[leto_ii]

I fully agree with your points. I think I may have been sloppy in expressing my criticism of the original article. At a certain level I think my comment was 'sociological' - the aliens, as intelligent creatures, should have a strong intuition that humans as other (somewhat) intelligent creatures can't be spending so much time and attention on something that is nothing more than a coin toss.

This sociological intuition should drive further inquiry into the mechanics of the game.

> if they are limited to that extent, then they are not going to determine that it is a game of rules, no matter what their philosophy and no matter how much they suspect that it is

Indeed, in that case the aliens are forever stuck. However, IRL you probably can't be absolutely sure that you're forever stuck, so in this case the 'philosophical' attitude might matter. An anti-realist might say - to hell with it, no worth trying, we'll never get better predictions out of more complex theories. A realist however, might pursue a theory not because it makes more accurate predictions, but because he/she has an idea that the theory is truly closer to the truth than the idea of a random coin toss. This intuition might take you through a dark period towards a higher reward (see moving from a local maximum to a valley, towards a yet unforeseeable global maximum).

PS:

flubert has a nice excerpt from Jaynes' Probability Theory: https://news.ycombinator.com/item?id=25648965

[mannykannot]

Thanks for bringing that to my attention - I have posted a reply there.

I think you are probably on to something when you suggest this article was influenced by Feynmann's analogy, but I am pretty sure that Feynmann was not suggesting the straw man that this article attacks.