|
Author
|
Topic: The Birth of MSR and the End of the Theory of Evolution
|
Bruce Fast
Member
Member # 924
|
posted 15. February 2006 15:31
Hmmm, mouses don't know how to make human femers! Mouses only know how to make mouse femers. For a mouse to make a human femer the "information" necessary to get from what a mouse knows (how to make a mouse femer) and what a human knows (how to make a human femer) must be put into the mouse. Now, when scientists take a mouse, and add human DNA (no epigenetic machinery), they can bribe the mouse to grow a human ear. Therefore whatever a mouse knows (how to grow a mouse ear) + whatever information is in the human DNA which was spliced into the mouse is enough information for the mouse to grow a human ear.
Let's get this into terms we both understand (based upon your profile which I just read.) Let's say that you have a black box. It does a pretty thing. Let's call your black box "human." Let's call your pretty thing, oh, word processing. Let's say I have a black box. It doesn't do the pretty thing that yours does. It does something similar however. Let's call my black box "mouse." Let's call my something similar "line editer." Now, lets say that you give me some information. Let's call that CD. If, when you put your information into my black box I end up able to do your word processing, then the only relevant difference between my black box and yours, as far as doing word processing goes, is in the information that is on the CD.
Human DNA, and the DNA alone, contains whatever difference there is between a human femer and a mouse femer. Dispite the existance of epigenetic machinery, epigenetic machinery doesn't factor in.
IP: Logged
|
|
Atom
Member
Member # 1840
|
posted 15. February 2006 15:40
You again miss the point. You assume "Mouse can only produce mouse femur." Is that a true statement? Based on the PEH and other front-loaded theories, a mouse may contain the code for tons of features not expressed in mice. They would then turn on or off these additional pieces of code as necessary.
So again, just because you only gave the mouse a small "switch" piece of data, doesn't mean it didn't already contain the code to interpet and execute this code to produce a human ear.
Once you see this point, your mistake will become apparent. Thanks for the sarcasm with the simple breakdown.
Atom
[ 15. February 2006, 15:44: Message edited by: Atom ]
IP: Logged
|
|
Atom
Member
Member # 1840
|
posted 15. February 2006 15:55
And again, even with your computer example, you are assuming all the code for word processing is contained on the CD.
You are incorrect.
The CD contains machine instructions, which an assembler would then need to execute. So even assuming the CD did not need an OS to run, and was written directly in binary, we could not say the bytes of information stored on the disc contained all the code necessary for the word processing function. You also need the machinery, and so would need to include the information for that in your count of information necessary to do word processing.
In the same way, I predict the genomic machinery has several built-in routines, which an organism can take advantage of. It can say "Make arc-7889 with a 30 degree twist" (a 36 byte string), and produce a twisting arc structure, based on a pattern it already contained. It only needed to be told which pattern to use.
But it would be incorrect to then say "The twisting arc we see only takes 36 bytes of data to specify its form." Yeah, 36 bytes of input plus all the gigabytes of data used to store the pattern, interpret the string, select the pattern based on that string, and execute the pattern.
[ 15. February 2006, 15:57: Message edited by: Atom ]
IP: Logged
|
|
Atom
Member
Member # 1840
|
posted 15. February 2006 16:04
Looking at the "human" ear on the mouse, you can see it lacks the internal shape complexity of a true human ear. Thus, even with the DNA given to it plus all the code it already contained for such structures (my epigenetic and/or latent genetic information), it was not enough to fully specify for a true ear shape. Notice the aperiodic changes of shape in a human ear, and contrast this with the simple "cup" structure on the mouse's back.
[ 15. February 2006, 16:08: Message edited by: Atom ]
IP: Logged
|
|
Bruce Fast
Member
Member # 924
|
posted 15. February 2006 16:34
quote:
Based on the PEH and other front-loaded theories, a mouse may contain the code for tons of features not expressed in mice.
Ho,Ho! This is hypothesis, conjecture, not scientifically accepted theory! If you can proove that mice have the information in them to create human femers, you will upset the evolutionary apple cart in a big way. However, IT AIN'T BEEN PROVEN!! NOT EVEN CLOSE! Prove it before you use it as a premise!
quote:
We could not say the bytes of information stored on the disc contained all the code necessary for the word processing function.
quote:
even with your computer example, you are assuming all the code for word processing is contained on the CD.
Niet!!! NO!!!
The information which differentiates my machine with a line editor, and your machine with a word processor is held on the CD. The CD is not a word processor!! It is only the information WHICH ALLOW MY MACHINE TO BECOME A WORD PROCESSOR!!!
Atom, you are trying to debate without first thinking!!! Slow down your chatter, engage brain, then output words!
If this conversation does not increase in intelligence by about ten fold, I will not continue to respond to it.
IP: Logged
|
|
Atom
Member
Member # 1840
|
posted 15. February 2006 17:53
Don't get so upset my man.
You're still mistaken in assuming that "that which differentiates two machines" equals "that which completely specifies one machine or the other". You're saying "It only takes X amounts of (DNA) bits to code for a structure", and I'm saying "You're not including all the bits of information needed to make sense of that DNA fragment, which we also need to take into account." I'm sure everyone else can see this point by now. If I'm wrong, someone else please chime in and show me what I'm missing.
As for the PEH type information, I'm simply saying that this is what I predict we'll find. I expect there to be front-loaded information of some form of another. This isn't mere speculation, either. The fact that organisms have been shown to code for many varieties (see FDOCC's work on Variation vs. Speciation), while keeping most of that information latent in individual forms, is positive evidence for my prediction. Standard estimates based on information theory also cause us to infer epigentic and/or latent genetic information. (The points I brought up earlier concerning input vs. output information.) In addition to all this, there are studies which are beginning to show genetic "tool kits" which organisms can use much in the fashion I have described. Once again, if I'm wrong, someone please correct me.
So my points still stand. I still say no to Darwinian Just-So stories and I am still waiting for you to grasp the difference between trigger information (that works using built-in information) and a completely sufficient specification. I'm not saying I've proven that this built-in information exists; I'm just saying I am on better empirical grounds for assuming its existence than you are in denying it. And I add that you have yet to prove your point, that the amount of information necessary to encode for our structure is remarkably small. You claimed it is, I showed why you're probably wrong, and you got upset.
Anyone else following this and like to chime in one way or another?
[ 15. February 2006, 17:57: Message edited by: Atom ]
IP: Logged
|
|
KBC1963
Member
Member # 1868
|
posted 15. February 2006 18:43
Gentlemen,
I know that this discussion is getting a bit heated and I hope we can continue with as much civility as possible because from my point of view it is imperitive for my MSR concept to be able to weather any storm of evolutionary assertion by simple logic and by attempting to find flaws in my system Bruce is forcing me to think about some things that may not have been considered as I was assembling this concept.
Atom I love your tenacity for running with this thing and I am indebted to your input for allowing me to be able to put it into words and your arguement does have merit. I will attempt at this time to pull the discussion back into line with the original concept with the thought in mind that I need to show by use of simple logic how MSR observation can be used as a tool to define observed phenomena in the world around us
with the intent to better theorize about the unknowns that currently plague both sides in this controversy. Give me a few moments to think about how I can get us back in line.
IP: Logged
|
|
Atom
Member
Member # 1840
|
posted 15. February 2006 18:55
Lol, sorry KBC for hijacking your thread. I think the main point that Bruce was trying to make (which I agree with as well) is that you do not have to code for every point in a compressible shape. You only need to specify a framework, a set of rules and input data.
Where we disagree is that Bruce claims the femur and ear represent compressible information. He claims that it doesn't take all that much information to code for structures, since it doesn't take that much DNA. But he conflates input DNA with total information, and ignores possible (I'd say probable) built-in information and epigenetic information. That all has to be included in your estimate of information needed to specify a structure.
But we can move it back to MSR specifically. I have no problem with that.
IP: Logged
|
|
Bruce Fast
Member
Member # 924
|
posted 15. February 2006 19:08
quote:
From my point of view it is imperitive for my MSR concept to be able to weather any storm of evolutionary assertion by simple logic and by attempting to find flaws in my system.
That truly is the point of this forum isn't it?
BTW, I submitted my own hypothesis on this forum a while back:
"Pentadactilism again challenges the modern synthesis" I would love to have experience half as much dialog on that topic as this topic has received. Please feel free to check it out, and turn it into an active thread as well.
IP: Logged
|
|
KBC1963
Member
Member # 1868
|
posted 15. February 2006 19:38
Atom,
you are not hijackin, as intelligent beings we often stray from paths as we consider points of interest along the journey.
I see your point about Bruces arguement which from what I have read is saying to me that by estimating genome size alone that my model of how forms must mechanically arise is flawed because the amount of data my model suggests should be present to allow form to occur is much larger than the box it comes from.
Am I correct Bruce?
When you consider it in only that way then yes my model doesn't fit the observation, However this is where a hidden assumption was made. You assumed that the "raw data" of the femurs form must always require that it remain as is without any change from point A [the DNA] to point B [the femur]. Let me at this time settle this using only simple logic and then we can move on to other possible errors in my MSR concept.
I will ask simply to you Bruce, "If you put your mind to it how small could you compress the "raw data" that was was measured at 10mb?"
I think at this point you see where I am going so let me point out something I have observed in the world of programming. We originally had no compression of data when the computer first got going but as it became more complex intelligent designers such as yourself found methods to make things smaller but still fully functional so there came to exist file compression technology and over time they just keep gettin better at reducing those file sizes and its to a point now that I can work out of a zip file without unzipping it, "real neat technology there".
Since I have observed that information can be coded in such a way that what comes out is greater than what you see on the outside I can infer that the DNA may posess the same functionality and that by some compression method [black box] that may in fact be beyond our current understanding it is able to contain the pattern that allows for femur formation.
My original point about mechanisms exhibiting actions that require constraint was simply to point out that the control system that we know must exist but as of yet have not been able to define is actually applying constraint at a cellular level. This point should be a given.
Now by what method this system applies these controls is debatable.
My logical assumption is that the structure that makes up any mechanical form must be built by basic components and in the case of the femur it is formed from bone cells and unless you constrain them at that level you would get puddle-O-cells. Do we agree on this point gentlemen?
I should be able to easily infer that at least 1 parameter exists for each cell in that form to allow for repeatability.
This is what MSR's allow us to realize as we consider mechanical forms. It allows for a better concept of what it takes to allow continuous repetition.
In this particular case Bruce I can't tell you how the constraint of each cell is managed but you must agree that our observation of form constraint certainly implys that the control exists to this level.
[ 15. February 2006, 20:08: Message edited by: KBC1963 ]
IP: Logged
|
|
Atom
Member
Member # 1840
|
posted 15. February 2006 20:31
After re-reading Bruce's original post, I think he makes a point. When estimating the MSR we may have to consider only the constraints between independent parts of the system. If two parts arose from the same process, or if one part was produced by a process based on the second part, then a single change could update both subsystems. For example, if we were to take molds of our teeth for mouthguards, we could be tempted to say constraints exist between the size and shape of the teeth and the inverse size and shapes of the mold. Since the shape of the mold is dependent on the shape of the teeth, this 1:1 constraint of shape should not be counted as an MSR feature. But if we instead had to make the mold by hand, without seeing the teeth in question, and then had to match shape, the 1:1 ratio would indeed be MSR. So I think the independence of parts or processes that produce the parts may be necessary to include a given constraint in our MSR estimation.
Now as you point out KBC, these "process" constraints also require code, to keep one part in sync with another. I need to think more about this to see how it affects the concept.
Thanks Bruce for the insight.
Atom
IP: Logged
|
|
KBC1963
Member
Member # 1868
|
posted 15. February 2006 20:46
Bruce,
As I mentioned in the last post we should be able to make the logical assumption of a cellular based control system based on observation of persistance of form.
Based on the logical assumption of such a systems existence what would you think would be the minimum computer code that could allow that type of physical constraint to occur?
My initial guess would be that it must be at the most comparable to the progam that runs our 3d cnc mill which is about 60mb at its basic level and since this programs technology is ever evolving for us at this time I might even say that maybe we could shrink it down to say 1mb or so. Now even at that level how much code in terms of bits is needed to implement 3d constuction?
To me even 1mb of code that shows continuity of function as well as continuity of existence is infinitly beyond random chance to ever form even to infinite time. I assert that random minor changes to such a system can only upset its balance and never add to it functionality.
The vision of rolling the infini-dice as I call my dice of infinite values to randomly change values inside the 1mb file never equates to function and balance.
I envision further a scene where someone calls up microsoft and "says your program just mutated and now it has this new additional function...." and I wonder how we could ever take for granted that random occurances could form a balanced system that would require accurate values needed with only the infini-dice as the source of the values.
IP: Logged
|
|
KBC1963
Member
Member # 1868
|
posted 15. February 2006 21:28
"When estimating the MSR we may have to consider only the constraints between independent parts of the system"
You of course have a logical point Atom, and when it comes to questions of bilateral symmetry where the forms are near mirror images of each other then we may restrain our application of an MSR in that case as the execution of the second forms construction may arise from the first forms file being read in a different way, although I would like to know exactly what it takes to implement mirror imaging.
I fully agree that looking for msr's that exist between two physically different components lends more credence to existence of the msr control across form boundaries.
The code or control system that creates an msr that that we can observe is truly a black box to us at this point but because of the mechanical observation of things like the joint between the femur and the pelvis we can make determinations of what must be happening at the cellular level of each component since all forms are simply the organizations of their smallest component which in this case is bone cells, right?
I believe it is logical to say that a specific part of the femurs form encoding is being controlled in such a way as to remain within a constant value of another set of values contained within the encoding of the pelvis.
Soooo... Leading from those logical assumptions based on msr observation I can conclude that billions of cells within each forms construction are being controlled and since I can infer that the control is at a cellular level then I believe it is logical to say that at least one parameter is needed per cell to allow for conformation of specified shapes.
Would you not agree that this one msr encompasses multitudes of parameters to execute?
With that in mind what would a small mutational change within just this one msr comprised of that many variables ever acomplish? shall we roll the infini-dice and see what happens? hehehe.
I feel that with closer study our observation of human mechanics by MSR will lead to a better understanding of how things must work in order to acheive an end and even though we may not see the specific tree within this forest, we can at least define the forests properties in detail.
IP: Logged
|
|
Bruce Fast
Member
Member # 924
|
posted 16. February 2006 01:07
quote:
I will ask simply to you Bruce, "If you put your mind to it how small could you compress the "raw data" that was was measured at 10mb?"
Wow, there's a hard question. Again I am bumping hard into the limitation that I know software, and I know innovation, but I don't know biology even to the bachelor's level.
However, let me take a whack at it.
I am seeing three kinds of compression:
1 - Code re-use. When we write software, we frequently call the same routine in a thousand different places. Biology clearly does something similar. There seem, for instance, to only be a few different kinds of muscles. If there is a muscle in any part of the body, quite obviously the same code that codes for that class of muscle is implemented.
2 - Let me call this one "algorithmic patterns." This seems to be a hobby of biological systems. Consider, for instance, the ant. Ants operate, as I understand it, in two modes: hunter, and gatherer. In hunter mode, the dear things meander about almost randomly hoping to bump into something yummy. When they find something, they turn on a scent trail and head back to their nest (they seem to have a good return navigation system.) In gatherer mode, they follow the trail to find the booty. It's an algorithm that takes very little code, but works incredibly well. Let me present another less appealing example. I recently had a cold. At some point I gave a large cough, and coughed up the infection (you know what I mean.) In the thing I coughed up, there was a whole network of my blood vessles. It would appear that my blood vessles have a "grow into any opportunity" attitude. This approach to the code of life ultimately requires very little code, just a good search algorithm does it all.
3 - Using DNA for multiple proteins. When I encountered this concept I was like "this can't have evolved." (Let me say that virtually the limit of my knowledge on this is found in "Evolution, a theory in crisis.") It seems that DNA codes in some really wierd ways. A section of DNA will code out to a protein, then the rna will split up into pieces and code up to some other proteins. Then, for some reason the sections of DNA used to code proteins will overlap -- for example protein 1 will code from offset 10256 to 10452 and the next will code from 10384 to 10825. I also understand that DNA will code different proteins based upon actions by different enzymes. (I bumped into this recently when studying histone-4. Histone-4 is virtually unchanged between plants and animals, so has not mutated in the last 650 million years.)
Now, let me expound on this third point of data compression for a bit. Years ago I linked two computers together with three electrical wires through the serial port. The connection was really weak, and if a computer had to take a breath to dump something to the hard drive, it would forget to read the data off the port, so data errors were common as dirt. To get reliable data, I did two things: I created packets, and allowed for the re-sending of packets that didn't checksum, and I compressed the data with zip prior to sending it. Either the file got to the other side 100% correctly or it blew up in flames of glory when I tried to unzip it. The moral -- really simple -- it is really really hard to mutate compressed data. The type of compression discussed in point 3 above has this sort of mutational immunity.
Now, how much data compression is going on in human DNA? Let's consider only type 3 'true' data compression (as opposed to algorithmic efficiency.) My understanding (rehearsing that I am not vaguely an expert) is that this sort of compression is common as dirt in the more complex organisms. My understanding is that "most" active DNA codes for multiple things somehow or other.
This stands as one of Denton's challenges to evolution, a challenge that I personally find to be very strong. Do some googling on histone 4, and on molecular clocks. You will find that this clock don't tick, and that applying the molecular clock hypothesis would suggest that the dear thing was around in almost it's current form back a few seconds prior to the big bang.
IP: Logged
|
|
KBC1963
Member
Member # 1868
|
posted 16. February 2006 12:33
Bruce,
You said: "Again I am bumping hard into the limitation that I know software"
I think my train of thought dealing with compression of information may be significant.
when I was asking how small 10mb might possibly be compressed I was asking you how small
you could compress it with the best known ability of current technology just so we may get an idea
of what is possible. The fact that we don't know how DNA functions, allows us to invoke known
and understood compression technologies as a possible method by which information such as 3d
patterns may be compressionably encoded within the structure of the DNA. I think gentlemen,
that if we can logically deduce by method of MSR identification that we are getting tons
more information out of the DNA than is apparent in its observable size that we can infer that
an as yet unknown form of compression is at work within the DNA's structure.
Do you guys suppose it is possible that science may be able to analyse DNA structure and
functions and be able to determine that the majority has been compressed in order to contain
all the information necessary to form an organism and continuously control its shape and function
for its entire life? and if this can be determined do we then have sufficient evidence to say that the engine of evolution could never have randomly acheived such an irreducibly complex
system based on known understandings of how information compression works, where it requires
a complex logorithm to compress it in the first place and then a complex unzip utility to
allow the compressed information to be accessed and read while still remaining compressed,
Just as I pointed out about being able to work with files without the need to decompress
them first.
I think that we may have hit on a revelation that I never considered could be possibly justified with MSR observations. Our interaction on the MSR possibilities may have just payed off big time regardless of the future uses of MSR.
Please let me hear each of your thoughts on this little gem of a concept and lets see if we can develop something.
Further thought: How would random mutation affect compressed information? are some things impervious to change by mutation such as the compressor and the reader? can they be?
Is it possible that there are two forms of mutation, one that can act at the point where
the mating of male and female halves of DNA is determining the specifics of the new
creature, and a different type such as used in the fly experiments where they used radiation
to spur mutation that appeared to be bad at every instance of its observation? I am supposing
that the radiation type mutation may actually be altering the overall compressed file structure
causing bad results by too massive a change occurring.
IP: Logged
|
|
|
Printer-friendly view of this topic