Ccogan: part 1: "literally infinite" problems

From: DNAunion@aol.com
Date: Sat Oct 21 2000 - 09:37:40 EDT

  • Next message: Susan Brassfield Cogan: "Re: To all uncommitted lurkers"

    > >Ccogan: Finally, I may as well point out that, if you understood the
    literally *infinite* richness that derives mathematically from the principle
    of repeated, cumulative variational branching, it's doubtful that you would
    claim that the theory is "simplistic."

    >>DNAunion: That is incorrect: there is not *literally infinite* richness
    produced by repeated cumulative variational branching. Had you said
    "infinite", in double quotes to indicate the word should not be taken
    literally, then your comment could be considered correct. But had you even
    said simply infinite, without double quotes, your statement would be wrong.
    And it is clearly wrong since you prefaced the word infinite with the word
    LITERALLY.
     
    >>Ccogan: My point is that there is *absolutely* no limit, in *principle*, to
    the degree of complexity that can be achieved by cumulative variational
    processes.
     
    >>DNAunion: Bold claim - now let's see if you can support it. Or will you try
    to wiggle out by solely relying on your "in *principle*" clause? If so, then
    please explain the principle that allows for the infiniteness.

    DNAunion: No response by Ccogan. From other of his statements, he is
    apparently relying on the "in principle" clause. But mathematically, in
    principle, a human could count by ones from 1 to
    999,999,999,999,999,999,999,999,999 - that is, there is no mathematical
    barrier in counting that high - but in reality, one could not do so. As my
    example makes clear (and should be obvious anyway), one cannot rely on such
    "in principle" mathematical truths (either his or mine) to apply in the real
    world.

    >>Ccogan: I should have emphasized that I was speaking in mathematical terms
    though. Oh, WAIT! I *did* emphasize that. Perhaps you missed the word
    "mathematical" only three words further along in the sentence?

    >>DNAunion: As I already pointed out, your use is incorrect BECAUSE you used
    the term mathematically. The possibilities are not INFINITE in the correct
    mathematical sense.

    >>Ccogan: Oh, but they are. Perhaps you forgot to take your math classes in
    school?

    DNAunion: Okay, Chris, are you now "Starting" with the derogatory personal
    comments? I have not said anything negative about your intelligence.

    >>Ccogan: Perhaps you took me to mean that there could in fact be an
    infinite number of *actual* variations created in a finite time or with a
    finite supply of materials. If that's so, then perhaps I did not make myself
    clear. I was not claiming any such thing.

    DNAunion: So, you are in fact stating that *in principle* the possibilities
    are infinite, but not *in reality*. You must then realize that only an
    *astronomically small* "fraction" of those infinite possibilities could
    *ever* occur - and even fewer in the finite number of years the universe has
    been in existence? If so, how does this help support an argument for
    evolution or the origin of life based on your proposed mechanism? So your
    original point was…..?

    >>DNAunion: Simple refutation. There are 20 amino acids. If they are peptide
    bonded into a 10,000 amino acid protein, then there are 20^10,000 possible
    unique arrangements of symbols (i.e., amino acids). This is many orders of
    magnitude larger than the estimated number of fundamental particles in the
    universe. But then there are 20 times MORE unique combinations that are have
    just one more amino acid in the chain. Then there are another 20 times MORE
    than that one when another single amino acid is added, and so on, and so on,
    and so on. All the possible unique combinations have not been hit, and never
    will, even if the universe gets to be trillions of trillions of trillions of
    trillions [you get the idea] years old.

    >>Ccogan: Actually, this does not refute my point at all. Obviously, since
    the amount of matter in the (known) Universe is finite, and the amount of
    time since the (probable) "Big Inflation" is finite, the *actual* number of
    possible combinations that can be achieved will necessarily be finite.
     
    >>DNAunion: Yep, just as I stated - your were wrong for saying "literally
    infinite".
     
    > >Ccogan: However, this is not particularly relevant to my point.
     
    >>DNAunion: Now the desperate wiggling and struggling to loose yourself from
    your own claim begins.
     
    > >Ccogan: My point is that there are no special limits to the complexity
    that can be achieved by the process that are inherent in the process itself.
    Any kind and degree of complexity that can be achieved by *any* process can
    also be achieved by a strictly evolutionary process of cumulative variation
    and replication, assuming that the materials are available for it.

    DNAunion: Again, if you are stating that there are no limits *in principle*,
    but there are in *reality*, then what is your point in regards to evolution?

     
    >>DNAunion: So show us that complete contents of latest issue of "Origins of
    Life and Evolution of the Biosphere" could come about by the process of
    NON-DIRECTED cumulative variation and replication. That would be a beginning.

    >>Chris: Okay, that's an easy one: Generate *every* possible string of
    characters of that length by a process of cumulative random variation.

    DNAunion: Nope - "Impossible". Using only capital letters A - Z and the
    numbers 0 - 9 (which leaves out %, (, ), ", ', ., ,, ? a, b, c, d, e, f, g,
    h, I, j, k, l, m, n, o, p, q, r, s, t, u, v, w, x, y, z, and all graphic
    information, etc.) there are 37 symbols to consider. The number of possible
    unique ways to arrange 96 of these symbols is about 3.5 x 10^150. Why is
    that number important? Because it is close to Dembski's estimate of the
    number of possible particle interactions since the birth of the universe (I
    just made a calculation in a few seconds on my calculator, using 10^80
    particles and 15 billion years for the age of the universe: (10^80 particles)
    x (4.7 x 10^17 seconds) = 4.7 x 10^97 "interactions between particles since
    the birth of the universe", which is far below Dembski's. So I believe his
    to be a good indicator.)

    What does that mean? That means (if I recall Dembski's algorithm off the top
    of my head) that if every elementary particle that ever existed in the
    universe interacted with another one at the rate of one per second, for the
    lifespan of the entire universe, that only 10^150 events ever occurred. And
    the equvilaent of this number is reached when a string of only 96 symbols is
    taken into consideration: add 4 more to make it an even 100, and the number
    of possible unique arrangements of the symbols must be multiplied by about
    another 2 million! And this goes for every increase of 4 symbols. So for a
    series of 200 symbols, there are over 10^313 possible unique sequences!
    Consider now that the issue I mentioned has about 100 pages, each of which
    probably contains on average, say 200 words at 5 letters each, and you end up
    with 100 pages x 200 words x 5 letters = 100,000 symbols. Thus, this number
    becomes the exponent, and there would be 37^100,000 possible unique
    sequences. There is no way that the specified sequence that comprises the
    issue of Origins of Life and Evolution of the Biosphere could come about by
    random chance - in the real world.

    Your "okay, that's an easy one" answer required that we generate *every*
    possible string of symbols about 100,000 units long. That is, in *reality*,
    impossible.

    This causes enormous problems for one of your claims from above:

    >>Chris: "Any kind and degree of complexity that can be achieved by *any*
    process can also be achieved by a strictly evolutionary process of cumulative
    variation and replication, assuming that the materials are available for it."

    We see in the Origins of Life and Evolution of the Biosphere a level of
    complexity that was produced (in just a couple months) by an intelligent
    process (which does count as *any* process) that could not be produced by
    your mechanism of choice. I can explain the specified information using only
    real world mechanisms, but you *must* rely on concepts involving the realm of
    the imagined.

     
    >>Chris: You can even start with a bit-string one bit long, if you want.
    Just keep the replication process going long enough to ensure that every
    possible string of the length of that work is generated.

    DNAunion: Sorry, but unless you expect the universe to exist for another
    quadrillion quadrillion quadrillion quadrillion quadrillion quadrillion
    quadrillion quadrillion quadrillion quadrillion …. years, then it is
    impossible to generate every possible 100,000-symbol sequence from a set of
    37 symbols. Your explanation exists in fairy-tale land, not in the reality
    we must deal with in scientific explanations for real world entities.

    >>Chris: Among the resulting strings there will be one that is that complete
    contents. You just have to find it.
    It would be easier if we could simply start out with a single bit and,
    treating like a number, keep incrementing the number until it was the last
    number possible to be represented in the same number of bits as the contents
    of the work in question. But, that might be considered a *directed* process.

    DNAunion: I will not even consider that possible objection: I don't need it.



    This archive was generated by hypermail 2b29 : Sat Oct 21 2000 - 09:38:03 EDT