At 03:37 PM 12/18/2000 +0100, you wrote:
>Silk here: Could Be???????????????????
>
>Information: A modern scientific design argument
>
>All the design in living things is encoded in a sort of recipe book
>with lots of information. Information describes the complexity of a
>sequence - it does not depend on the matter of the sequence. It could
>be a sequence of ink molecules on paper (book) - however the
>information is not contained in the molecules of ink but in the
>patterns. Information can also be stored as sound wave patterns (e.g.
>speech), but again the information is not the sound waves themselves;
>electrical impulses (telephone); magnetic patterns (computer hard
>drive).
>
>The anti-theistic physicist Paul Davies admits: `There is no law of
>physics able to create information from nothing' (this issue, p. 42).
Chris
Nor does it ever *need* to be created. With different emphasis:
Nor does it *ever* need to be created.
There is no law of *anything* able to create information from nothing. In
fact, there is no law of anything that requires that information be
created, period.
>Information scientist Werner Gitt has demonstrated that the laws of
>nature pertaining to information show that, in all known cases,
>information requires an intelligent message sender, a conclusion
>rejected by Davies on purely philosphical (religious) grounds.
Chris
Besides, it's utterly false on empirical grounds. The Sun sends information
in every photon, as does an atom that releases a photon when an electron
drops into a lower energy level. Either the author of the article got Gitt
wrong, or Gitt is an absolute idiot.
>Thus a
>modern version of the design argument involves detecting high
>information content. In fact, this is exactly what the SETI project
>is all about - the Search for ExtraTerrestrial Intelligence involves
>trying to detect a high-information radio signal, which they would
>regard as proof of an intelligent message sender, even if we had no
>idea of the nature of the sender.
>
>In living things, information is all stored in patterns of DNA,
False. It is stored in every electron, neutron, and proton in the organism,
*and* in the exact energy levels of each, *and* in the exact organization
of all of these particles, *and* in the *actions* of these particles.
>which
>encode the instructions to make proteins, the building blocks for all
>the machinery of life. There are four types of DNA `letters' called
>nucleotides, and 20 types of protein `letters' called amino acids. A
>group (codon) of 3 DNA `letters' codes for one protein `letter'. The
>information is not contained in the chemistry of the `letters'
>themselves, but in their sequence. DNA is by far the most compact
>information storage/retrieval system known.
By what measure? This seems extremely doubtful, particularly since the
methods of compressing information that are in common use in computers do
not even *approach* the compression of known methods of storing data. I'd
be willing to bet that the information storage of DNA could be increased by
about an order of magnitude if it could use the techniques that could be
applied in a good computer software system. But, because the code is
linear, and does not provide the subroutine calling facility of any
computer, what would be or could be put into a single subroutine and
invoked where needed often has to be duplicated in DNA.
One of the things I hate about these cutesy hyper-popularizations of
scientific and technological discoveries and inventions is that most of the
absolutely critical details are left out, so there is no way the reader can
possibly tell which of perhaps several possibilities are really intended by
such statements as "DNA is by far the most compact information
storage/retrieval system known." Does it mean in *physical* terms (i.e.,
DNA vs. computer hard drives?). Does it mean in terms of ratio of stored
bits to final output bits (in which case, the formula for the Mandelbrot
set or the formula for calculating pi has it beat *infinitely*)?
It appears from the remarks below that what is meant above is purely
*physical* capacity. But, even then, it is not true. IBM, several years
ago, was able to place *individual* atoms at specific locations on a
substrate. It was not efficient from other points of view, but the amount
of space taken by such information storage would be many times smaller than
that of DNA, which uses several atoms per nucleotide. Even if we simply use
each nucleotide as a two-bit information holder (one combination of two
on-off states represented by each nucleotide), we are still using several
atoms per bit. The IBM technology could easily use the space of *one* atom
per bit (and, for about half of the bits, it would not need the atom at
all). The *actual* storage space needed in reality involves at least the
entire volume of the cell, which is needed to "package" the
information-storing DNA. Chip technology, of course, currently requires a
larger substrate, but Hewlett-Packard and others have been working for some
time on molecular storage and processing technologies, so, even in terms of
sheer physical compactness, DNA's current status is not likely to hold up
much longer.
>Now consider if we had to write the information of living things in
>book form. Dawkins admits, `There is enough information capacity in a
>single human cell to store the Encyclopaedia Britannica, all 30
>volumes of it, three or four times over. Even the simplest living
>organism has 482 protein-coding genes of 580,000 `letters'.
Chris
I don't think this is an "admission." I think it's an *assertion*.
>Let's suppose we had the technology to go the other way, and store
>books' information in DNA - this would be the ideal computer
>technology.
Chris
No, it wouldn't, because it is too unstable. The lack of stability is good
for evolution, but not good for storing stuff we want to *stay* the same no
matter how many times it's copied. A simple cosmic ray can screw it up so
that succeeding generations of copies all have the error caused by the
cosmic ray in them.
>The amount of information that could be stored in a
>pinhead's volume of DNA is equivalent to a pile of paperback books
>500 times as tall as the distance from Earth to the moon, each with a
>different, yet specific content. Putting it another way, a pinhead of
>DNA would have a billion times more information capacity than a 4
>gigabyte hard drive.
Chris
This is impressive, I admit. But, keep in mind that life probably arose at
the molecular level, whereas we are starting at a macroscopic level and
having do devise technologies that are smaller and smaller. If you compare
the rates of progress, computer technology is progressing at a rate many
thousands of times faster. Further, to store the same amount of information
in functional terms, computer technology does not need so much repetition
(as indicated above).
>Just as letters of the alphabet will not write the Annals of Ennius
>by themselves, the DNA letters will not form meaningful sequences on
>their own.
Invalid analogy. To make the analogy correctly, try this:
Just as letters of the alphabet will not write meaningful text
by themselves, the DNA letters will not form meaningful sequences on
their own.
The problem is, when the analogy is corrected, it no longer supports its
claim. Further, no replication mechanism is specified for the letters. If
we simply copy a string of letters and allow an occasional change, such as
replacing, deleting, or adding a letter (or space, etc.) they *will* form
meaningful text. It is all but absolutely impossible for them not to, given
enough repetitions of the process, even if we start with a "string" of
letters containing no letter at all, or even a space. If the copying
mechanism merely counts off repetitions of this nothing but occasionally
adds a letter, it is trivially easy to demonstrate that it *will* generate
meaningful text.
>And just as the Annals would be meaningless to a person
>who didn't understand the language, the DNA `letter' arrangements
>would be meaningless without the `language' of the DNA code.
True, but so what? The language is built into the laws of chemistry. For
the particular molecules and structures of molecules involved, there cannot
be any other "language." And, obviously, if the processing mechanism does
not understand the language of any particular genome, it will not be
reproduced, so, at each step of the way, from the simplest autocatalyzing
molecule to the DNA of today, each major failure of the information storage
mechanism to be processed correctly would be weeded out, so that only the
ones that did it successfully would still exist. Fortunately, there's more
than enough "slop" in the mechanism and in the "language" that small,
incremental steps in both the language and the language processor can go
nearly hand-in-hand.
Hmmm. Am I missing something? Where is the design argument? None is given
or even really implied. I suppose that one is hinted at, but no actual
argumentation is given, and the one that seems to be suggested by the
passages you quoted is pretty thin.
This archive was generated by hypermail 2b29 : Mon Dec 18 2000 - 23:24:30 EST