What would help is understanding his methods vs nit picking the models he is testing.
Can you humor me a think a little out of the box here. @Mercer is claiming unless a model has phylogenetic analysis it is not useful. Is it possible phylogenetic analysis is simply the best we could do at a time of relatively poor understanding of molecular biology?
Could you not essentially always provide such a rationalization if you donât like the currently accepted model?
I think science is better served in this case with competing models given the current level of uncertainty.
Can you humor me and try not to misrepresent othersâ positions?
I am claiming that unless a model accounts for all of the extant data, it is not useful.
Thatâs not a controversial point for anyoneâexcept those who are desperately trying to ignore most of the evidence.
I could not agree with you more, which is generally what my theory attempting to do. Now, can you please tell this to John so he can focus on critiquing other predictive elements of my theory rather than focus solely on nested hierarchies. 
I admit that this is a tongue-in-cheek response.
Funny, I noticed the same thing coming from them every time I interact with them. Itâs all or nothing if thatâs not fully worked out.
What do you mean by âa predictive modelâ? What do you mean by âfully formed animalsâ? And of course you donât need to start from animal populations. Animals are opisthokonts, so you could start with an ancestral opisthokont population, and that isnât even the start.
Itâs only valuable if it leads to real discoveries. I suggest that it wonât. Thereâs too much evidence, ignored by Ewert, that shows common descent. Ewertâs graph doesnât count actual genes, just families, nor does it consider the sequences of genes and other parts of the genome, which fit that nested hierarchy.
If I have a fully formed population of chickens I can predict (through a simple mathematical model) how many eggs they will generate over the next month.
The discovery that Winstons model points to is how genes work together to form a unique functioning animal. What are the gene sets required to form an egg laying chicken? If I knock out the genes shared with zebra fish will a living, egg laying chicken form?
No, as you would if you knocked out the genes NOT shared with zebrafish.
Itâs utterly meaningless. Have you ever even read a knockout paper, Bill?
And Winston did not discover anything. Do you notice that you have to twist the meanings of well-defined words beyond all recognition?
Once again, @colewd is providing the useful service of illustrating the mentality of the person targeted by the propaganda created by the likes of Winston Ewart. The intent is to persuade Bill and others like him that ID makes legitimate scientific claims and, as we can see, the approach succeeds. Some creationists respond best to a more detailed simulacrum of science than that provided by Answers in Genesis.
What, again, do you mean by âfully formedâ? What if anything does this have to do with population genetics or any alternative? Or with the questions I asked?
Now, if I want to know how many eggs a population of chickens will generate next month, the best predictor is to observe chickens. But another predictor would be to observe related species, i.e. to use phylogeny as a guide. Most birds (probably all birds) lay one egg per day until the clutch is done. At least some will keep laying indefinitely if you remove each egg after itâs laid. So phylogeny would suggest one per day per hen working to complete a clutch. Dependency graphs seem completely useless in this regard. But we can take phylogeny even further. One egg per day seems universal in birds, but it also seems to apply to some non-avian theropods too. Oviraptor laid eggs in pairs, and this was likely one egg per day per ovary. While birds have only one functional ovary, the ancestral state would be two ovaries. So the state of one egg per day per ovary seems to predate birds by quite a bit. Isnât phylogeny handy? What your question does is show how useful phylogeny is and how useless Ewartâs dependency graph is.
No it doesnât. And you would have no idea how to go about trying to study that.
Silly question. Iâm willing to bet that some of the genes are required (i.e., no viable chicken without them) while others are important (chicken at a severe disadvantage without them) and still others are not important (chicken still works, though not quite as well). For the latter, consider the human case of red/green color blindness. Anyway, this has nothing at all to do with Ewartâs diagram, which is not useful in finding out any of that or even in deciding where to look.
The nested hierarchy is one of the most basic observations in biology. If your model canât explain why we see one of the most basic features of biology then your model has a serious problem.
Would this be the same methodological naturalism that you use to explain nearly all of nature? Do you also say that germs causing infectious diseases is the best explanation we can come up with under methodological naturalism?
It seems that you want to throw out logic and reason for no other reason than them leading you to conclusions you donât like.
That discovery was made decades ago when we learned how DNA, RNA, and proteins worked.
Also, you are now using methodological naturalism to explain how biology works. Whatâs up with that?
I showed you a simple toy predictive model. Population genetic models are slightly more complex but built with the same principles of observation and mathematics.
Exactly right. While phylogenetics may be interesting to show how related animals lay eggs the best predictor is to observe the organism you are counting on for your eggs.
I agree the dependency graph is not the right tool for the job here. While phylogenetic analysis is interesting it is not likely to help with building a better model.
Where Ewerts dependency graph is interesting is in understanding the genes that cause this egg laying that is mathematically predictable. If we have another egg laying animal what would its dependency graph look like?
Maybe, but it puts into question that the pattern observed is the result of random (non directed) genetic change
That would be methodological naturalism, wouldnât it? In methodological supernaturalism you start with the belief that a trickster god is only making you see chickens lay eggs when in fact they are being deposited by invisible egg fairies. The egg fairy model is far superior to the scientific model because it better explains why we see eggs.
You showed me irrelevant trivialities. You have no clue about population genetics.
How is this a point relevant to anything weâve been talking about? Why are you even bringing chickens into the discussion?
Word salad. No response is possible.
Once more you are confusing the pattern of change with the causes of change. Can you at least acknowledge that thereâs a difference and that weâre talking about the former, not the latter?
I have shown a model where the output is predictable (eggs) vs a model where the output is probabilistic (allele frequencies). I am simply trying to answer your question of what I mean by âpredictive modelâ.
Chickens were simply used as an example of a predictive model of existing populations of animals.
Inside Ewerts dependency graph are the genes and gene families that allow predictable egg laying to occur.
How do we eliminate âcauseâ of the pattern from the discussion and still explain the pattern?
You havenât shown a model at all. You just mentioned chickens and eggs. You predicted nothing.
But the graph has nothing to do with egg-laying and doesnât suggest or lead to any sort of model of predictable egg laying, which you havenât produced either.
We donât. You really are unable to distinguish between the cause of the pattern, i.e. the tree, and the causes of the individual elements of the pattern, i.e. the changes. Iâve tried to explain this to you possibly as many as a hundred times over the course of several years, and you still canât get it. Does Ewartâs diagram explain why you canât?
You understand how to form a egg generating model given a population of chickens so letâs move past this one.
This is all true but not the claim I was making.
Ewertâs graph shows a better fit (dependency graph) than a tree for the gene families he selected. As far as I can tell, the cause of the pattern is based on a starting point of animal populations (kinds) as theologically described in Genesis 1. Many of the animals are designed with a unique gene set.
Simply saying that common descent only explains the tree pattern is fine but it is a limited explanation of the gene pattern. It does not eliminate the theological claim that animals were created by God according to their kind.