A Modest Proposal (By a Somewhat Modest Engineer)
by JJS P.Eng."Front-loading is the investment of a significant amount of information at the initial stage of evolution (the first life forms) whereby this information shapes and constrains subsequent evolution through its dissipation. …
Front-loading, by definition, is about designing the future through the present. It is about imposing some kind of constraint on evolution, or more simply put, it is using evolution to carry out design objectives."
The Design Matrix, Chapter 7, p. 147
The above quote raises interesting questions from an engineering viewpoint. How did the front-loading designer "carry out design objectives" when designing the first life forms? What design philosophy or methodology was used? What potential obstacles had to be overcome? These are questions that logically follow from the above statement.
As an engineer, I find the teleological/non-teleological evolution debate fascinating. However, what role (if any) is there for engineers in the investigation? I believe that I have a modest proposal that outlines the engineer's role in the debate that taps their skills and knowledge. But first, we need to find out what engineers do and how do they go about doing it.
Wikipedia defines engineering as (emphasis mine):
"… the discipline and profession of applying technical and scientific knowledge and utilizing natural laws and physical resources in order to design and implement materials, structures, machines, devices, systems, and processes that safely realize a desired objective and meet specified criteria."
Not only does the engineer require a good grasp of the scientific knowledge applicable to his/her area of expertise, the engineer requires technical skills ("utilizing natural laws and physical resources") to apply that knowledge to produce tangible products used by real-world consumers ("[to] realize a desired objective and meet specified criteria").
The article goes on further to explain (in a generalised way) engineering methodology (emphasis mine):
"Engineers apply the sciences of physics and mathematics to find suitable solutions to problems or to make improvements to the status quo. More than ever, Engineers are now required to have knowledge of relevant sciences for their design projects, as a result, they keep on learning new material throughout their career. If multiple options exist, engineers weigh different design choices on their merits and choose the solution that best matches the requirements. The crucial and unique task of the engineer is to identify, understand, and interpret the constraints on a design in order to produce a successful result. It is usually not enough to build a technically successful product; it must also meet further requirements. Constraints may include available resources, physical, imaginative or technical limitations, flexibility for future modifications and additions, and other factors, such as requirements for cost, safety, marketability, productibility, and serviceability. By understanding the constraints, engineers derive specifications for the limits within which a viable object or system may be produced and operated."
Again, it is stressed that (skilled) engineers need "to have [comprehensive] knowledge of relevant sciences". Part of the design methodology is "to identify, understand, and interpret the constraints on a design in order to produce a successful result". Once the constraints are understood, engineers can set operating limits for use of their designed products. These limits are, in part, defined by the various natural laws and physical/chemical mechanisms involved.
Quick note: while these definitions may not fully capture the various intricacies of engineering* (short generalised summaries rarely do), they are sufficient for this post's purposes.
The engineering profession is made up of several distinct disciplines. Wikipedia provides a fairly comprehensive list of these disciplines, such as aerospace engineering, structural engineering, sparkies electrical engineering, and mechanical engineering. Absent from the list, but no less important, is biological engineering and its various subsets. Note that while each discipline has its own specialised niche, there are basic and overarching engineering principles that provide the foundation for each field of engineering practice.
Finally, I wish to provide a reference to reverse engineering (the reasoning for this will become clear soon):
"Reverse engineering … is the process of discovering the technological principles of a device, object or system through analysis of its structure, function and operation. It often involves taking something (e.g. a mechanical device, electronic component, or software program) apart and analyzing its workings in detail, used in maintenance or to try to make a new device or program that does the same thing without copying anything from the original." (emphasis mine)
To summarise, an engineer requires substantial knowledge of relevant sciences and the technical skills to apply that knowledge to the real world. The engineering profession has split into various specialised disciplines that require both specialised scientific knowledge relevant to their field and an overall knowledge of basic engineering principles and methodology. Reverse engineering can be part of the engineer's technical toolkit.
With this in mind, I propose that engineers from various fields can draw upon their scientific knowledge and technical skills to create a new engineering discipline: Evolutionary Biological Engineering (EBE). The following is a Wikipedia-like definition outlining what an EBE does and his/her potential role in the investigation:
"Evolutionary Biological Engineering (EBE) is a branch of engineering that explores how biological life could have been designed using mechanisms and principles identified by many decades of evolutionary biological research. Implied in this definition is the assumption of an intervening entity (i.e. designer). A common problem encountered with this assumption is the many variables that such a designer introduces to the investigation. Front-loaded evolution (FLE), as described by Mike Gene in The Design Matrix, is proposed as a way to simplify the amount of additional variables.
Evolutionary biological engineers require a comprehensive understanding of evolutionary principles and mechanisms, which are used - along with general engineering principles and design methodology - to explore potential design methods that could have been used by a (hypothetical) human-like designer and to investigate how certain design obstacles could have been overcome. EBE methodology and principles may draw upon those used by other engineering disciplines (eg. computer, electrical, mechanical, structural).
EBE also makes use of reverse engineering to learn more about the technological principles behind life's (potential) design."
To state the obvious, EBE is in the "infant" stage**, where the methodological and theoretical foundation must be laid. FLE provides such a foundation (or starting point).
All constructive input is welcome. In the meantime, this beaver has a rabbit to track down. 
*For those who are interested in more details about engineers, I find the writings of Henry Petroski are particularly illuminating, even for the layman. For the brave few who wish to look more closely at structural engineering, I highly recommend the book Structures: or Why Things Don't Fall Down by J.E. Gordon.
**When geotechnical engineering was in its infant stage approximately 50 to 80 years ago, it had a relatively sparse experimental database to draw upon compared to other, more mature, disciplines. This is not necessarily the case with EBE, since it can draw upon over a century worth of experimental and scientific findings.
January 13th, 2009 at 12:58 pm
It looks as if the answer you will get is that if an engineer designs an object that is self maintaining, able to supply its own energy needs and has reproductive and information storage capacity, then one will get a succession of progeny from the initial design.
Comment by Bradford — January 13, 2009 @ 12:58 pm
January 13th, 2009 at 1:04 pm
Good day Bradford.
I would say that is a rather broad answer. I'm trying to get a closer look at the details (design methodology, constraints, obstacles, etc.).
Comment by JJS P.Eng. — January 13, 2009 @ 1:04 pm
January 13th, 2009 at 1:07 pm
IOW, I'm trying to approach the issue with higher resolution.
Mike Gene: the gift that keeps on giving.
Comment by JJS P.Eng. — January 13, 2009 @ 1:07 pm
January 13th, 2009 at 1:25 pm
JJS:
Embryonic stage would be a more modest characterization. Like many creationist/ID proposals, FLE is likely to get aborted or be stillborn (modestly stipulating that this hasn't happened yet)
"Reverse-engineering" principles have been applied in evolutionary biology for a long time, both by engineers (e.g. John Maynard Smith) and non-engineers. Ironically, the results suggest that there was no engineer.
Comment by Raevmo — January 13, 2009 @ 1:25 pm
January 13th, 2009 at 1:27 pm
Have at. First question: what constraint caused the approach of designing the end results, and then, instead of just building it, creating this insanely complicated unfolding process? And how can we confirm such a constraint once an engineer has proposed one?
By the way, can you point to your prior research? I've seen a lot of engineering-like analysis of biological systems, so I'm somewhat surprised to hear you suggest that this is a new idea. In fact, if I were to guess, I'd guess that current theory — you know, the one that doesn't involve teleology — is in line with engineering every bit as much as it is in line with chemistry and physics, so I'm very interested in reading your research that shows this isn't the case.
Comment by don provan — January 13, 2009 @ 1:27 pm
January 13th, 2009 at 1:43 pm
Raevmo;
In true Rudyard Kipling style Raevmo gets to introduce an embyronic stage without having to account for it.
Engineers don't have the benefit of just so techniques. They must assume a starting point.
Comment by Bradford — January 13, 2009 @ 1:43 pm
January 13th, 2009 at 1:58 pm
Current theories entail concepts like minimal function. How is that atelic?
Comment by Bradford — January 13, 2009 @ 1:58 pm
January 13th, 2009 at 2:20 pm
Bradford:
I was being generous — otherwise I would have said cloned from a creationist starting point.
Comment by Raevmo — January 13, 2009 @ 2:20 pm
January 13th, 2009 at 2:21 pm
Bradford:
What theories and how is that telic?
Comment by Raevmo — January 13, 2009 @ 2:21 pm
January 13th, 2009 at 2:25 pm
Good day Raevmo:
Can you use the probabilistic design method to inform us of the probability of such "abortion" occuring?
I never said reverse engineering has never been used in evolutionary biology, just that it would be part of the toolkit for the EBE.
Correct me if I'm wrong (since I don't have "the results" in front of me, but the question of OOL has not been sufficiently answered yet, and all that "the results" suggest is that there are natural mechanisms that could explain the development of certain biological organisms. Both FLE and EBE are focused at the initial conditions for OOL and the potential design methodologies employed by the front-loading designer.
Comment by JJS P.Eng. — January 13, 2009 @ 2:25 pm
January 13th, 2009 at 2:43 pm
Good day don.
Possible constraints include material properties (proteins, DNA, RNA, etc.) and the environment the first cells were placed in.
I find it interesting that you are more inclined to critique the design process rather than find out more about the design methodology.
As opposing to actual engineering analysis of biological systems?
The problem with that line of thinking is that physics and chemistry are not as dependent on engineering terms as biology is (Chapter 3, The Design Matrix). What Mike, myself, and others want to know is if this analogy is legitimate and if it is (Mike makes a good argument for it in Chapter 3), can engineering shed light on OOL?
"
I quoted Mike because it seems like you and every other critic expects the teleological argument to be at the same level of sophistication as the non-teleological argument. IMO, the teleological argument has been neglected for over 100 years. Does it not make sense to provide constructive criticism that reflects the actual level of sophistication, or are you afraid the teleologists will actually catch up, so you feel the need to bludgeon it.
By the way, I am planning on more posts that investigates a potential design methodology and a potential obstacle that every engineer faces.
Comment by JJS P.Eng. — January 13, 2009 @ 2:43 pm
January 13th, 2009 at 2:51 pm
And a good evening to you, JJS:
Is that how they call guessing in engineering? Well, I guess a spontaneous abortion will occur as soon as it is realized that no distinguishing predictions have been forthcoming for a while. I guess the probability of that happening is >95%.
I never said you said that. I thought it was worth pointing out for those readers who were not aware of this.
You are right, the question of OOL is still quite open. However, there are ideas like the RNA world that have some empirical support (such as the discovery of self-replicating RNA molecules). I'm not aware of any FLE models that make unique predictions that have been confirmed, but I'd be corrected gladly if I'm wrong. I also had the impression that FLE was not just focused on OOL, but on the whole of evolutionary history. At least on this forum, I've seen many claims of FL during later stages of evolutionary history.
Comment by Raevmo — January 13, 2009 @ 2:51 pm
January 13th, 2009 at 3:06 pm
How is it telic?
Comment by don provan — January 13, 2009 @ 3:06 pm
January 13th, 2009 at 3:34 pm
Have I discourage you? Have I criticized your suggestion? No. Quite the contrary. I asked for some of your ideas and for more information about how you know what you're proposing hasn't already been tried.
But don't expect me to jump on the bandwagon. IMO, the reason teleology has been ignored for 100 years is that in the thousands of years before that in which it was the primary argument, it tended to produce results that were of little use and even worked against gaining useful knowledge.
That's what I get for being polite. You're suggesting that no engineering has been done, so I polited suggested that what I saw was merely engineering-like to avoid rudely suggesting they were counter examples. Meanwhile, you don't explain why you are suggesting different than those examples regardless of whether they're engineering or merely engineering-like. Why is that?
I think you misread my question. I was asking why you think engineers have overlooked evolution until now while physicists and chemists have gone over it in detail. I would assume engineers would already be quite involved for exactly the reasons you say. Can you explain why they haven't?
I must me missing something. Is this really more useful than something entirely vacuous, like "there might be constraints of some kind"? What about "material properties" might make the impossibly complex approach of front loading prefered over simple construction? Just to think of one thing, engineers invariable have to test and debug their constructs, but you are suggesting an approach that makes testing and debugging impossible. That doesn't sound like engineering at all, so you should have truly excellent reasons for saying engineering is responsible for an approach that leaves out two critical steps in engineering as we know it.
What have I critiqued? I was entirely supportive! I've accepted your proposal and am asking you to explain one single fundamental aspect of your suggested engineering approach. Should it concern me that failing to answer it makes you defensive?
Comment by don provan — January 13, 2009 @ 3:34 pm
January 13th, 2009 at 3:50 pm
Well, as defined by the quote from The Design Matrix in the OP, front loading starts with the origin of life, making OOL a given because it is itself an example of a front loading event.
Comment by don provan — January 13, 2009 @ 3:50 pm
January 13th, 2009 at 4:13 pm
don:
Sure, but just front loading the OOL (by seeding earth with self-replicating molecules, say) is not enough to guarantee the evolution of humans — evolution is too stochastic for that. Even many IDists seem to realize that. Therefore repeated injection of "information" is needed to secure the evolution of humans, who are thus made in the image of the Designer. I seem to recall that Mike Gene has often argued that single-celled organisms were front loaded with genes that more or less guaranteed the evolution of multicellularity — an obvious prerequisite for human evolution.
Comment by Raevmo — January 13, 2009 @ 4:13 pm
January 13th, 2009 at 4:44 pm
Perhaps an front loading advocate could help us with this. I understand front loading as being a way to avoid leaving anything to chance.
Have I missed this? My impression is that the whole point of front loading it to avoid having to inject information during the process.
By the way, "in the image of the Designer" not only loses you points by allowing the "Not Religion" chant (as I'm sure you realize), it also loses points because it gives ID too much credit: teleology suggests only an unidentified purpose, never, ever a purpose as specific as producing a being with some identifiable target appearance.
Comment by don provan — January 13, 2009 @ 4:44 pm
January 13th, 2009 at 7:13 pm
I may be wrong, but I have a feeling Mike is quite modest (well relatively) in his claim. He is not suggesting that front-loading was engineered to produce humans, but only that it designed to give organisms the basic building blocks (including multicellulity, as you say).
JJS, this would be very important to your research. Are you proposing a designer who front-loaded a basic toolkit to aid evolution, or a designer who had the human form in mind?
If the latter, would you not expect to see far more human-like features right across biology? If the human brain was coded into first life, why is it not seen anywhere in the insect world, for example? Can you estimate how long the DNA would have to be to code for the original single-celled organism, and humans, and include systems (I am thinking redudancy, perhaps you have other ideas) that would ensure the human DNA survives for 4 billion years?
With regards to why the supposed aliens would take this approach, a possible answer is that they identified hundreds of planets in the galaxy with low oxygen atmospheres. They engineered this first life to be able to survive that environment, and to evolve so that they would alter the environment to be oxygen rich, and then further evolve to produce "hgher" lifeforms. Once the system has been designed once, it can be used again and again on such planets all the way across the galaxy, perhaps seeded by simply firing rockets at the desired planet. No need to leave their own planet, or go faster than light speed.
Comment by The Pixie Again — January 13, 2009 @ 7:13 pm
January 13th, 2009 at 8:22 pm
JJS P.Eng.
Great way of putting it, but I have some questions for you.
* If this is the infant stage, could you provide what you'd imagine a more mature stage would look like?
* Where does artificial selection fall when considering such a scope? Is it part of EBE? In a different category altogether?
As an aside, UD's latest post deals with a study where 'pre-existing evolutionary potential' in the past is discussed and realized. I'm surprised at how I'm seeing Mike Gene's name pop up more and more in comments there, and among people I normally wouldn't expect to fully appreciate his book. (Hopefully his site returns soon, it was apparently hacked.)
Comment by nullasalus — January 13, 2009 @ 8:22 pm
January 14th, 2009 at 1:22 am
Interesting post, JJS.
I have not read "The Design Matrix."
Regarding the proposal, the creation of a new field Evolutionary Biological Engineering (EBE), the term seems to be a misnomer. Engineering fields are concerned with engineering. A study of engineering requires that examples exist to study, and an engineer.
In all other engineering fields, the engineers are human. So it seems that one would require the existence of biological engineers who are human, in order to study biological engineering (BE).
Likewise, human evolutionary biological engineers, who have actually engineered life that evolves, seem necessary to study EBE.
In any field of engineering, the first engineers are often hacks. Sometimes they are intelligent hacks, and sometimes they are seemingly inspired hacks. One such engineer, a clever and possibly inspired hack, Lee De Forest, invented an amplifier whose operation he was intellectually ill-equipped to rigorously explain.
Attempts at engineering often lead to catastrophic failures due to unintended consequences. The Tacoma Narrows Bridge, and the 35W Bridge in Minnesota come to mind. In both cases, and certainly in the later case, one could argue that engineering optimization for cost was a significant factor in the failure(s). In engineering, optimization with insufficient margin often leads to unintended consequences, but engineers often take pride at optimal, or near optimal, solutions, with the margin being pulled from a hat or treated unquestioningly as a constraint tantamount to a law of nature.
The point being, as humans endeavor in the field of biological engineering, there will be catastrophic failures. In the case of a bridge, the damage is contained. In the case of a human created reproducing life form released into the biosphere, the consequences could be devastating.
One point I am trying to make, in order to engineer a life form front loaded to adapt via evolution to future environments, it seems necessary that one be able to engineer a life form, period.
Are we there yet?
"Reverse engineering" is another term, and, again, I think it is sometimes a misnomer, or at least too imprecise. Modern history provides examples that in the absence of respect for intellectual property, it is possible for a group of clever persons to "reverse engineer" and then manufacture devices that were actually engineered, without having an ability to engineer themselves. I do not think this is engineering, any more than I think parrots are able to converse in English.
Granted, real engineers are able to reverse engineer, and re-engineer (as Edwin Armstrong was able to do with De Forest's amplifier), but the term "reverse engineer" makes no distinction between an Edwin Armstrong and a copycat.
I don't know of any instance where humans reverse engineered a designed object that was not designed by a human, either in the engineering sense or the copycat sense of "reverse engineered." (Well, okay, we've figured out how some primitive tools used by animals work.)
Some do believe that alien vehicles have been acquired and reverse engineered by secret government organizations, but I am not among them.
In conclusion, before we can study evolutionary biological engineering, it might be necessary as a species to first engineer life that can evolve, learn from the mistakes, and hope we don't induce a biological Armageddon.
Meanwhile, studying how biological systems work is probably not too much different from what you proposed, except the existence of an original engineer is not necessary.
In the end, evolanders will be unimpressed by replacing God with "intelligent designer" with "human-like designer" with "engineer".
Yet I think most would agree that studying how biological systems work is a worthwhile pursuit, even if some are reluctant to apply what we learn to engineer new life forms that evolve.
Comment by William Wallace — January 14, 2009 @ 1:22 am
January 14th, 2009 at 1:29 am
By the way, where is this server located, New Brunswick or Nova Scotia? It's only 11:22 p.m. where I am at.
Comment by William Wallace — January 14, 2009 @ 1:29 am
January 14th, 2009 at 12:11 pm
Good day nullasus.
I don't know if this answer will totally satisfy your question, but I see one such "mature stage" being the development of potential design sketches that demonstrate the evolutionary path from A to B. For instance, take a single celled organism as A. The design sketch could show, based on environmental and evolutionary mechanistic influences, the designed path to achieve B. This "sketch" could take various forms (a "tree", a flowchart) with each step accounting for the various factors.
It would depend on the "design" goal of the front-loading engineer (side note: I am going to use the term engineer instead of designer since where I work, designer is a five dollar term for CADD tech). Most "artifical selection" I am aware of are done by humans. If humans are the "ultimate goal" of the hypothesised design, then artificial selection is somewhat of a mute point.
It is my belief that Mike Gene has done more to advance the teleological argument in the last 10 years than any IDist.
Mike does have a new site, and has a slight hope he can salvage the material from his former site. On the bright side, Mike did say he has the material from his idthink.net site (which also got hacked).
Comment by JJS P.Eng. — January 14, 2009 @ 12:11 pm
January 14th, 2009 at 12:35 pm
Good day WW. Your post on Information Theory looks interesting. I look forward to reading it as soon as time permits. (So many comments to respond to, so little time)
You have quite the long response. Let's see if I can provide a satisfactory response.
I've had some time to quietly think some more about EBE. It is merely a way to contribute to the teleological investigation that Mike Gene started with The Design Matrix (BTW, I highly recommend it). Mike uses the analogy of a criminal investigation to describe the search for teleological clues. In his Amazon review of DM, Tom Gilson correctly (IMO) states "we're still in the detective stage, not the judge and jury stage."
Engineer do two things: design and analyse. Analysis is studying an existing object or item; design is concerned with the creation of an object or itme.
With these two things in mind, evolutionary biological engineers not only analyse existing natural objects (possibly using reverse-engineering techniques) to conduct a forensic investigation, but instead of determining the reason why something failed, he/she is determining the reason why something worked: in this case, life.
I have a post in the works that suggests a potential design methodology by the front-loading engineer.
It is guesswork. It may amount to nothing more than following a hunch. But, like Mike, I want to follow the rabbit trail.
As for "engineering life" in the present, I believe there is someone attempting that but I can't recall his name (help anyone?).
Is it necessary for there to be human engineers designing life first before an EBE investigation takes place? IMO, no. One can look at life using sound engineering principles and methodology (both general and discipline-specific), and come up with workable and testable hypotheses.
(For you critics out there, one such testable hypothesis would be "Given environment alpha, organism X was designed using engineering methodology A to evolve to organism Y via mechanisms 1, 2, 3… in a specific order." I'll grant you can't test whether engineering methodology A was ever used, but it would become a potential explanation if it was shown that organism X ever evolved to Y via the hypothesised pathway)
Comment by JJS P.Eng. — January 14, 2009 @ 12:35 pm
January 14th, 2009 at 12:45 pm
JJS:
Being one of the privileged few who received The Design Matrix in its very first mailing and signed by the author, I am not surprised in the least that Mike's sites have suffered serious hack-attacks. Obviously, the die-hard Culture Warriors consider him a real danger despite his having been mostly ignored for too long by the 'Big Names' associated with DI and the political/theological "Movement." It's good that his ideas are finally getting some play in those circles.
Comment by Joy — January 14, 2009 @ 12:45 pm
January 14th, 2009 at 1:01 pm
Now let me attempt to answer everyone else…
IMO, FLE's primary focus is on OOL. But to examine the likelihood of FLE, Mike suggests using the present to explain the past (Ch. 7 and 9 of DM). This means examing the FLE design at later stages, not just the beginning. The history of life is complex and focusing on one particular part of that history can make one myopic.
With EBE, I am trying to tease out the FLE hypothesis further by examining potential design methodologies and obstacles that may (or may not) give a better understanding of FLE.
Comment by JJS P.Eng. — January 14, 2009 @ 1:01 pm
January 14th, 2009 at 1:18 pm
To don provan: If I came across as rather snarky, I apologise. At the same time, I get the impression through your questions that you expect FLE to be a "mature scientific theory" rather than one of many OOL hypotheses.
That said, I have a couple of posts in the works that provide follow up to this one. My main objective with these posts are to look at FLE through the eyes of an engineer. Mike Gene has provided evidence for FLE in The Design Matrix. However, he tempers this by saying that the evidence merely makes FLE plausible, not probable, not a "fact" (see Explanatory Continuum in Chapter 2 for more details).
A guess: we just generally don't care. Whether evolution is true or not has zero effect on the way we work. For me, this is an intellectual hobby. Nothing more, nothing less. If this annoys you and others, it's no skin off my back. However, I do appreciate constructive comments from design critics, which I must say I have received more from at EE than here (a shout out to dave, freelurker and aghunt).
Funny. Your first comment starts out "Have at." I interpretted that as a confrontational statement. Using phrases like "insanely complicated" and "impossibly complex" (your second comment), aren't exactly "supportive".
But perhaps I misunderstood your intent. How about we start again? I suggest that I re-respond to your first comment. Deal?
Comment by JJS P.Eng. — January 14, 2009 @ 1:18 pm
January 14th, 2009 at 1:20 pm
This sounds fine. I suspect that you will find that the only methodology A that ever fits will be the trial and error approach of modification with evironmental sorting, but let us know if you find anything else.
By the way, if you answered my question about prior art, I missed it. What does the literature tell you about previous results along these lines? This actually sounds a lot like what biologists do normally, although they might not call it "engineering". But in addition, as I mentioned, I'd expect actual engineering work in this area, as well. Can you contrast your approach to previous approaches? Or aren't there any?
Comment by don provan — January 14, 2009 @ 1:20 pm
January 14th, 2009 at 1:22 pm
(emphasis mine)
Just a guess, but I would say a rigorous examination of probabilistic design/theory was never made by either side. Feel free to provide me with a paper that shows me otherwise.
Comment by JJS P.Eng. — January 14, 2009 @ 1:22 pm
January 14th, 2009 at 1:25 pm
Sorry don, I had to reply to this one.
In Chapter 7 of The Design Matrix, Mike mentions the front-loading designer could have made intelligent uses of chance. Since you've read the book, I'll leave it up to you to look the passage(s) up.
Comment by JJS P.Eng. — January 14, 2009 @ 1:25 pm
January 14th, 2009 at 1:39 pm
Good day Pixie.
You flatter me, Pix, by using the word "research". As I have said earlier in this thread, this is an intellectual hobby.
That said, what I am proposing is in line with what Mike has proposed with FLE. However, as an engineer, I want to get more into the mind of the front-loading engineer (i.e. design methodologies, obstacles, etc.). Part of the FLE investigation is determining the design objective(s)/goal(s).
Not necessarily. Given constraints of materials (proteins, DNA, RNA, environment, etc.), evolutionary mechanisms, and the proposed starting point of single-celled organisms, it would not be surprising that human-like features would not appear for quite some time and thus would not be abundant "right across biology".
Comment by JJS P.Eng. — January 14, 2009 @ 1:39 pm
January 14th, 2009 at 2:00 pm
I expect you to have thought it through enough to have some idea how it could actually be productive and whether it's something that's already been done. But if you're presenting it before that stage, then you can take some of my comments as suggestions for how to make you off-the-top-of-your-head proposal into something real.
But wait. "OOL hypotheses"? You've been describing this as a technique, not a hypothesis, and you said you were going to look at evolutionary pathways. How is this "one of many OOL hypotheses"? I suppose this could be a typo or something, but I think it tips your hand: you're presenting this as a technique, but you think of it as an answer. Until now, I thought you'd at least believe the answer your technique gave you, but now it looks like you've invented the technique to give you the answer you want. Am I wrong?
OK, good. Don't forget to tackle my question which was, "Why in the world would an engineer use such a complicated approach to begin with?"
Many people are very interested in evolution. So the real question is whether you actually know nothing like this has been done before.
Why would it annoy anyone? But keep in mind, ID is full of rank amateurs that think their ideas are novel and that their objections haven't be refuted with evidence even though they have no clue. Don't be one of them. I don't mind if you want to believe with all your heart that no one's had your ideas before and that nothing's already known that would refute them, but do keep you mind open to the possibility.
It seems hard for IDists to believe, but scientists really did arrive at the conclusion of evolution based on evidence, lots of it; they didn't just make it up and then spend all their lives inventing ways to fool people into believing it. Some kind of engineering approach as you suggest might bring new light to some issues, but not if it's applied only to support a specific answer, ignoring contradictory evidence in the pursuit.
Nope. I just meant "go ahead, be my guest", but, it's true, with the subtext of "but don't expect me to get excited about it because I don't think you're going to get anywhere." But I've been explaining why I felt that way, so you know how you can change my mind.
OK, this is really important, then. It is just a fact that a front loaded solution is very, very, very complicated compared to construction. Yes, I think this makes it extraordinarily unlikely that you'll be able to justify it in engineering terms. If you didn't realize you were facing this challenge, now you do.
But this isn't criticism. I just asked how you thought you might get around this challenge, an obvious first step of any engineering evaluation of front loading. I wasn't just claiming you'd be unable to. You've made clear yourself that haven't yet presented anything concrete enough to actually criticize.
Fine with me. I admit I'm skeptical, but I promise to continue being impartial.
Comment by don provan — January 14, 2009 @ 2:00 pm
January 14th, 2009 at 2:14 pm
Using chance is not the same as leaving the results up to chance. That's the logical difference between the teleological explanation of front loading and non-teleological explanation of evolution: front loading achieves some non-random goal, evolution just ends up where the chance variations take it.
Did I somehow imply I've read The Design Matrix? Mike didn't send me a copy, signed or otherwise, so I haven't. I know enough about Mike's ideas that I don't feel a need to pay for the book myself.
Comment by don provan — January 14, 2009 @ 2:14 pm
January 14th, 2009 at 2:50 pm
It could have something to do with front loading at point of origin.
What do you think a hypothesis is? How is a tentatively held conclusion, subject to testing, any more an answer than the "answers" of critics?
This looks like an effort on your part to interpret the words of another in the most uncharitable way possible. Am I wrong?
Comment by Bradford — January 14, 2009 @ 2:50 pm
January 14th, 2009 at 4:28 pm
I guess I'm the only one that took JJ seriously and thought he was proposing a new branch of engineering, as he said. Apparently Bradford has seen through his duplicity and recognizes that he's proposing front loading and all this hoopla about "EBE" is just a cover story.
I don't understand. You're agreeing it's a hypothesis, so there's nothing uncharitable about that interpretation. So you must mean I'm misinterpretting the OP. Let's look:
Am I misinterpreting that? It sure as heck doesn't look like a hypothesis to me. It looks like a way to investigate a hypothesis.
Comment by don provan — January 14, 2009 @ 4:28 pm
January 14th, 2009 at 5:07 pm
I thought you were scientifically literate dp. Not very if you think there is a problem with using an engineering approach to apply a front loaded hypothesis to the origin of life.
Comment by Bradford — January 14, 2009 @ 5:07 pm
January 14th, 2009 at 7:02 pm
OK, don, from the top…
Stop me if this sounds familiar.
Possible constraints are the material properties of proteins, DNA, RNA (i.e. they do not possess unlimited capabilities, but rather finite capabilities that are defined within specific parameters), evolutionary mechanisms (again, these are not a magic wand; these mechanisms have limitations), and the initial environment the first cell was placed in.
This is just a guess, but to demonstrate his engineering brilliance?
I would think that quantifying the capabilities of the aforementioned material properties and mechanism should've been done by now. If not, I'd say it's way past time to find out. Plus, these particular constraints are what they are (i.e. they define the particular object or mechanism).
As for environmental constraint, include it in as a boundary condition (constraint) in an experiment and let 'er rip.
This differs from the non-teleological theories at the OOL level. My proposal was intended to build upon Mike Gene's FLE hypothesis, which focuses mainly on the OOL issue.
Comment by JJS P.Eng. — January 14, 2009 @ 7:02 pm
January 14th, 2009 at 9:08 pm
I guess my point about this wasn't clear enough. My question wasn't what constraints you can imagine the designer might possibly have faced generally. The question I was asking was about the constraints he was facing induced him to use front loading. Are we clear that there's a huge cost to designing a front loaded delivery system on top of designing the biological systems themselves?
If anything, the constraints you list pose immediate problems that we'd expect an engineer would work past directly, making front loading an even less likely approach. I'd be tempted to go so far as to say you've listed some constraints on front loading that a direct approach would avoid, raising even more questions about why front loading would be used.
OK, but let's express it as a hypothesis: "The designer is conceited and likes to show off." I'm not sure how such a hypothesis is useful, though; it kinda sounds like more of an excuse not to worry about something you can't explain rather than an actual explanation. How could we confirm it, for example?
Evaluating specific constraints assumes their existence. My question was how we would go about confirming that your suggested constraints were, indeed, part of the problem the intelligent engineer actually faced. I'm also unclear about how we can evaluate something as a constraint when we don't know the intelligent engineer's capabilities. Are we going to hypothesis His abilities, too?
How is this different than existing work in this area? "Engineering" doesn't seem to play a role in the experimental approach you're suggesting here.
Then I don't understand why you're talking about "proteins, DNA, RNA" as "constraints". In the context of OOL, they are solutions to constraints, aren't they? Are you assuming some other constraint that forces the intelligent engineer to use DNA and RNA and proteins?
I hope this helps you see where you need to clarify and expand. I admit, I find some of this very confusing, but hopefully you can help me look at it the right way.
Comment by don provan — January 14, 2009 @ 9:08 pm
January 14th, 2009 at 10:03 pm
Oh don, don't you know that
godfront-loading designers work in mysterious ways! Why we just aren't able to fathom the designers reasons! Or, lame cop-out #2, the designers goal was [insert any feature that happens to match current biological diversity].Personally I think any designer who uses front-loading is a fool! No good engineer would use this technique, engineers invent things like planned obsolescence. I mean, you're never gonna sell version 2.0 if version 1.0 will change itself into a new version all by itself.
Comment by Todd Berkebile — January 14, 2009 @ 10:03 pm
January 15th, 2009 at 12:32 am
Todd,
Like these guys?.
These designers invested a significant amount of information into the initial state in order to exploit evolution. Sounds like front loading, and they don't seem like fools.
Comment by chunkdz — January 15, 2009 @ 12:32 am
January 15th, 2009 at 1:05 am
chunkdz:
This is an excellent example for proving Don's point. These guys didn't build each precursor antenna, they ran the program to completion and then only built the final product. They didn't make antennas that physically reproduce themselves or that improve themselves after being manufactured. Instead they built a single final product which was the intended result of the evolutionary process.
Comment by Todd Berkebile — January 15, 2009 @ 1:05 am
January 15th, 2009 at 2:35 am
Todd,
I wasn't talking about Don. I was talking about your assertion that "any designer who uses front loading is a fool." Do you think these designers are "fools"?
Comment by chunkdz — January 15, 2009 @ 2:35 am
January 15th, 2009 at 2:40 am
Todd,
What's a program? If information is information, who's to say it's more or less cost effective to use silicon or carbon?
Comment by chunkdz — January 15, 2009 @ 2:40 am
January 15th, 2009 at 2:53 am
Now, now, please lay the blame for this idea where it belongs…on the guys with business degrees, not the ones with engineering backgrounds.
Comment by William Wallace — January 15, 2009 @ 2:53 am
January 15th, 2009 at 2:56 am
P.S. The problem of antenna design could also be tackled with neural networks (and it seems like it would be a better approach, seem being the operative word.) Not that this affects either of your points.
Comment by William Wallace — January 15, 2009 @ 2:56 am
January 15th, 2009 at 6:36 am
JJS:
I guess it depends on what you mean by rigorous. But here's a paper with a stochastic OOL model that you might find interesting:
Szathmary & Demeter 1987. Group Selection of Early Replicators and the Origin of Life. Journal of Theoretical Biology 128, 463-486.
Abstract:
Comment by Raevmo — January 15, 2009 @ 6:36 am
January 15th, 2009 at 8:24 am
[...] also noticed that JJS has a blog entry about something he calls Evolutionary Biological Engineering (EBE). I’m curious to see where he goes with this. Most of the criticism over there seems to be [...]
Pingback by An Engineer in the Matrix « The Design Matrix — January 15, 2009 @ 8:24 am
January 15th, 2009 at 11:21 am
Correct me if I'm wrong, but it seems to me that the main objection to the FLE hypothesis in this thread is "I think an engineer would do it differently". That's cool. State your alternate "engineering" hypothesis then, keeping in mind as you postulate more than one intervention, the number of variables increase.
FLE is an attempt to limit the number of variables inherent in teleological arguments. IMO, Mike did a fantastic job of accomplishing this. If you know a better way to limit the number of variables in a teleological argument, then enlighten us, please.
Comment by JJS P.Eng. — January 15, 2009 @ 11:21 am
January 15th, 2009 at 11:22 am
I got a chuckle out of Todd's statement, "No good engineer would use this technique, engineers invent things like planned obsolescence." In my experience, I've never seen structural, process, mechanical, nor electrical engineers "invent … planned obsolescence". IMO, they'd either laugh at that statement
or would be deeply offended by it 
. But hey, you're free to make whatever statement you want, just as I am free to laugh at it.
Comment by JJS P.Eng. — January 15, 2009 @ 11:22 am
January 15th, 2009 at 11:31 am
It seems to me that the main basis of FLE is the assumption that an engineer might do things a certain way. I was simply poking fun at that idea by suggesting a more human motive like greed. I'm glad you got a chuckle; that was my goal.
Comment by Todd Berkebile — January 15, 2009 @ 11:31 am
January 15th, 2009 at 11:35 am
Yeah, I often build an entire universe just to run an evolution simulator. Can't trust those computers, they run too fast. Can't use an infinitely clever god-brain either, to busy crunching sports statistics. Oh, I get it! You're saying God is a government contractor! Obviously he was working on a cost-plus basis, that's why he so overdid the simulation!
Comment by Todd Berkebile — January 15, 2009 @ 11:35 am
January 15th, 2009 at 12:23 pm
JJS P. Eng:
2 questions:
1. What would an engineer need to know about the constraints relevant to a self-replicating molecule?
2. What would an engineer need to know about the constraints relevant to an information storage and retrieval system?
Comment by Bradford — January 15, 2009 @ 12:23 pm
January 15th, 2009 at 1:59 pm
This isn't front loading. This is setting a design goal, then performing random variations and selection. So, in engineering terms, it's more like evolution as biologists have identified it with its design goal of gene survival.
Comment by don provan — January 15, 2009 @ 1:59 pm
January 15th, 2009 at 2:13 pm
Quoting from chunkdz's link:
Front loading is centered around a process and references initial conditions of that process. The issue is what is the causal genesis of initial conditions? The causal genesis should clue us into the nature of the causes. A mind is involved somewhere in the process of setting up an evolutionary algorithm. Forces of nature can front load the conditions that gave rise to that volcano which just exploded or a mind to the artillary shell that did the same.
Comment by Bradford — January 15, 2009 @ 2:13 pm
January 15th, 2009 at 2:22 pm
Todd,
You refused to answer this, but no matter. We've established that front-loading IS a useful design method:
I think using a network of tiny replicating microcomputers is brilliant. It's an exceedingly effective way to algorithmically search evolutionary space.
Comment by chunkdz — January 15, 2009 @ 2:22 pm
January 15th, 2009 at 2:33 pm
Provan,
Yes, it is.
ie: investing a considerable amount of information into the initial state…
…and using it to constrain and exploit evolution.
What definition of front loading are you using, Provan?
Comment by chunkdz — January 15, 2009 @ 2:33 pm
January 15th, 2009 at 2:49 pm
If you're going to claim that you have an approach that uses engineering to reveal truth about the origin of life or the species, then I expect that approach to be impartial. Specifically, I expect you to be thinking up as many ways an engineer might do it differently and be ruling them out or in. The fact that you are resistant to that and even claim someone else should do it makes me suspect you are not as open minded about the answer as you'd like me to think.
Your basic idea is sound, but when you act like you see your approach as being suitable only for supporting your favorite hypothesis, I start to suspect that, like many IDists before you, you are just going to cherry pick the one or two things that support what you seek to prove and ignore uncountable numbers of contrary things that would cast a negative light on your chosen hypothesis.
The existing theory postulates zero interventions. I'm sure you've heard of it: it's the hypothesis that says the engineering technique being used is mindless variation coupled with environmental sorting. At the very least, I would expect your effort to engage that engineering proposal from the get-go without anyone else having to bring it up.
Indeed, that's the problem with teleological arguments. And Mike doesn't really solve it, he just pretends that all future interventions can be front loaded in advanced. That doesn't really reduce the number of interventions, it just gathers them all together into one big "poof", and then conveniently moves that poof back in time where it is impossible for us to confirm it.
If you're going to approach front loading as an engineering problem, the first thing you need to clarify is whether there's any engineering differences to look for between a normal mindless variation and a front loaded variation. You can only speculate about the front loading event itself, but events in the more recent past can be looked at for artifacts of front loading. Or can they?
Comment by don provan — January 15, 2009 @ 2:49 pm
January 15th, 2009 at 3:09 pm
Front loading builds that state into the system unfolding. The approach used for this antenna development just established a goal — I don't actually think it was that much information, was it? — and selected for it. As I said, that's evolution. If you want to call it front loading, then evolution is also front loading with the goal of maximized gene survival.
I'm assuming that front loading is a distinct proposal from evolution. If front loading is just another way to describe the existing theory, then why bother? It seems to reduce "front loading" to nothing but a term to replace OOL.
But perhaps I'm just missing something. If this is an example of front loading, then can someone explain where the selection mechanism is in biological front loading? In the antenna designed, it's in a computer, and we can see it at any time during the development process, even after the process has completed. Is there something similar in biological front loading? If so, where can we look for it?
Comment by don provan — January 15, 2009 @ 3:09 pm
January 15th, 2009 at 3:15 pm
If an engineering approach is to reveal truth about the origin of life then it must encompass design models whose generating source would be constrained by forces of nature and which would compare anomalies in such models with models of logical thought processes.
I have yet to see any evidence that JJS has a closed mind to any possibilites.
That's a testable assumption.
Comment by Bradford — January 15, 2009 @ 3:15 pm
January 15th, 2009 at 3:17 pm
Provan,
I think it's important that you not dodge this question.
Comment by chunkdz — January 15, 2009 @ 3:17 pm
January 15th, 2009 at 3:19 pm
Then why not specify what you believe are the prerquiste conditions needed to kick start an evolutionary process and link them to their corresponding forces of nature. What would have to be front loaded is a major issue.
Comment by Bradford — January 15, 2009 @ 3:19 pm
January 15th, 2009 at 3:22 pm
Provan,
Does your definition of front-loading put it at odds with what you just described?
Comment by chunkdz — January 15, 2009 @ 3:22 pm
January 15th, 2009 at 3:34 pm
If I had to speculate (i.e. give an engineering guess):
1. The engineer would need to have comprehensive knowledge of the available material resources (amino acids, proteins, molecules involved in replication, etc.), the risks and benefits of various self-replication means, the potential environment(s) the first cell is to be placed in, and probably more than what's coming to mind right now.
2. I'm going to do a bit of a cop-out here and say that Mike Gene has (indirectly) addressed this in The Design Matrix, Chapter 4 and 7(?)
Comment by JJS P.Eng. — January 15, 2009 @ 3:34 pm
January 15th, 2009 at 3:47 pm
While that may be an interesting task for some engineer (or other) to take up and examine, I think you may be missing the point: FLE limits the number of interventions to 1. From my engineering standpoint, that is the most effecient way of doing things. Do it once and do it right. If you think otherwise, you're welcome to present your case. Hey, I'll even give you a forum to present it at EE.
IMHO, the zero-intervention theory contains zero engineering since engineering requires a mind. So therefore, the "existing theory" is mindless and contains zero engineering.
Replace "pretends" with "hypothesises and we're in agreement, but I get a sneaky suspicion that's not your intent.
Comment by JJS P.Eng. — January 15, 2009 @ 3:47 pm
January 15th, 2009 at 4:32 pm
My definition of front loading is, of course, uninteresting. And as I've already said, if someone proposing a hypothesis of front-loading defines front loading in a way that does not put it at odds with the prevailing theory, then that makes it nothing but an attempt to substitute a new name for a theory that already has a name.
Comment by don provan — January 15, 2009 @ 4:32 pm
January 15th, 2009 at 4:56 pm
It is theoretically based predictions and empircally confirmed outcomes which would distinguish theories, not definitions. Prevailing theories render no definitve conclusions about some matters in any case.
Comment by Bradford — January 15, 2009 @ 4:56 pm
January 15th, 2009 at 4:57 pm
I think we need to define "intervention" more clearly. I would say you are using the term to mean "creation event", but considering prevailing theory, the more useful definition would be any event that doesn't agree with the prevailing theory, regardless of whether it involved direct intervention through some kind of active manipulation or indirect intervention through front loading. In that sense, the number of interventions doesn't change, they're just all packaged up in one creation event you call "front loading".
Oh. Well, that's your problem, then. You are implicitly investigating what happened in a way that can only produce an answer involving a mind.
But, of course, I've already very clearly described the engineering approach accepted as the current theory, and everyone can see for themselves that it is engineering and that it doesn't require a mind. To deny that reduces you to just eliminating by fiat evolution as the soul explanation. The fact that an intelligent engineer might do it differently is not convincing except to someone presupposing an intelligent engineer. (And you can't deny it's engineering, since you just describe it and its effectiveness in your "probabilistic design" entry.)
Yes, my bad. What I was trying (and failed) to say is that he pretends this makes a difference. A proposal that calls for multiple, individual interventions is not significantly different; there's no particular reason to assume front loaded interventions less detectable then direct interventions. In either case, we'd look for what makes them different than what would occur "naturally".
But we shouldn't argue about this until we can clarify this larger issue of whether "front loading" implies we should see anything different than evolution alone operating in exactly the way biologists currently accept. You can avoid my complaint that interventions are interventions by denying there are any interventions after the front loading, but I claim such a proposal shouldn't be called "front loaded" because it's really just a claim that at some point something created evolution, and everything since then has played out just exactly as biologists current think it did. Is that the proposal or isn't it?
Comment by don provan — January 15, 2009 @ 4:57 pm
January 15th, 2009 at 4:58 pm
don provan,
There are two ways of looking at front loading:
1. If it's results are not at odds with the prevailing theory, then it is an attempt to show another way other than the prevailing theory that the effects could be accomplished, rather than merely substituting a new name for a theory that already has a name. As far as I can tell, Mike Gene seems to subscribe to this perspective on front loading. In this case, one could just as easily choose to believe a telic explanation for evolution vs. a non-telic explanation for evolution since evidence for one can easily be counted as evidence for the other. Thus, there is no way to *scientifically* determine which is the better explanation and which one you choose will depend on other extra-scientific considerations.
2. It's results are at odds with the prevailing theory, because proponents contend that the prevailing theory is not up to the challenge of producing certain effects without front-loading by a foresighted system/mechanism. This is where I fit in. As an example, is there any evidence that an evolutionary algorithm will produce a proofreading system to minimize and counter errors in information compiling absent any front loaded organization seeded by a previous foresighted system/mechanism? Or, to bring it down to a more fundamental mathematical level, will law and chance *absent* any front-loading by a previous foresighted system/mechanism produce specific effects (ie: CSI)?
Comment by CJYman — January 15, 2009 @ 4:58 pm
January 15th, 2009 at 5:09 pm
It never fails to amaze me that people advocating hypothesis think it makes any sense at all to ask someone else to define what the hypothesis they're proposing says. If there's any leeway at all for me to define "front loading", that concedes that there's no legimate proposal to begin with.
So if you'd actually like to participate, why don't you try to clear up how I should understand front loading. I've already explained a couple different ways that the antenna development differed from anything I'm seeing in proposals for biological front loading, so start there and explain why an external measure of succcess is like front loading and different from evolution. As I explained, the way I'm looking at it, it's just the reverse. So how should I look at it?
Comment by don provan — January 15, 2009 @ 5:09 pm
January 15th, 2009 at 5:11 pm
Or what about evolution itself? Will evolution occur at all without front-loading? One way to test this is to take a sample of background noise and an arbitrary collection of laws and run it through a "selection filter" based on more background noise and arbitrary laws (or the same laws and noise as before) and see if any evolution will take place.
Any other suggested ideas to test the power of evolution?
Comment by CJYman — January 15, 2009 @ 5:11 pm
January 15th, 2009 at 5:20 pm
dp:
Sounds like parodies of ID about to kick into gear. Creation event? As in the formation of a self-replicating molecule for example? That's consistent with prevailing theory. Why not apply an engineering mindset to that and see where our replicator goes.
Comment by Bradford — January 15, 2009 @ 5:20 pm
January 15th, 2009 at 5:26 pm
That's simply not correct. We can directly observe evolution and its mechanisms, and have significant evidence of the process over geologic time-scales. The Theory of Evolution is strongly supported.
Without a clear and rigorous metric for CSI, the claim is unsupportable.
Comment by Zachriel — January 15, 2009 @ 5:26 pm
January 15th, 2009 at 5:26 pm
So it does seem to me as if this is the way it's been treated by Mike and JJ, but that means by definition it doesn't tell us anything we don't already know. It literally makes no difference.
If that's really what we're talking about, then "front loading" is an inaccurate, even deceptive, term. The correct term is "creation". I don't say that to defile "front loading", but merely because I see nothing about the proposal as thus described beyond "something at some time created evolution". We can still use these engineering ideas to study the creation, but if "front loading" means nothing different than "evolution", then calling it front loading doesn't change the engineering picture, either.
Comment by don provan — January 15, 2009 @ 5:26 pm
January 15th, 2009 at 5:34 pm
Last time I checked, FLE explicitly assumes only one intervening act. The vast majority of the other telic explanations assume several (more than one) intervening acts. So I fail to see your point (italicised above).
No, I am explicitly investigation whether FLE is plausible.
Actually don, you described the stochastic approach (pure chance + time). IOW, no engineering involved. For further proof, take another look at the engineering definition from the above post:
The non-telic process you describe is nowhere near an engineering approach.
There is no denial here, just confusion on your part on what constitutes engineering. Mike and I have (in plain English) stated that engineers make use of chance. The above post describes how structural engineers make use of chance in their designs. You are conflating blind/pure chance with using chance. That's your problem, not mine.
That "something created evolution" is not the claim. FLE claims that the front-loading engineer used evolution to obtain design objectives and assumes the evolutionary mechanisms exist prior to the intervention.
Comment by JJS P.Eng. — January 15, 2009 @ 5:34 pm
January 15th, 2009 at 5:38 pm
Provan,
Since you were so sure that the antenna designers were not using front-loading, I presumed you knew what front loading is. Your subsequent comments reveal that you don't, or have a very different definition than the DM definition.
If you ever care to reveal what you think "front-loading" is, we might actually have something to discuss. Otherwise, carry on with bashing something that you seemingly don't even understand.
Comment by chunkdz — January 15, 2009 @ 5:38 pm
January 15th, 2009 at 6:06 pm
CJYman: Thus, there is no way to *scientifically* determine which is the better explanation and which one you choose will depend on other extra-scientific considerations.
Zachriel:
Like usual, you comment as if you didn't read what I had said or else you seem to purposefully not respond to what I said and/or not understand what I said. Obfuscation at its best. You truly have turned it into an art.
I agree that the TOE (along with its observations and mechanisms) is strongly supported. My point was that those people who take perspective #1 see no difference between evidence for a-telic vs. evidence for telic evolution. Evidence for one does not rule out the other as they are able to work any evidence for evolution into either framework. You've been around long enough to read through Mike Gene's work and see that this is indeed the case. BTW, I do not prescribe to #1. As I explained, I am a proponent of viewpoint #2.
CJYman: Or, to bring it down to a more fundamental mathematical level, will law and chance *absent* any front-loading by a previous foresighted system/mechanism produce specific effects (ie: CSI)?
Zachriel:
I agree. The only difference is that you think, without providing any evidence to back up your assertions, that CSI is not clear and rigorous enough. However, I have already shown that:
1. Foresight (modeling the future and generating pre-set targets that are not yet obtained) does exist and does produce effects that can not be mathematically described as resulting from chance processes or merely the result of law as a mathematical description of regularity or the result of the physical properties of the units involved.
2. These effects of foresight may be measured as CSI.
3. CSI is measured as an improbable specified pattern which uses up all probabilistic resources and thus rules out chance as a best explanation. Once mathematical descriptions of regularities and patterns based on physical properties of the units are ruled out, law is also ruled out.
4. So far no one can show that a collection of law and chance absent previous foresight will produce CSI.
FYI, a door opening and closing and represented as 101010101 can be described as law-like by a mathematical description of regularities (an algorithmically compressible pattern) and is thus ruled out of consideration. Yes, I have slightly updated my view of CSI and this issue of algorithmic compressibility is one area in which I disagree with Dembski since Langton's ant seems to produce algorithmically compressible patterns based on arbitrarily assigned initial conditions. However, I still use compressible patterns as an example of how to measure for CSI without consideration for law-like processes. In fact, I have shown the equation for CSI and how it is used.
CSI is merely a measurement of a specified pattern against all specified patterns and then measured against the information content of all trials (probabilistic resources).
Comment by CJYman — January 15, 2009 @ 6:06 pm
January 15th, 2009 at 6:20 pm
Yes, I understood that. It's vacuous, at best.
That's fine. But you forgot the rigorous metric.
Comment by Zachriel — January 15, 2009 @ 6:20 pm
January 15th, 2009 at 6:27 pm
CJYman:
"1. If it's results are not at odds with the prevailing theory, then it is an attempt to show another way other than the prevailing theory that the effects could be accomplished, rather than merely substituting a new name for a theory that already has a name."
don provan:
It makes a big difference, not in result [assuming evolution absent front loading is indeed up to the task] but in method — what caused the initial conditions.
1. Telic = modeling of the future was involved and the initial conditions were set so that a future target would be obtained.
2. A-telic = background noise and an arbitrary set of laws fashioned a ratcheting filter which then produced the same results as if an engineer had front-loaded the initial conditions to obtain some target(s).
don provan:
Sure, "creation" of the initial conditions. This happens regularly when utilizing EAs to solve problems. Another word is front loading, since you are literally front loading information (as a specific organization to produce specific ends) into the initial conditions.
don provan:
You seem to have missed the point. Front loading was never meant to replace evolution nor was it meant to = evolution. It was meant to show what causes evolution or how evolution can be used to achieve specific ends. In fact, front loading can not mean the same thing as evolution, since you can front load information and then not continue with the subsequent unfolding (evolution) of that information. Front loading happens first, evolution happens next and unfolds the front loaded information.
Comment by CJYman — January 15, 2009 @ 6:27 pm
January 15th, 2009 at 6:37 pm
CJYman: My point was that those people who take perspective #1 see no difference between evidence for a-telic vs. evidence for telic evolution.
Zachriel:
According to what I have explained, only as vacuous as stating that evolution is a-telic. But, then again, that is why I subscribe to #2 as opposed to #1.
CJYman: The only difference is that you think, without providing any evidence to back up your assertions, that CSI is not clear and rigorous enough… FYI, a door opening and closing and represented as 101010101 can be described as law-like by a mathematical description of regularities (an algorithmically compressible pattern) and is thus ruled out of consideration.
Zachriel:
Ummm … no I didn't. I've previously explained it to you complete with the math and examples. Whatever … obviously neither of us will change his mind, so I guess its your assertions absent a negation of the math and concepts vs. my explanation and examples.
Oh, BTW, how do you define "rigorous metric?"
Comment by CJYman — January 15, 2009 @ 6:37 pm
January 15th, 2009 at 6:37 pm
Just a reminder on the vacuous issue. Claims that ID is vacuous start with an already existing self-replicating capacity. So the default presumption has been that front loaded features just exist without an empirically supported cause. That's why an engineering approach is called for. It allows for explanations that are currently excluded a priori.
Comment by Bradford — January 15, 2009 @ 6:37 pm
January 15th, 2009 at 7:52 pm
The evidence indicates that, whatever its origin, life evolved through natural non-telic processes. Of course, some agent may have manipulated the path of evolution, but we have no evidence of this.
A rock erodes from a mountain and tumbles down. Science considers these processes to be non-telic. Maybe the rock was meant to fall at a certain time and in a certain way, but there is no scientific evidence of this. Saying teleological explanations are of equivalent scientific merit—absent evidence—is incorrect.
Scientific theories of biological evolution are non-telic in just this way. We have robust explanations of how the process works, and no evidence of teleology.
So you won't provide an explicit metric for calculating CSI.
It means being able to take a phenomena (within the domain of applicability—which Dembski claims is "objects"), make appropriate observations and calculate in a unambiguous manner a result. In this case, I would rather expect CSI to be a formula with well-defined expressions. You might try searching the scientific literature for applications of CSI.
This is Dembski's formula for specified complexity:
χ = –log2 [ BIGNUM · ϕS(T) · P(T|H) ]Comment by Zachriel — January 15, 2009 @ 7:52 pm
January 15th, 2009 at 9:12 pm
JJS:
So at what point did this single intervening act occur? Was it the introduction of the first self-replicating molecules? Or were there already naturally occurring self-replicating molecules, but did the intervention introduce the first cells? Was it before or after the evolution of the genetic code? Any ideas? Anybody?
Comment by Raevmo — January 15, 2009 @ 9:12 pm
January 15th, 2009 at 9:42 pm
Could you humor me and repeat it for my benefit?
If you don't mind, I don't think we should let Zach define "rigorous metric". English, on the other hand, defines it as a measurement clearly enough defined that anyone making it would, barring errors, come up with the same answer.
Comment by don provan — January 15, 2009 @ 9:42 pm
January 16th, 2009 at 11:26 am
Show of hands: Based on the above comment, does Raevmo demonstrate any understanding of what FLE actually hypothesises?
Comment by JJS P.Eng. — January 16, 2009 @ 11:26 am
January 16th, 2009 at 11:30 am
Normally, I wouldn't stoop to comments like the one I just made, but I'm in a pissy mood this morning.
You want to bring it, go ahead, just remember three things:
1. I'm triggerhappy this morning and will have no (immediate) regrets in flushing comments down the memory hole.
2. If you don't want your comments flushed, please demonstrate some understanding of FLE and the basic concepts I've laid out in the last two posts.
3. Don't f#$& with the
babysitterengineer!Comment by JJS P.Eng. — January 16, 2009 @ 11:30 am
January 16th, 2009 at 12:24 pm
JJS:
Let me rephrase then. As far as I understand, FLE hypothesizes that The Designer somehow seeded our planet with the first life forms. My question is: what did these initial life forms look like?
Specifically: were the initial life forms cellular? Or do self-replicating molecules count as life forms? Something in between? Did the first life forms already have a genetic code or were they RNA-based?
Comment by Raevmo — January 16, 2009 @ 12:24 pm
January 16th, 2009 at 12:42 pm
Raevmo:
I don't think these questions should be the major focus. Instead of attempting to distinguish between life forms and non-life we ought to be foucused on whether or not we can predict a fundamental evolutionary trajectory of self-replicators even in ideal replicating conditions. My answer to that is a definitve no. IOW, aside from claiming that molecular properties, favoring replicating capacity would have an obvious selective advantage, there is little more to be said.
Again, even if you assume they were RNA based that would not explain subsequent causal steps en route to a code. What are the prerequiste conditions to generating a code?
Comment by Bradford — January 16, 2009 @ 12:42 pm
January 16th, 2009 at 12:58 pm
How can they not be the major focus when the question is one of engineering? What engineer issues can you investigate for an event you know nothing about? When I asked about constraints, JJ specifically mentioned RNA and DNA, and while I still don't understand how he thinks they relate, it tells me that JJ, at least, thinks RNA and DNA might play a role either before or after and, therefore, questions such as the ones Raevmo is asking should in fact be a major focus.
Exactly. An absolutely fundamental engineering question that you seem to be suggesting we can't even begin to answer.
Comment by don provan — January 16, 2009 @ 12:58 pm
January 16th, 2009 at 12:59 pm
Thank you for re-phrasing.
I concur with Bradford's reply.
Comment by JJS P.Eng. — January 16, 2009 @ 12:59 pm
January 16th, 2009 at 1:03 pm
I would put heavy emphasis on might. When you asked me about constraints, I gave you potential constraints which implies that the constraints I listed may not have been constraints. This is an investigation that's being conducted, not a trial.
Comment by JJS P.Eng. — January 16, 2009 @ 1:03 pm
January 16th, 2009 at 1:04 pm
Bradford:
What do you mean by a "fundamental evolutionary trajectory", and why is it important?
Well, that settles it then.
Again, just because we don't know how it happened doesn't mean it's impossible or even unlikely. You really need to try and understand this simple principle.
Comment by Raevmo — January 16, 2009 @ 1:04 pm
January 16th, 2009 at 1:13 pm
Sure. But am I wrong to conclude you think this area should be a major focus? Or is Bradford correct that you intend to apply engineering principles to an event you know nothing about?
Comment by don provan — January 16, 2009 @ 1:13 pm
January 16th, 2009 at 1:18 pm
Again, even if you assume they were RNA based that would not explain subsequent causal steps en route to a code.
I did not write impossible or unlikely. A theoretical framework that provides the expectation that generations of self-replicating molecules leads to any discernable pattern would be a start.
Comment by Bradford — January 16, 2009 @ 1:18 pm
January 16th, 2009 at 1:21 pm
Don't misattribute comments to me. Ignorance is evidenced in your comments.
Comment by Bradford — January 16, 2009 @ 1:21 pm
January 16th, 2009 at 1:22 pm
JJ,
Can I ask a simple question? A front loading event occurs at time T. Which of the following is true?
A. Everything after time T will look as if it's an unfolding of natural processes.
B. Certain events after time T will be contrary to what we expect from natural processes.
Thank-you for your help.
Comment by don provan — January 16, 2009 @ 1:22 pm
January 16th, 2009 at 1:28 pm
If human input occuring at time t alters subsequent events is the historic trail natural?
Comment by Bradford — January 16, 2009 @ 1:28 pm
January 16th, 2009 at 1:32 pm
It was a conclusion based on your position: if asking questions about what the event involved isn't a major focus, then whatever we're doing, it will involve not knowing anything about the event suitable to engineering analysis. Right?
If that's not what you meant, then explain what you did mean. Slinging mindless insults just tells me you're not really interested in contributing.
Comment by don provan — January 16, 2009 @ 1:32 pm
January 16th, 2009 at 1:34 pm
don, neither answer fully describes the FLE hypothesis. Again, I encourage you to spend the $20 (or so) to obtain The Design Matrix and read it, especially Chapter 7. In Chapter 7, you will see that Mike explicitly states that FLE uses evolution to carry out design objectives. The question(s) being asked is not whether FLE and the "prevailing theory" look the same, but what are the clues that point to the design objectives and how evolution was used.
Comment by JJS P.Eng. — January 16, 2009 @ 1:34 pm
January 16th, 2009 at 1:39 pm
Your twisted versions of what I write are non-contributory.
Comment by Bradford — January 16, 2009 @ 1:39 pm
January 16th, 2009 at 1:40 pm
Does it matter? Is my question not clear unless I make a ruling on this point? If so, then my ruling is that humans and their actions should be considered natural for the purposes of my question.
Comment by don provan — January 16, 2009 @ 1:40 pm
January 16th, 2009 at 1:41 pm
JJS, speaking from years of experience, there is an unmistakable attempt to derail your thread.
Comment by Bradford — January 16, 2009 @ 1:41 pm
January 16th, 2009 at 1:59 pm
Bradford, duly noted.
don, you got flushed because of your playground comment about Bradford. If anyone is demonstrating ignorance, it is you.
Last warning, don, or you get banned from the thread. See comment for ground rules.
Comment by JJS P.Eng. — January 16, 2009 @ 1:59 pm
January 16th, 2009 at 2:09 pm
Back to topic, don.
Comment by JJS P.Eng. — January 16, 2009 @ 2:09 pm
January 16th, 2009 at 2:10 pm
JJ, can you go into more detail about how we can apply engineering principles to an event without first asking what it involved? Yes, I understand there are lots of unknowns, but how can you apply engineering principles without first resolving or at least conjecturing about those unknowns? You conjectured about RNA and DNA, and that was OK, but Raevmo conjectured about what the initial life forms might look like, and suddenly he's showing a lack of understanding of front loading. Even with all you've said, I don't see how those two ideas are significantly different. Since you feel very strongly that they are, this should be another easy way for you to illuminate what it is I'm missing about your ideas.
Comment by don provan — January 16, 2009 @ 2:10 pm
January 16th, 2009 at 2:15 pm
don, my conjecturing and Raevmo's conjecturing follow two different trajectories. My focus is on FLE while Raevmo is focusing on one specific non-telic hypothesis of OOL.
Again, read Chapter 7 of The Design Matrix.
Time to move on.
Comment by JJS P.Eng. — January 16, 2009 @ 2:15 pm
January 16th, 2009 at 2:16 pm
Are you saying my question isn't on topic? I meant it to be a basic issue that anyone would have to understand before they can follow you. Either the engineering issues you want to look at impact subsequent events in an observable way or they do not. That makes a fundamental difference to how we apply engineering to the event, doesn't it?
Comment by don provan — January 16, 2009 @ 2:16 pm
January 16th, 2009 at 2:23 pm
Defining life is a semantic issue. Ascertaining pathways to result y from initial condition x while invoking process z is a matter subject to testing. My point to Raevmo centers around the way a process relates to autocatalytic RNA.
Comment by Bradford — January 16, 2009 @ 2:23 pm
January 16th, 2009 at 3:04 pm
Let me put it in engineering terms. This thread and FLE in general is primarily involved in conceptual design. You are looking to skip ahead to detailed design. Yes, details are important, but so is determining if a conceptual design is plausible.
Last time. Move on.
Comment by JJS P.Eng. — January 16, 2009 @ 3:04 pm
January 16th, 2009 at 3:39 pm
Raevmo: Ironically, the results suggest that there was no engineer.
Specifically what results do you believe suggest that?
Comment by kornbelt888 — January 16, 2009 @ 3:39 pm
January 16th, 2009 at 3:42 pm
Actually, this has nothing to do with a "creationism". I see no mention of Bibles or other holy book or personal revelations by which JJS intends to guide such an endeavor.
Comment by kornbelt888 — January 16, 2009 @ 3:42 pm
January 16th, 2009 at 4:08 pm
JSS:
You can put a stop to naysayers by making your EBE have the goal of actually designing a front-loaded life form, being free to use what is presently known about earth's life forms. In other words, it should not be about how an engineer designed life, but about how we could achieve the same thing. All of the gainsaying is irrelevant and unnecessary in that light.
Maybe that's what you have in mind.
Comment by kornbelt888 — January 16, 2009 @ 4:08 pm
January 16th, 2009 at 4:13 pm
Bradford:
It seems to me that if you want to reverse-engineer FLE, then it makes a big difference whether we have to reverse-engineer an RNA world or life with the current genetic code.
Comment by Raevmo — January 16, 2009 @ 4:13 pm
January 16th, 2009 at 4:41 pm
Raevmo:
I agree but don't think you have to decide the issue of what life is beforehand.
Comment by Bradford — January 16, 2009 @ 4:41 pm
January 16th, 2009 at 6:32 pm
I believe Mike Gene hypothesized that it was the first cells that would have been designed, and wisely so since the cell is the most basic unit of life that we know of.
It might be just as wise to use that as our foundation rather than muddying the waters with fanciful ideas of self-replicating molecules being front-loaded.
Comment by chunkdz — January 16, 2009 @ 6:32 pm
January 16th, 2009 at 6:55 pm
I'll go along with this if all the other Telic Thinker's agree with it.
This is a very clear cut answer. If JJ agrees with it, we could have saved a lot of time if he'd simply provided it as the answer to Raevmo's original question instead of saying "Show of hands: Based on the above comment, does Raevmo demonstrate any understanding of what FLE actually hypothesises?" Ridiculing Raevmo did nothing to help me understand this very simple yet critical point.
Now is it proper to ask what there was in the environment for the front loading designer to work with? That's the other aspect of Raevmo's original question which isn't covered by your clear statement about what was produced in the front loading event.
I certainly agree with this. But there's a problem here: we're discussing applying engineering principles, so what are we to apply them to? JJ's been very open and honest about all the conjecturing he's doing in order to speculate about engineering applications, and I don't see any way to approach EBE without that kind of guessing. But how do we distinguish open ended conjectures from fanciful ideas? They seem like exactly the same thing.
Comment by don provan — January 16, 2009 @ 6:55 pm
January 16th, 2009 at 7:24 pm
Me, too. Specifically, I assumed it meant something relate to the English word "front loading" and that it fit in with everything else I know about ID and it's version of "information theory".
How so? Do you have anything at all to offer other than "I say this method of antenna design is the same as front loading"? I've explained exactly why I thought it differed from my understanding of front loading, so how is me trying to present a definition going to make it any clearer?
I've already revealed what I thought front-loading is, and you said I was wrong. If I specifically produced a definition, you'll just say that's wrong. The only way we can make any progress is if you start telling me what's right.
I explained this very clearly, I thought: the antenna design method puts a virtual antenna into a simulated environment. The information is front-loaded into the environment, not into the antenna. Is the front loading concept that the environment is set up to produce the desired results?
Comment by don provan — January 16, 2009 @ 7:24 pm
January 16th, 2009 at 7:33 pm
Provan,
Can you link me to it?
Comment by chunkdz — January 16, 2009 @ 7:33 pm
January 16th, 2009 at 8:09 pm
Most ID-friendly posters around here have read Mike's book, and either think along the same lines, or are very friendly to Mike's ideas. Do yourself a favor and get a copy. I think you'll at least come away with a much better idea of who and what you're dealing with here, if not find Mike's view interesting. Hard to believe anyone who hangs out here as much as you hasn't already.
Comment by kornbelt888 — January 16, 2009 @ 8:09 pm
January 16th, 2009 at 10:20 pm
1. Call me JJS, call me JJ, call me for a beer (Canadian preferably), but please don't call me JSS. Slip of the typing finger, kornbelt?
2. I get a sneaky suspicion that even if a front-loaded life form was designed by engineers in the present, the
peanut gallerynaysayers would still be yapping.3. Isn't someone already trying to design "human-made life"? The name escapes me…
Comment by JJS P.Eng. — January 16, 2009 @ 10:20 pm
January 16th, 2009 at 10:29 pm
…if you actually would have read Chapter 7 of The Design Matrix (and no I don't get a cut from all book sales
)
Sure it's proper. But as I've said before, environment is a rather broad term. Can you focus your question a little more so we don't waste another 100 comments.
*raises the beer mug* I'll drink to that!
Comment by JJS P.Eng. — January 16, 2009 @ 10:29 pm
January 16th, 2009 at 11:01 pm
Craig Venter
Comment by chunkdz — January 16, 2009 @ 11:01 pm
January 16th, 2009 at 11:50 pm
One thought experiment I’ve used to explore the problem of the origin and evolution of life from an engineering perspective is the idea self replicating “Von Neumann” machines. Specifically I’m thinking about using such machines to “terra form” another planet like Mars to make it more biologically habitable.
The basic scenario is really quite simple. A small number of machines capable of living-off-the-land and replicating themselves are sent to Mars and begin to replicate. In addition to replicating them selves they also begin to build “Environmental Gas Generators” or EGGs The purpose of the EGGs is to release gases from processed surface material into the thin Martian atmosphere to thicken it and increase the global temperature. Once the atmosphere was thick enough then plant life would take over and finish the job decreasing CO2 levels and increasing the O2 level in the atmosphere.
One problem that presents itself is what I call the sorcerers apprentice effect: The machines continue to replicate until they overrun most of planets surface. Obviously, to control this something has to be written into the main program (gene analog) of these machines so that they begin to slow down once the atmospheric pressure reaches a certain level. Maybe at some point the program will have them evolve a cannibalizing version of the machine which will begin to clean up the surface. Once the machines have finished do they go extinct? Or, do they evolve to serve another purpose? Building fully automated manufacturing facilities, perhaps?
This to me is a good illustration of how front loading might work. The first cells, however they got here, must be robust enough to adapt and evolve to changing conditions. However, a certain level of fore knowledge needs to programmed into the genes of these cells or this process simply won‘t get very far. Just saying that it was one lucky accident after another is not a very good explanation. Actually, in my opinion, it is not an explanation at all.
Comment by JOHN_A_DESIGNER — January 16, 2009 @ 11:50 pm
January 17th, 2009 at 9:50 am
Comment by Zachriel — January 17, 2009 @ 9:50 am
January 18th, 2009 at 4:38 pm
Zachriel:
Yes, yes, I've heard your assertions sans evidence many times already. Life evolved through non-telic processes? Care to provide some evidence that CSI can be created by chance and law absent previous foresight?
It is obvious to me that you are merely being hyper-skeptical, where obviously nothing will change your mind. All you have are assertions. You've never backed up any claim that CSI can be generated by chance and law absent previous foresight. However, a very common observation is that CSI describes patterns resulting from foresight. Furthermore, there is no evidence that CSI can be produced by chance and law absent previous foresight, and it is common knowledge that the organization of nucleotides is neither merely described by regularities nor by the physical properties of the nucleotides themselves. Thus, law defined as mathematical description of regularities or as patterns caused by the physical properties of the material used is ruled out.
Oh, and that doesn't even address perspective #1 (which is what you were responding to initially and which I don't subscribe to), which states that evidence for telic can be taken as evidence for non-telic evolution and vice versa. According to viewpoint #1, saying that life evolved according to non-telic processes is just as vacuous as stating that life evolved from telic processes unless you can provide evidence for atelic evolution which also eliminates telic evolution. Do you have such evidence? If so, run it past Mike Gene, since he is the prime promoter of viewpoint #1 and has been apt at showing that any evidence for a-telic evolution can also be seen as evidence for telic evolution.
But, if you are going to discuss the subject with me, you'll have to address viewpoint #2 which states that evolution of CSI can't even occur in the first place absent previous intelligence.
Zachriel:
For sure, these processes are well within the domain of chance and law, and hence non-telic. There is no need for foresight.
Zachriel:
Of course, when did I ever say otherwise? It also cuts the other way. Saying that non-telic explanations are of equivalent scientific merit — absent evidence — in explaining something such as an essay is incorrect as well. We both know that science deals in evidence, so I'm not sure why you bring this up.
Zachriel:
Incorrect, as modern theories of evolution do not even deal with the telic vs. non-telic cause of evolution. As you state, the modern theories of evolution merely explain how the process works (with plenty of evidence of previous application of foresight). What you are trying to sell here is akin to stating that scientists understand how computers operate, thus there is nothing teleological about them. Well, of course, once the computer is set up, it operates according to natural law and there is nothing telic about how electricity flows through logic gates. But that is missing the point of where the computer and sufficient organization of logic gates comes from in the first place. Thus the question I ask you is, "what evidence do you have that evolution of CSI will occur absent previous foresight?"
Please provide evidence that chance + law [absent foresight] will produce CSI or an EA which will then produce CSI. Maybe start with some background noise and an arbitrary collection of laws.
CJYman: I've previously explained it to you complete with the math and examples.
Zachriel:
Eh?!?!?!?! Where'd you get that from? Did you even read what I wrote? I've already done so on my blog and elsewhere specifically for you, but whenever I do you either ignore me or spout obfuscations about me lying or some such nonsense. I even explained what I meant by "better than chance performance in search" for you about five times in a row and you first ignored me and kept asking that I explain it to you and then you actually lied by stating that I didn't care to explain "better than chance performance."
Why should I expect anything to change now?
CJYman: BTW, how do you define "rigorous metric?"
Zachriel:
Ummmm …. sure … and I've already shown you exactly how it works on my blog quite some time ago. I've already explained CSI and I've shown how your criticism (yes only one that I can remember) is invalid.
Why would I search the literature for it, when it seems that hardly anyone cares to understand CSI, much less even critique it on its own merits? Why don't you find the best critique of it in the literature and we'll discuss it. So, do you finally wish to discuss CSI or not? I don't have much time to waste, which is why I'm not willing to engage in your normal method of obfuscating in circles.
Comment by CJYman — January 18, 2009 @ 4:38 pm
January 18th, 2009 at 6:44 pm
Believe it or not, that's a significant stumbling block for many IDers.
We can't address that claim until we have a clear metric for CSI.
That's fine. So, let's try a few examples using Dembski's definition, and see how to determine ϕS(T) and P(T|H). First, what is the specified complexity of the simple ordered pattern,
11111111...I would assume that ϕS(T) is a small integer, and P(T|H) would be a uniform probability distribution. So assuming the sequence is 1000 digits long, χ > 1 or a positive result. Hmm, is that right? Or do we choose a different P(T|H)? Perhaps we take the average distribution of digits across the sequence. In that case, it seems to work. Do we take the average distribution of digits in all cases?
Comment by Zachriel — January 18, 2009 @ 6:44 pm
January 18th, 2009 at 9:24 pm
I, for one, have done nothing but try to understand and discuss CSI on its own merits. So I would ask whether it might not be you that are unwilling to discuss it on its own merits, preferring instead to treat it as an icon to be defended at all costs regardless of its merits.
We could just start with CSI's Wikipedia entry which has a section summarizing the principle criticisms.
Indeed, most critics feel exactly the same way about people that defend CSI.
Comment by don provan — January 18, 2009 @ 9:24 pm
January 19th, 2009 at 1:46 am
Hello don provan,
You state:
Ask whatever you wish. Let's discuss. Or click on my handle and read my introductory explanation of the topic in the post "Specifications (part 1) … what are they" below "a case for a naturalistic case for intelligent design" in the left margin.
I am merely convinced by the concept and mathematics behind CSI and Conservation of Information and I have not seen a good argument against these concepts being measurements of the effects of foresight.
By all means, let us discuss the merits. I only ask that we give each other the benefit of the doubt, do not engage in character assassination, and debate the merits in a straightforward manner.
don provan:
Which do you find the most convincing?
CJYman: So, do you finally wish to discuss CSI or not? I don't have much time to waste, which is why I'm not willing to engage in your normal method of obfuscating in circles.
don provan:
I'm sure that is the case. However, I was responding directly to Zachriel. We have a little bit of back-and-forth history when it comes to discussing these issues.
Comment by CJYman — January 19, 2009 @ 1:46 am
January 19th, 2009 at 2:40 am
CJYman: Care to provide some evidence that CSI can be created by chance and law absent previous foresight?
Zachriel:
Very true.
CJYman: Why would I search the literature for it, when it seems that hardly anyone cares to understand CSI, much less even critique it on its own merits?
Zachriel:
I've already used this as an example on my blog. I calculated the CSI. I posted it on Feb 11, 2008 under "Specifications (part 1) … what exactly are they." Apparently you responded to it. I've already shown the problems with your door slamming pattern under "specifications (part 2) …" The rest of your criticisms are actually my criticisms as well, since describing patterns with human language is extremely ambiguous, which is why algorithmic compressibility works so well as a non-ambiguous example. A pattern's algorithmic compressibility can be an objective measurement and all we have to do is make sure we are comparing measurements from the same programming language (ie: print 1 X 20). This of course is the same for *all* measurement in that comparing measurements can only be done once an agreed upon arbitrarily constructed system (ie: imperial) is being used.
Dembski: Can objects, even if nothing is known about how they arose, exhibit features that reliably signal the action of an intelligent cause?
Zachriel:
As long as I am understanding you correctly, the answer is "no" in order to calculate the specificity the average distribution of digits is not always taken into account. In this case however, the average distribution is important since the pattern is specified as a result of it being algorithmically compressible (as a result of its digits' average distribution). Since it is the digits average distribution which causes it to be specified, we need to measure its specificity based on its algorithmic compressibility. We need to discover the probability of finding that pattern based on its deviation from statistical randomness. The rest is explained on my blog and I quote:
"Let’s return to our first example – the long pattern of 1s. Dembski has stated that the higher the compressibility of the pattern, the higher the specificity. Why is that the case? That is the case, since the less compressible the pattern is, the more it becomes algorithmically complex and the more random it becomes. These algorithmically complex patterns are the types of patterns that will be generated by random processes.
For example: It is way more likely for a pattern with the same compressibility as “100101111101100001010001011100" to be generated by a random flip of the coin than for a pattern with the same compressibility as “111111111111111111111111111111" to be generated. Why? Because, the only other pattern that can be formulated with the same algorithmic compressibility as the pattern of 30 1s is “000000000000000000000000000000,” which can be compressed to “print 0 X 30.” However, there are many more other patterns with the same compressibility as the first, more random pattern (assortment of 50% 1s and 50% 0s which are sorted in a truly random fashion as per the rules of statistics*). So, it is easier for chance the “find” one of the less compressible patterns, because there are more patterns with the same lower compressibility than there are patterns with the same higher compressibility — in the above case there are only 2 out of 1,073,741,824 patterns which have the highest compressibility and those two patterns are shown above in the repeating 1s and the repeating 0s."
The rest of the calculation is on my blog.
Then, the main difference between calculating for a specification based on algorithmic compressibility versus calculating for a specification based on function is as follows:
"When measuring for a functional specification (within a set of functional "islands"), you apply the same equation, however, when measuring the specificity you take into account all other FUNCTIONAL patterns (able to be processed into function *by the system in question*) that have the same probability of appearance as the pattern in question. You do that instead of taking into account all equally probable compressible patterns, since you are now measuring for functional specificity as opposed to compressible specificity. Therefore, you can only measure for functional specificity and then a specification based upon a high understanding of the system and pattern in question."
So the main thing to remember is that when measuring for specificity, measure all specified patterns of the same probability (based on what causes the patter to be specified whether function, or compressibility [average distribution], or other) against all possible patterns not specified.
Comment by CJYman — January 19, 2009 @ 2:40 am
January 19th, 2009 at 7:57 am
CJYman wrote (quoting Dembski, I suppose):
Doing so requires knowing which patterns are functional and which aren't. That demonstrates that CSI is a measure of someone's knowledge and not an objective physical quantity.
To make it simple, I can return to my question from last year, when I asked you to compute the amount of CSI in the binary sequence 1100100100 0011111101 1010101000 1000100001 0110100011 0000100011… Your response was
Zachriel then revealed to you that the sequence was simply the binary form of the number π=3.1415926… Knowing that, one could immediately conclude that the sequence was quite specific.
Let me slightly rephrase the question by putting it in a more familiar context. In his paper, Dembski discusses the specificity of a password from the Da Vinci Code, where a character uses Fibonacci's sequence. His granddaughter does not know that beforehand, bust she correctly guesses the password using a hint. I could equally tell you that my string was one of the famous mathematical constants. Your answer then would be different.
Let's get back to functional CSI. How do you figure out which protein sequences are functional? You don't have an option of trying them all since that would amount to an exhaustive search of the entire space. The smallness of functionality islands is assumed, not demonstrated. That could well lead to a false positive.
Comment by olegt — January 19, 2009 @ 7:57 am
January 19th, 2009 at 8:53 am
I assume you mean Kolmogorov complexity. K(T) is the length of the string's shortest description in a given description language. That's fine. So how do we calculate P(T|H)? You almost seem to be saying that we use K(T) for both ϕS(T) and P(T|H).
That wasn't a calculation, but a qualitative explanation. I don't want a vague description. I want the precise methodology. Don't worry, I do read maths, and will ask if I have any questions.
Comment by Zachriel — January 19, 2009 @ 8:53 am
January 19th, 2009 at 1:36 pm
Zachriel:
Yes and no. Yes, the number of specified patterns S(T) was based on Kolmogrov complexity. In this case K(T) as the Kolmogrov Complexity is used to calculate the number of specified patterns since the pattern is specified as a result of it being algorithmically compressible. You can use any objective measure you wish to formulate an event as an independent pattern, and in this case K(T) was used. Just remember to calculate specificity of all patterns of the same probability using the same measure K(T), since you are merely finding how many patterns are specified according to the objective metric you've chosen (against the same non-ambiguous programming language) and then multiplying that by the probability of finding that pattern by brute force search P(T|H).
However, no, P(T|H) is not calculated using K(T). Since P(T|H) is the probability of finding that pattern by chance it is merely equal to the probability of finding that pattern by brute force search without taking the specified nature [in this case K(T)] of the pattern into consideration.
Zachriel:
I'm not sure what you're missing, but its all there.
specification? = – log2 [(M*N) * s(T) * P(T|H)]
specification? = – log2 [(probabilistic resources -- number of trials available multiplied by length of trials) * number of specified patterns of same probability as pattern in question * brute force search probability of pattern in question] > 1
I'm not sure which part of this you have a problem with. It is only subjective if the person using the equation is defining the specificity of a pattern based on ambiguous, subjective criteria. Garbage in, garbage out. The same can be said of any equation. Use it correctly and you get correct answers; use it incorrectly and all you get out is garbage.
K(T) is obviously objective as long as the same well defined objective programming language is used across the board, which is why K(T) is a measurement available in the scientists' toolkit.
Comment by CJYman — January 19, 2009 @ 1:36 pm
January 19th, 2009 at 2:08 pm
I don't want a vague description. I want the precise methodology. Don't worry, I do read maths, and will ask if I have any questions.
CJYman: brute force search probability of pattern in question
You do understand that "brute force search probability of pattern" is not a clear mathematical description. I assume you mean a uniform probability distribution. Is this correct?
Comment by Zachriel — January 19, 2009 @ 2:08 pm
January 19th, 2009 at 2:18 pm
I'm asking how your engineering approach starts to address this question, or, if that isn't the first question it asks, what questions does it ask first? You've rejected our efforts to find a starting point, so where do you start? How can you determine plausibility without knowing resources, abilities, or end result?
Comment by don provan — January 19, 2009 @ 2:18 pm
January 19th, 2009 at 2:18 pm
Edit isn't working. The blockquotes are off in the previous post. These are my comments.
You do understand that "brute force search probability of pattern" is not a clear mathematical description. I assume you mean a uniform probability distribution. Is this correct?
For the long binary pattern 01010101…. ϕS(T) is a low integer. And assuming a uniform probability distribution, P(T|H) is very small. Hence it's CSI. Maybe it represents coin flips, and someone cheated, design. But it might represent a orderly, but naturally-ocurring pattern. It still doesn't seem to work.
Comment by Zachriel — January 19, 2009 @ 2:18 pm
January 19th, 2009 at 2:22 pm
olegt:
First, no, I am not quoting Dembski. I am attempting to put the concepts in my own words — the best way to understand something.
Well, sure, a probability technically isn't a physical quantity. But what's your point?
And yes, CSI is a measurement of someone's knowledge of which patterns are specified and which one's aren't, however all measurement is based on someone's knowledge (especially as it pertains to probabilities) so again, what's your point? That is why I said that the pattern in question and the system which causes it to be specified should be well tested and understood. Again, garbage in, garbage out. Are you saying that it is theoretically impossible to discover which patterns are specified and which aren't or is it just hard sometimes to discover this? If it is possible then measuring for CSI is possible.
Your criticism says nothing of the effectiveness of CSI any more than it is a criticism of *all* measures of probability — if not all measurements, period. It merely shows that people can use the equation incorrectly by plugging in the wrong numbers based on insufficient data. So what, which measurements (equations) do not have this problem? Just make sure you are straight forward with your data which is used to discover how many patterns are specified.
The less accurate the input, the less accurate the output will be … of course.
olegt:
First, you didn't paste the part of my quote where I explained that a cryptanalyst working in conjunction with a mathematician may be able to discover that the pattern is indeed specified without anyone telling them. This is the beauty of calculating for CSI. If you can't discover a specified pattern, the equation will not result in CSI. This helps to rule out false positives while also getting rid of some patterns which may actually be the result of intelligence. But of course that is why it is a filter. It may not catch every sign of intelligence, but when it does catch something that pattern is indeed most likely a sign of previous intelligence.
olegt:
Uhuh, and once one used cryptographic methods to discover that, then one could continue to potentially measure for CSI. That's why we can't continue with the measurement based on lack of knowledge. I mean why would someone calculate for any probabilistic measurement with insufficient knowledge about the pattern? Why are you asking for the ID Proponent to do so?
So, why do you keep insisting that somehow the equation for CSI is supposed to have magical qualities which can discover if a pattern is specific? That is a job for the cryptanalyst. However, once the cryptanalyst has discovered a specified pattern, then he may be able to continue with a measurement for CSI.
olegt:
Excellent question for the biologists here. How would one go about quantifying the search space for proteins? I agree that it is possible that the smallness of functional islands may be assumed, however that assertion would have to be argued on its own merits, however this is not a weakness of the equation for measuring CSI which is what I am discussing here.
Can CSI be measured with sufficient information? Using Kolmogrov Complexity, I have shown that it can. Now, is there a relation between CSI and foresighted systems when law as a mathematical description of regularities or a pattern based on the physical properties of the material used is ruled out?
Comment by CJYman — January 19, 2009 @ 2:22 pm
January 19th, 2009 at 2:28 pm
Zachriel:
Maybe it will be more familiar to you as "exhaustive search." Either way, "brute force search" and "exhaustive search" are both used as terms in computer science. Yes, a uniform probability distribution where there is no search bias to a target.
If you wish to discuss the implications of bias to a target and how that relates to ID, then we could begin to discuss Conservation of Information.
Comment by CJYman — January 19, 2009 @ 2:28 pm
January 19th, 2009 at 2:58 pm
Resources: proteins, known bacteria ancestors/fossils, etc.
Abilities: these should be pretty standard. Properties of proteins are well-known. Evolutionary mechanisms are what they are.
End result: do I even need to answer this? And before you go off on me, asking about end results in evolution is rather self-evident.
Comment by JJS P.Eng. — January 19, 2009 @ 2:58 pm
January 19th, 2009 at 4:07 pm
For the long, repeating binary pattern 01010101…. With H being a uniform probability distribution, P(T|H) is very small. Hence it's CSI.
Comment by Zachriel — January 19, 2009 @ 4:07 pm
January 19th, 2009 at 4:57 pm
CJYman, when you mention your example on your blog, I'm not sure which one you're referring to, so I'll assume it's the one with the string of thirty 1's.
Here's a sincere question: Why did you use the value of 30 for M*N?
I see two things wrong with that:
(1) The number of flips in a single trial has nothing to do with the total number of trials.
(2) If we reduce your example to three coin flips, the CSI in "111" is -log2[3 * 2/8] = 0.4, which is greater than 0. So we should conclude design for "111", and likewise for "000". But the odds are 1:4 that chance will produce one of those two patterns, so we have a 1:4 chance of a false design inference. So much for "no false positives".
Comment by R0b — January 19, 2009 @ 4:57 pm
January 20th, 2009 at 12:25 am
Sorry, I completely flubbed (2) above. Here's a corrected version:
(2) If we reduce your example to five coin flips, the CSI in "11111" is -log2[5 * 2/32] = 1.7, which is greater than 1. So we should conclude design for "11111", and likewise for "00000". But the odds are 1:16 that chance will produce one of those two patterns, so we have a 1:16 chance of a false design inference.
Comment by R0b — January 20, 2009 @ 12:25 am
January 22nd, 2009 at 2:34 pm
JJ,
I think what I was missing is that you are proposing front loading as an engineering technique that EBE would evaluate in operation. Is that right? I thought you were proposing (or at least allowing for) evaluating front loading itself and how it was engineered, but all this head butting makes me think you don't consider the implementation of the front loading event open to study by EBE.
Comment by don provan — January 22, 2009 @ 2:34 pm
January 22nd, 2009 at 2:46 pm
I think EBE can be a framework to study various hypotheses, including FLE. It was never my intention to say that FLE is beyond evaluating. You do have legitmate points, but you also have to keep in mind that the FLE hypothesis is more in the investigation stage rather than theoretical testing stage, or in engineering terms, FLE is conceptual design, not detailed design.
Comment by JJS P.Eng. — January 22, 2009 @ 2:46 pm