Thank you for including me in your mail list and discussion.
I must commend you for the struggle and know from past experiences how tired
one can become trying to argue creation to those who have faith instead in
evolution.
I have a few comments to make about the recent discussion. First information
theory and the discusion on information content. There is a large body of
work that has already been done in this area due to the advent of genetic
algorithms (GAs). GAs are an interesting way to solve global optimization
problems. The basis of GAs are a few basic concepts borrowed from the
concepts of "evolution." The originator and creator of the seminal work is
Dr. John Holland, and his students. I would recommend searching this body of
work to prevent reinventing the information content wheel.
I myself am a creationist. I also am a Chemical Engineer, and a Knowledge
Engineer. I have used GAs to solve complex chemical plant optimizations. This
work has convinced me all the more that evolution is unlikely. In order to
create a system that actually evolves good answers is a difficult task in the
extream.
One must first get the encoding of the information to be evolved correct.
You can only evolve answers from the "chromosome" you design. If there is
any error here it will prevent anything interesting from showing up in the
results. Next there is the problem of how to cross-over the chromosomes to
preserve information and not destroy it. Finally random mutation is used to
search the state space of possible solutions.
It is important to note that mutation (randomnesss) is responsible for
searching the state space, but must be kept very small to prevent distruction
of the information being perfected by selection and crossover. The real work
of evolution is done by slection and crossover... not by random mutation.
The argement that mutation could not have caused evolution is in my view
somewhat miss guided, as mutation does only at most 3% or so of the work of
evolution. Selection does the rest. Any more mutation than that maximal value
and the genetic information you are trying to pass on the the next generation
is distroyed.
Now haveing said that... let me say again that the key to successfull
evolution is in the meta-data... the design and set up of the problem... the
conditions... the physical laws of crossover... the design of the chromosome
and and the careful use of miniscule mutation. All of these are done only by
intellegent design. Without this design nothing evolves. So even if
evolution were to be true, it would be the result of a very beautiful metat
data design.
Consider now the complexity of the human genome. In some posts we were
discussing simple strings like
THIS IS A EXAMPLE MESSAGE CONTAINING INFORMATION .
This example is simple to the extream. Do you know how many weeks of full
screens full of this sort of sentance it would take to view one complete
chromosome?
Lots of weeks. Imagine the design of this thing. Where the heck did it come
from? Some can say from space.. on a meteorite... but that just moves the
problem out into space. Where did it come from out there?
Now.. finally.. supose that evolution is a way to manipulate this incredible
design to investigate the potential for viable life. Does this prove that
that is all a fluke of chance? Not at all. Here is my example:
I design a chromosome to represent the combination of set points I might run
in my chemical plant. I design the criteria for sucess. (the optimization
objective function). This is a chemical process that takes about 4 hours to
line out at steady state so that I can get the results which are used to rate
the success of the chromosome representing the potential set points.
Now I run the GA. A typical problem is tackled by creating a population (say
300 individuals) of potential solutions .. chromosomes. I can not wait nor
will any plant manager let me actually jack his plant around to see the
results of these 300 test sets. So instead I create a math model of the
chemical plant and I run the 300 possible sets through my math model. I keep
the best.. kill the rest.. and allow the best to "bread" by crossover. I toss
in a tiny amount of mutation to assure the state space is searched. Each
such event Generate.. Test... Select... Crossover... Mutate is called a
generation. I must run 1000 or so generations to get a decent solution.
That means 1000 *300 = 300,000 potential plant steady state conditions I
have studied to arrive at a solution. Hopefully this effort has "evolved" a
nice set of set points for my plant that achieve my optimization objectives.
Now since the chemical plant takes 4 hours to line out, that means had I
actually physically done this in the plant it would have taken me 300,000 * 4
= 1,200,000 hours of chemical plant operating time to have conducted this
optimization experiment. That means 136 years of operation!
However, I did not run the plant for 136 years to find my solution... I
solved it in less than 15 minutes of compute time on my PC. I then walked
into the control room and instructed my operators to set these set points
here here and here. It all happend in less than an hour!
You see.. even if evolution is proven as a viable method of manipulating the
population it does not mean that a powerful God did not design the experiment
in his head... play out a zillion generations is that mind far more powerful
than a PC to find a pattern that was right for Earth, and then simply speak
the end results into existance complete as revieled in the book of Genesis.
Even if it is found to be true.. which I think highly unlikely because of
the complexity involved, it will in no means bother my faith... in fact it
will make it stronger that such a complex and wonderful design could have
beeen created in the first place.
One thing strikes me funny about evolution... usually we scientists are
interested in simplicity. Simplicity is the halmark of great discoveries.
However, evolution is so utterly complex as to be the anti-thesis of
simplicity. Yet it is still viewed somehow as science.
I do not see evolution as science or even as a theory. The old cliche
"Theory of Evolution" is a big miss-nomer. A theory at least can be
dissproved. Evolution can not be disproved... even if you do they adapt it
to include a new answer to the argument. A theory on the other hand can be
dis-proved. Thus evolution must be viewed as a faith only. A faith that can
not be proven or diss-proven. I am sure it will make many evolutionists out
there very angry for me to say they live by faith... but I believe they do.
In my view it is much simpler and therfore more scientific to believe in
creation. The complexity of evolution is too extream to be taken on faith
alone.
J.P. Havener
Principle Chemical Engineer and Knowlege Engineer
Pavilion Technologies Inc.