Evolutionary Algorithms and ID

I’m denying the form was pre-specified anywhere. “Here is an insect leg which superficially looks like a human designed gear therefore the insect leg was designed too” is about as dumb and fallacious an argument as they come.

No serious scientist would ever deny that the structure found in the planthopper is a true mechanical gears. It is really astonishing to see how far from reason some people are ready to go in order to avoid an inconvenient truth. Yes, truly astonishing !

Because the 800-pound gorilla in the room is being ignored. The ID specification is not independent.

3 Likes

Not independent of what?

See here: Evolutionary Algorithms and ID - #131 by Dan_Eastwood.

so we cant conclude design just by looking at gears? ok. what about something more complex like a pc?

131 is an assertion.

Specification is based on the function not the sequence. By most all observation we have multiple sequences that can perform the function. The issue is how many can vs how many sequences are possible. In the case of beta lactamase I ran alignment on 5 species of the same protein version and only 6% of the amino acids were aligned.

If it worked as science, you would use it to make emprical predictions. You don’t.

1 Like

That leaves almost infinite room for cheating.

A lot more than you’ll admit.

You don’t seem to want to know the former.

That’s an absurd way to estimate. In the case of beta-lactamase, catalytic antibodies can be obtained from only 10^8 clones. Unlike yours, that’s a real number.

That’s an underestimate given the structural constraints of antibody structure.

2 Likes

Weak sequence preservation appears to correlate with your number.

No, it has nothing to do with it. Sequence preservation is a horrible way to estimate the target space when you could be reading >5000 different catalytic antibody papers.

You don’t want to know, though, do you?

I have read a paper you cited. Gpuccio is coming up with an op that will cover catalytic antibodies.

1 of >5000?

Great news. I am looking forward to read it!

Sorry. There just seemed like there was debate concerning the topic of Evolutionary Algorithms where ID folks were saying they designed and the non-ID folks were saying they weren’t (at least not directly). I am trying to understand the distance between the groups. I figured one way might be to take some examples that ID folks would agree are not examples of design and work from there. Maybe it’s horrible idea, who knows.

My only real exposure to Evolutionary Algorithms are genetic algorithms use in ultrafast (femtosecond) laser spectroscopy to create laser pulses “shaped” to interact with a particular bond. The algorithms were used because people couldn’t predict what the best shape would be (it doesn’t have a functional form that we know of). Presumably a person (a grad student no doubt) could sit there and test each possibility out via trial and error but the EAs were used to find solutions much much quicker. In these cases, the system was set up, but the solution (the sequence, so to speak) was not designed.

Can you think of anything on the biochemical scale? I’m imagining that if there is a spectrum between “not designed” and “designed” there would be things that would be hard to tell. Can you think of any examples or is that always clear (using FI or some other measure of “design”)?

OK thanks, I just wondered.

2 Likes

But @Giltil, it doesn’t look to me like @Timothy_Horton was denying it was acting as a mechanical gear (interlocking teeth on a rotating structure) but was instead saying that the claim of a design inference was based on superficial similarity, at best.

This is also the problem that I have with the gear analogy. The bit that makes a gear a gear is very simple. Interlocking teeth and rotation are not why I would conclude a gear was human designed. It’s things like the perfect circle, homogenous material, English lettering, that lead me to conclude it’s human designed. In other words, it’s all the incidental properties and not the function that leads me to suspect design, and those properties are exactly the ones that aren’t shared with biological gears.

8 Likes

Well, i didn’t say we can’t conclude, I just said I didn’t conclude anything. I generally don’t go along my day pondering whether everything I see is “designed” or not. I generally don’t care one way or another.

But then I did answer your question, I thought:

So I would say that a person could probably have a good chance to spot a human-designed gear, but I don’t think that means that all gears are designed. See my previous response to @Giltil for more details.

As to the PC, yes, I don’t think it’s a surprise to say that a PC is designed.

3 Likes

What you are saying is quite strange. Nobody is arguing that the planthopper gear was human designed! Only that in all probability it has been designed by a mind, period. So what is the point of insisting to find in this insect gear all the incidental properties you attribute to human gear?
Talking of gear in the planthopper case is not an analogy; by all definitions, the bug gear is a true mechanical gear. It is a gear in term of function and it is a gear in term of structure/organization.
Below is another image of the bug gear to awaken our sense of wonder.
image https://www.insidescience.org/sites/default/files/insect-gears-top.jpg

Why? Explain how you make that inference.

Obviously you must draw some analogy to human designs. Humans are known to have designed gears, so if we find them in an object not designed by humans, we should think they are designed [even if not by humans] because… why?

Nobody is denying they are gears. You are wasting a lot of time arguing they are gears, and nobody has said they’re not. They’re clearly gears, and if you can dig up a quote by someone saying they’re not then I’ll join you in dismissing this person’s opinions as blatant denial.

What I want from you is the inference that shows design is the best explanation. Preferrably by including a phylogenetic analysis of related planthopper insects and the structures they have in place of these gears. If you’re going to make an inference to the best explanation, you need to compare the ability of different explanations to account for the data.

Have you, or anyone, does such an analysis in this case?

2 Likes