Wednesday, September 09, 2015

Robert Marks II: Still Refusing to Reply One Year Later

Wow, has it been a year already? A year ago, I wrote to the illustrious Robert Marks II, asking him about a claim he made: "we all agree that a picture of Mount Rushmore with the busts of four US Presidents contains more information than a picture of Mount Fuji".

No reply.

No reply after three months.

Or six months. (Just an auto-reply.)

Isn't this just typical of creationists? Make wild claims and refuse to back them up when challenged.


Mikkel Rumraket Rasmussen said...

Hey Jeffrey, you should re-post your information theory challenge to creationists some time, with an update that show there still has not been any creationist that passed it.

Test your knowledge of Information theory.

Even at a intuitive level the creationist claims about information seem, for lack of a better term, obviously wrong. To say that mutations can't create information is to say that pressing buttons on a keyboard to write a sentence will not produce an informative one (it would simply be analogous to insertion mutations, adding "letters" one at a time). I honestly don't understand how a thinking human being can even get themself to say something so trivially and demonstrably wrong.

lukebarnes said...

Why not throw him a lifeline? There is *some* true claim hovering the vicinity of that sentence. An alien arriving on Earth could learn something about how natural mountains form on this planet by examining either mountain. But with Mount Rushmore, they could *also* learn about the intelligent beings that occupy Earth - what they look like, that they have the technology, intelligent and motivation to carve their faces into a mountain, that the particular individuals on the mountain are probably important for some reason. This is what most laypersons would understand from Marks's statement. Implicitly, more information = more things that intelligent beings could learn. There is a bayesian theory-testing sense in which Mount Rushmore is evidence for propositions about Earth that Mount Fuji isn't.

Having identified the sense of "information" that Marks's claim presupposes, and then show that it is some combination of:
* not able to be made mathematically precise,
* not reflected in Kolmogorov's theory,
* not relevant to information in biological systems, and / or
* not a reliable indicator of intelligence.

And then we'd all learn something, even if Marks doesn't. (Obviously, your request for an explanation from Marks is completely reasonable and there are no excuses for his silence.)

Jeffrey Shallit said...

This is what most laypersons would understand from Marks's statement.

Marks does information theory, so I doubt very much he was speaking of information in the informal sense you describe. And I see no evidence at all he was talking about aliens.

Remember that he was talking about pictures. Luke, are you really going to claim that every possible picture of Mount Rushmore contains more information than every possible picture of Mount Fuji? What if the picture of Mount Rushmore has resolution 50 x 50 x 8 bits, and the picture of Mount Fuji is 1024 x 780 x 32 bits? What if the picture of Mount Fuji shows people climbing it? I don't think you've thought this through at all.

Gordon Davisson said...

I'm not sure about the pictures, but it occurs to me that the shape of Mt. Fuji itself must have quite a lot of "active information", since an evolutionary algorithm (or any other reasonable optimization algorithm) would find its way to the mountain's summit much faster than a random search would. The reason is simple: the landscape's altitude functions as what you might call (paraphrasing Dembski) a Euclid oracle, giving the distance to the summit.

Clearly, Mt. Fuji must be intelligently designed. Either that, or a product of an intelligently designed search; I suppose you could consider volcanism + erosion to be a weird sort of search algorithm, so maybe they're intelligently designed.