Jim Manzi has what looks to be an interesting new book out. David Brooks describes the premises.
4 thoughts on “The Value Of Trial And Error”
Criminal trolls from the UN are taking the lead?
Anyway, you’d think the premise would be pretty obvious. Science used to be contemplation followed by measurement. Now it’s what makes you feel good at a party without any need to measure (because you can always produce phony data to support your point.)
Social science; when real science isn’t good enough.
“Social science; when real science isn’t good enough.”
Well, there are some pretty good reasons to avoid -strict- real science rules of causality vs correlation, true random sampling in advance, control groups, double-blinds etc.
“Hey Ken, you’re randomly selected for a trial. We can’t even tell you what the trial is about to avoid screwing things. In fact, let me knock again and start over. -Insert Goons-.”
They’re generally only used on enemy prisoners and are called war crimes.
Of course Al.
I’ve got a beef with real science as well. The problem with both is humans. We aren’t really designed well for doing science. The biggest problem being we’re really good at discovering patterns that often have no relation to what actually is. Asimov made the point that it’s not that science knows the truth; it’s that it gets closer to the truth as time passes. Which I agree with except it doesn’t acknowledge all the veering away from truth that happens (and is often supported for generations.)
If you look at truth as a binary question; what we all know with absolute certainty is not true. Asimov said we shouldn’t do that. I say it’s useful to keep us humble.
Trial-and-Error?
It’s how most of us used to learn about life. That was before bicycle helmets, non-competitive sports, schools without ‘letter grades’ and helicopter parents of course.
Criminal trolls from the UN are taking the lead?
Anyway, you’d think the premise would be pretty obvious. Science used to be contemplation followed by measurement. Now it’s what makes you feel good at a party without any need to measure (because you can always produce phony data to support your point.)
Social science; when real science isn’t good enough.
“Social science; when real science isn’t good enough.”
Well, there are some pretty good reasons to avoid -strict- real science rules of causality vs correlation, true random sampling in advance, control groups, double-blinds etc.
“Hey Ken, you’re randomly selected for a trial. We can’t even tell you what the trial is about to avoid screwing things. In fact, let me knock again and start over. -Insert Goons-.”
They’re generally only used on enemy prisoners and are called war crimes.
Of course Al.
I’ve got a beef with real science as well. The problem with both is humans. We aren’t really designed well for doing science. The biggest problem being we’re really good at discovering patterns that often have no relation to what actually is. Asimov made the point that it’s not that science knows the truth; it’s that it gets closer to the truth as time passes. Which I agree with except it doesn’t acknowledge all the veering away from truth that happens (and is often supported for generations.)
If you look at truth as a binary question; what we all know with absolute certainty is not true. Asimov said we shouldn’t do that. I say it’s useful to keep us humble.
Trial-and-Error?
It’s how most of us used to learn about life. That was before bicycle helmets, non-competitive sports, schools without ‘letter grades’ and helicopter parents of course.