Re: Making your mind up

On 19/04/2024 03:36, Mark Isaak wrote:

On 4/7/24 8:01 AM, Martin Harran wrote:

snip

One exception to that is your suggestion of a
random number generator when the two options look more or less equal
but your problem is that that randomness is very antithesis of
determinism.

 I don't think that's true. A process can be both random and determined. But that hinges on definitions of random, and is outside my area of competence.
 

I don't know if I ever talked about this on this forum but I had an epiphany on definitions of "random" when looking at regression plots ages ago. Which is that "random" isn't a word that applies to a thing, it applies to *the relationship between two things*. So if you plot the temperature of a place over 100 years against the days in the year you'll get an up-and-down trend showing summer and winter, and if you control for that trend you'll be left with a cloud of "random noise" that represents the year-to-year variation in temperature. On the other hand if you plot the same temperature points against the year you'll get a trendline representing the year-to-year variation (maybe a rising trend for example), and if you control for it you'll be left with random noise that represents the within-year variation. The same thing can be "random noise" or "trend" just depending on your choice of x-axis! Because "randomness" describes the correlation between two variables (or more specifically, the lack thereof).
I think that extends to almost all uses of the word "random", there is almost always a "with respect to..." hidden in there that clarifies what variables it is one is claiming are uncorrelated. For example how is a coin toss random even though it's deterministic? Well, it's random *with respect to* any guesses the thrower and observers can make as to what the outcome of the throw will be. It's not random with respect to precise position and velocity of the coin a millisecond before it lands, but that's not what it's being asked to be random with respect to.
Or in the context of evolution, the "random" in "random mutation" means "random with respect to whether the mutation is beneficial, harmful or neutral for the organism".
In the context of Martin Harran's comment on using a random number generator to make the decision I'd say that the randomness in question is with respect to all of the pro-and-con factors that otherwise would go into making the decision. For example if we're debating whether to go on vacation at the beach or in the mountains and we can't decide, I could say "OK let's just pick the one that's closest" and we wouldn't think of that as "random"; it's choosing one pro-or-con factor to prioritize above others. Same with "We'll go with what you prefer". On the other hand if I say "some neutral third party will hide a shoe in the house, first to find it gets to decide" that *could* be random even though hiding and finding the shoe aren't what we'd think of as random processes, because who finds the shoe (and therefore, what decision gets made) is presumably uncorrelated to any of the reasons that might otherwise have contributed to the decision. If it's not uncorrelated (because you're better at finding things, or I bribed the third party to tell me where they hid it) then we no longer think of the decision as having been "random".
 From that point of view the "random" decision is indeed completely compatible with determinism, the same way a random coin toss is.