functions can not be
extracted away by public institutions ruled by social engineers
without a failure of the whole society.
Some functions can, e.g. slavery, and some have unintended consequences.
Why do you require perfect prescience? Consider how Sickle Cell
Anemia has been found to offer a limited immunity to malaria in humans
<http://en.wikipedia.org/wiki/Genetic_resistance_to_malaria>. Nature is
all about trade-offs and not about "perfect" schemata.
That is why conservatives rely on nature where progressives rely on
reason and this is the reason why the latter fail.
Conservatives are those who find themselves on top and call it 'natural'.
LOL, where did you find that definition? Almost all "people at the
top" are liberals. You might like to look at some demographic studies
... "Liberal with others peoples money..." Conservatives are an entirely
different kind of idiot.
But natural evolution does not start from scratch it has to modify
previous designs for new needs, while reason without the help of
tradiction, operates from scratch. But this is not a drawback but an
advantage for evolution,since the problems of the past may not happen
in the present, but probably they will happen in the future.
So now you are imagining foresight in evolution. Isn't it much
simpler to just recognize it is a constraint that random variation is
unlikely to change many things at once so as to 'start over with a
clean sheet'.
Could you write up a toy model of how that might happen?
Evolution design not for today, but from all times to al times,
taking into account past, present and probable future events.
So that's why 99.9% of all its designs have gone extinct? Evolution
doesn't design at all, it just randomizes and nature selects. The
waste is enormous, but there's no accounting department to care.
We agree on that!!!
Natural adaptation exceeds not only the capabilities of rational
design but also the rational understanding, as I have demonstrated
You've only demonstrated your own prejudice against reason.
No, maybe just a prejudice against your particular form of
"reason", Brent. Have you looked at Alberto's stuff? It is very
sophisticated. I have a few problems with it, but at least he is making
an honest attempt!
Evolution produces many designs that are suboptimal, because natural
selection only requires that a design be 'good-enough':
Arthropods make an elastic protein, rezulin, which is much more
elastic than that in molluscs (abductin) and in vertebrates (elastin).
Thus a fly can flap it's wings with less energy loss than a horse can
run or a scallop can swim.
And both of those implementation are using cells structures that
are very similar!
The nerves in a mammals eye join the photoreceptor cells at the end
facing the pupil. They run across the inside surface of the retina and
exit together. Where they exit they create a 'blind spot' in our field
of vision. So, first, they partly obstruct the retina and, second,
they create a blind spot.
Try to design an eye that uses liquid supplied nutrition (and
excreta) that works as well and integrates well with the rest of the way
that organisms are constructed. Please.
People swallow and breathe through a shared passage. A design that
results in many deaths due to choking on food.
Enough survive to reproduce.
Oxidation of fatty acids unnecessarily reverses the handedness of
methylmaloyl.
What biologically compatible mechanisms might prevent this?
In the biosynthesis of some plant alkaloids, reticuline is formed in
the S configuration and then inverted to the R configuration; a step
which could have been avoided by just using the S form for all the
alkaloids. Lysine is biosynthesized via two different parallel
processes when one would have served.
How might Nature segregate the enantiomers?
The designer of a larger and heavier vehicle on the softer surface
uses more wheels to avoid sinking into the surface. Yet among animals
the small ones, arthropods, have six or more legs; while the large
animals have two or four. Why don't large heavy animals like elephants
and rhinoceri have six or more legs?
To use a phrase from software: Backwards compatibility. Nature has
not found a way to get a liquid based nutrition and excretion system to
cross a rotation interface.
When animals with fur get cold little muscles attached to each hair
folicle cause the hairs to stand up thus thickening the fur and
increasing its insulative value. Humans also have these muscles which
cause 'goosebumps' when we're cold. This increases our surface area
and increases our loss of heat.
In the human male the urinary tract passes through the center of the
prostate gland; a gland given to swelling. This stupid design causes a
lot of pain and difficulty for older men.
Men where never meant to live so long!
--
Onward!
Stephen
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en.
