|
Author
|
Topic: Biological Information
|
Evan
Member
Member # 164
|
posted 02. December 2002 13:23
I would love to see examples. In fact it would be particularly instructive to see an example where both types of analysis were done (design and "peer-review") in order to see the difference between the two.
Added later: let me make it clear that what would be good would not be more "approaches to quantifying biological information" - which would just be more theory, but rather an actual example of something biological, with some tentative numbers and the means by which those numbers are, or even might be, obtained.
[ 02. December 2002, 18:11: Message edited by: Evan ]
IP: Logged
|
|
Daniel Edington
Member
Member # 421
|
posted 02. December 2002 21:08
quote:
I offer the outline of a method of directly measuring the improbability of generating a protein. This would be a direct measure of the improbability of occurrence. Clearly a direct measure of improbability is far superior to a crude estimate. I answered your question.
I did not ask for a vague outline, as such will not result in any actual experimentation. I want one of two things: 1) Actual numbers, resulting from the "direct measure of the improbability of occurrence." of something like human cytochrome c. 0r 2) I want a detailed experimental protocol for measuring the "improbability of occurrence" of something like human cytochrome c.
all of this with references to the primary scientific literature would be grand.
I really don't think I'm asking that much.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 04. December 2002 11:41
Before addressing the issue raised by Evan, a few comments on mathematics, mathematical expertise, and the mathematical nature of the proposals offered here. To begin, the mathematical concepts, definitions, and techniques being discussed here produce mathematical conclusions which directly contradict a number of ideas that are widely accepted in both evolutionary biology and ID.
MATHEMATICAL CONTRADICTIONS
Mathematics does not permit unexplained contradictions. To somewhat oversimplify, an assertion can not be both ‘mathematically impossible’ and ‘mathematically valid’. It can in fact be argued that one of the major benefits of using mathematics in scientific analysis is to identify contradictions and logical inconsistencies that would not otherwise be recognized.
The mathematical concepts being introduced here are intended as useful tools in the scientific analysis of biological information processing. The mathematical concepts and techniques being introduced are also, however, designed to identify and evaluate contradictions and logical inconsistencies in currently accepted scientific ideas.
The mathematical concepts introduced here appear to support conclusions such as:
1. It is mathematically impossible for most evolutionary changes to have arisen from RM&NS processes.
2. It is mathematically impossible to logically infer the existence or actions of an external designer from the observation and analysis of biological design.
3. It is mathematically impossible for biological information to be consistent with or conform to the laws of thermo-dynamics.
4. It is mathematically impossible for most evolutionary change to have arisen from Darwinian processes.
5. It can be shown mathematically that teleological causation is a form of deterministic permanent and universal causation.
These conclusions would appear to directly contradict a number of widely held views and beliefs. The list of mathematical conclusions that appear to directly contradict widely held scientific opinions is long. Design science mathematics makes it relatively easy to identify apparent mathematical contradictions in both evolutionary biology and in conventional ID. Confirming or disproving an apparent inconsistencies is, of course, much more difficult. The purpose of this post is to consider the ‘how’ and ‘who’ of evaluating mathematical inconsistencies.
HOW ARE MATHEMATICAL INCONSISTENCIES EVALUATED
The discussion here is not concerned with the specific techniques, but with the general principles of mathematical evaluation. Mathematical evaluation is based on objective, verifiable, explicit analysis on which mathematicians can reach consensus. Specifically, mathematical evaluations are not based on subjective, undocumented, personal opinions of authority figures or individuals claiming to be authority figures.
The issue of validity established by ‘subjective peer review acceptance’ may be controversial in scientific analysis but it is not or should not be controversial in mathematics. The validity of a mathematical assertion should not and can not be established by a general agreement among authorities based on subjective, non-explicit, non-reviewable arguments.
WHO EVALUATES MATHEMATICAL INCONSISTENCIES
There are two features of ‘who’ evaluates mathematical inconsistencies which are worth noting here. First, mathematical inconsistencies are evaluated by ‘mathematical experts’, and not everyone qualifies as an expert. Second, mathematical evaluation is a process performed by a group of experts, not a process performed by an individual.
EXPERTISE
From our earliest days in school, we know that there are huge differences in mathematical ability. We also know, that even for individuals with natural mathematical ability, learning to perform specific types of mathematical analysis requires training and experience. Finally, we know that even with ability and knowledge, actually performing specific types of mathematical analysis can take a great deal of time and effort.
When new mathematical concepts are introduced, it takes time, effort and resources to develop the expertise needed to perform mathematical evaluations. Design science mathematics is not a dramatic departures from existing mathematics, but it involves concepts and techniques which will be unfamiliar to many mathematicians.
There is an unfortunate tendency in academia to assume that everyone with a credential who has read a few papers and can spout appropriate terminology is a mathematical expert. There is am equally unfortunate tendency to assume that people without certain credentials and who don’t use certain terminology are not experts. Mathematical expertise is based on the ability to actually perform and understand mathematical analysis.
GROUP VERSUS INDIVIDUAL
Mathematical evaluation is a group or social process by which a group of mathematical experts reach agreement or consensus on an issue. One individual solving a problem and believing the solution is valid is not mathematical evaluation. Mathematical evaluation requires that a group of experts confirm the findings proposed by an individual. In most complex applications, many different individuals will contribute to the formulation and refinement of a proposal, as well as confirming a proposal.
There is a common misconception that mathematical evaluation is primarily the work of a single individual. From this perspective, a single individual makes a discovery, performs analysis and a group of less important experts looks at the discovery or analysis and gives it blessing.
While one individual may play a significant role in initiating a particular mathematical evaluation, the process of evaluation is a group process. Any proposal to perform a type of analysis will be based in large measure on the prior analysis performed by many different individuals. After a proposal has been presented for evaluations, particularly if the proposals is a complex one involving many features, many individual experts will be involved in evaluating and refining the many features.
SUMMARY
Design science mathematics was developed primarily as a tool to assist in the scientific analysis of biological systems. A secondary feature of design science mathematics is that it appears to support mathematical conclusions which directly contradict many widely accepts scientific ideas. Properly evaluated by mathematical experts, these mathematical conclusions provide a useful basis for evaluating the soundness (or lack of soundness) and usefulness of design science. Similarly, these conclusions provide a useful basis for evaluating flaws (or lack of flaws) in many existing forms of scientific analysis.
IMO, both Darwinian theories of evolution and the peer review standards science which support it are seriously flawed. One of the key benefits, again IMO, of the proposed definitions of biological information and biological information processing is that it reduces a number of key issues to mathematical questions. Since there is greater general agreement on the standards used to evaluate mathematical questions, it should be easier to resolve the differences between evolutionary biology and design science.
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 04. December 2002 12:21
Warren writes,
quote:
The mathematical concepts, definitions, and techniques being discussed here produce mathematical conclusions which directly contradict a number of ideas that are widely accepted in both evolutionary biology and ID.
The problem here is that so far no mathematical conclusions have been reached, because no actual mathematics has been done.
Yes, we have some theory.
Do we have any actual numbers? No, we don’t. Until we see the theory applied to some actual situation, with real numbers, we have no idea whether the conclusions (of which there are none so far) contradict anything at all.
We need the theory to be applied to something before we can reach any conclusions.
So, now that Warren has made these additional preliminary remarks, I will look forward to examples.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 05. December 2002 11:11
Quote: The problem here is that so far no mathematical conclusions have been reached, because no actual mathematics has been done.
One of the benefits of a discussion forum is feedback. Based on feedback you can judge both your audience and how clearly you have communicated your message to your audience. Based on the above comment, it is clear that I have failed to communicate the essential features of the design science approach to biological information. A second benefit of a discussion forum is that you can go back and try again.
As I attempted to explain in terms of mathematics, the design science approach involves a new definition or conceptualization of biological information and the processes operating on biological information. Expressed in less mathematical terminology, the design science approach involves a new way of looking at or visualizing biological information processing. More accurately, design science involves a refinement in the standard visualization of biological information processing. Possibly, it will be easier to understand the design science approach based on visualization rather than mathematical conceptualizations.
THE STANDARD VISUALIZATIO OF INFORMATION PROCESSING
The standard or basic visualization used to describe information processing involves three components- input, processing, and output. Essentially all efforts to analyze and model information processing are based on some variation of this three part model.
Darwinian evolution is based on this visualization. The Darwinian input is an organism, the processing is selective reproduction based on variation and selection, and the output is the new ‘evolved’ organism. The conventional ‘genetic view’ of biological information processing is similar with input equal to DNA, processing equal to some type of mutation and selection process, and output equal to modified DNA.
The thermodynamics visualization of biological information processing is based on a similar simplified input-output visualization where biological information has the logical structure of energy, processing is some standard physical force or forces and output is some altered form of the input energy.
The stimulus-response approach to the study of behavior can also be characterized as an input-output visualization.
THE GOAL OF SCIENTIFIC ANALYSIS
The input-output visualization fits nicely with idea scientific theories expressed in terms of mathematical algorithms. The goal of science, viewed from this perspective, is to 1)find a method of identifying and quantifying input, 2)find mathematical method of identifying and quantifying output, 3)measure or observe quantified values of input and the corresponding output, and 4)finally identify mathematical algorithm F such that the F(I)=O fits the observed relationships between input and output.
The Darwinian, the genetic, the thermodynamic, and the stimulus-response visualizations of biological information processing all succeed, at least in part, in developing methods of identifying and quantifying input and output. Similarly, these visualizations all make it possible, at least to some extent, to observe and measure pairs of input values and associated output values.
THE FAILURE OF INPUT-OUTPUT VISUALIZATIONS
As a basis for scientific analysis, all the standard visualizations have been failures. No one has succeeded in processing algorithms capable of simulating and/or predicting the output that will be generated by input. It is important to remember that despite the best efforts of scientists, existing input-output visualizations have, to date, failed to produce hard science, predictive models or theories of biological information processing.
The debate over Darwinian and genetic theories of evolution is not over whether there is a predictive model or theory of evolutionary change. No one seriously claims that anyone has discovered a predictive algorithm F which is capable of scientifically predicting either genetic or phenotype changes. The debate with respect to evolutionary theory is not over hard science models or theories, but whether the pseudo mathematical or metaphysical RM&NS type theories should qualify as ‘scientific theories’.
THE DESIGN SCIENCE DIFFERENCE
The goal of design science, like the original goal or all sciences using the input-output visualization is to actually find a processing algorithm(or set of processing algorithms) F that can predict the values of output from the values of input. There are two basic reasons why design science is considered a different type of science based on different standards. First, design science standards are intended to eliminate the rationalizations that are used to justify treating pseudo mathematical models like RM&NS as scientific theories. Second, design science standards are intended to recognize the concept of solution sets which are essential to formulating models and theories of complex causation.
REFINING THE INPUT-OUTPUT VISUALIZATION
The fundamental problem with standard input-output visualizations is that they are too simplistic. The standard input-output visualization allows for one type of input, one type of output, and a single processing algorithm. The first step in finding algorithms which can simulate biological information processing is to expand the input-output visualization to allow for multiple inputs, multiple outputs, and multiple types of processing. To account for this increased complexity, biological information processing is visualized as the processing occurring within a complex computer or logic machine.
From work with computers, most of us will be familiar with complex logic machines. Such machines can have many different types of input and many different types of output. Input and output would include ‘internal’ input and output coming from and going to internal storage files. Multiple processes performed by these biological logic machines would take the form of different subroutines within the overall programs controlling biological information processing.
TO BE CONTINUED
It is, or should be fairly easy to visualize biological units as complex logic machines performing complex biological information processing. Variations of this visualization have been widely used in analyzing the functioning of nervous systems and the functioning of the human mind.
It should be obvious, but apparently is not, that the key issue is using this visualization or this general approach, is not ‘showing calculations’. The key issues are 1)unraveling the structure of this complex logic machine, 2)finding methods of isolating and analyzing various sub-functions, and 3)finally finding predictive algorithms that can describe the functioning of the biological logic machines.
Enough for one day, I will expand on the ‘biological logic machine’ visualization tomorrow.
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 05. December 2002 11:36
Warren, please see my opening post in the new thread "General Concern about ID Research" - it contains my reply to your latest post here as well as more general comments which also apply to your work here.
[ 05. December 2002, 13:42: Message edited by: Evan ]
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 06. December 2002 11:33
My apologies to the moderator for starting the thread "Concerns about ID Research" when it just turned out to be a continuation of this thread.
I think it is clear that Warren is not able at this time to connect his theoretical ideas to any real-world observables, much less supply any rough estimates of numbers that would allow us to quantify the idea of biological information (the topic of this thread) or biological complexity.
My general concern that in fact no one knows how to do this is valid, I think, but not, perhaps, fit for discussion as a opening post. If anyone ese besides Warren has some ideas about how to quantify biological compexity (a critical step in actually applying the explanatory filter), I hope they will offer their “brainstorm” about this sometime.
IP: Logged
|
|
|
Printer-friendly view of this topic