Before looking up all of these facts, it would’ve been interesting to attempt to Fermi estimate them; that could’ve helped you uncover more stuff that’s mistaken in your model of reality.
Isn’t that like trying to estimate unknown unknowns? If one has already reached the point of thinking, ’wait how many major plane crashes have there been in the USA in the past decade, exactly? Why… I can’t seem to think of any!″, one has already done most of the perspective changing.
What’s interesting about the Fermi Estimate post is that its examples encourage you look for predictors that are unexpectedly reliable, rather than those that first jump to mind.
That I haven’t heard of many plane crashes in the past decade, this sounds like something I might hear on a post arguing an opposite point. “Sure, you’ve haven’t heard of plane crashes in a decade, but why suspect that reliable predictors are in the neighborhood of your daily activities rather than your knowledge about the world? And now I will eye you knowingly until you learn something of your cognitive biases!”
Although, perhaps there aren’t any estimated-statistics approaches that would be any good here, that wouldn’t rely on other more incidental bits of information you happen to possess. Sure, you could try to list the top causes of flight disasters (human error and mechanical malfunction?) and estimate the likelihood these things occur, and also estimate how many of these flights result in large scale deaths. But there may be too many variables; either that, or I have ways to go in making Fermi estimates. In any case, it would be hard to incorporate the time variation of flight disaster outcomes. For all I know, flight safety has skyrocketed in the past decade due to widespread process improvement resulting from studying past disasters. And how could I ever predict that effectiveness, or predict how long it would take to come about?
That I haven’t heard of many plane crashes in the past decade, this sounds like something I might hear on a post arguing an opposite point. “Sure, you’ve haven’t heard of plane crashes in a decade, but why suspect that reliable predictors are in the neighborhood of your daily activities rather than your knowledge about the world? And now I will eye you knowingly until you learn something of your cognitive biases!”
Please see my sibling comment on why the availability heuristic delivers in spades about jumbo jet crashes.
If every car crash received headline coverage due to hundreds of people dying simultaneously horribly, followed by sporadic articles as the investigation of that car crash went on and profiles of people who died in the crash, then not being able to think of any recent car crashes probably means that there weren’t any, not that the entire media industry collectively decided to stop covering them.
So, you don’t commute much? I can think of a few specific ones I’ve driven past in the last few months, and I know that I’m not even remembering them all. Plus hearing traffic reports, etc.
Of course it’s more reliable to look them up, but the point of doing a Fermi estimate here is to figure out what else you might be mistaken about. For example, if you do a Fermi estimate about how much a rocket costs and it turns out to be really inaccurate, you might learn that that’s because you have a really inaccurate picture of how much engineers get paid, or how many engineers it takes to build a rocket, or how much the materials for something like a rocket cost, or…
Anna and I actually did a Fermi estimate of the fuel for the Apollo 11 mission over dinner last week, and we were off by a factor of two. Some of the available inputs:
A crude estimate of the potential energy of mass lifted from Earth’s surface to a distance of many times its radius
The heights reached by jet aircraft using fuel amounting to only a very small portion of their mass
A crude estimate of the energy content of gasoline (one approximation is to energy content of food, and/or the energy output of humans), with adjustment for the need to carry oxygen into space
Images of rockets launching, which show that the fuel tanks are much bigger than payload, but not thousands or millions of times bigger
Knowledge of the price of consumer gasoline, or the price of oil
The existence of science fiction writers with physics backgrounds, SpaceX, the L-5 societies, and other groups seriously pushing for advancements to slash cost-to-orbit
Rough knowledge of NASA’s budget, either directly or by bounding it relative to known US budget items
Knowledge of the enormous cost of producing military aircraft and naval vessels, which can be in the hundreds of millions to billions of dollars
The existence of an ecology of NASA contractors condemned for their enormous costs (these would be trivial if fuel was the major cost)
But those aren’t the main reasons for the costs. The main reasons are that rockets have not benefited from reduced costs through mass production—something that would be very hard to estimate if you didn’t know or guess that.
Right, and attempting to do a Fermi estimate and then checking it would help you see that that was something you didn’t know. (Maybe I’m not getting my point across well here. My point wasn’t that such a Fermi estimate would have been in any way accurate. My point is that identifying what made it inaccurate would have helped you update more.)
Before looking up all of these facts, it would’ve been interesting to attempt to Fermi estimate them; that could’ve helped you uncover more stuff that’s mistaken in your model of reality.
Isn’t that like trying to estimate unknown unknowns? If one has already reached the point of thinking, ’wait how many major plane crashes have there been in the USA in the past decade, exactly? Why… I can’t seem to think of any!″, one has already done most of the perspective changing.
What’s interesting about the Fermi Estimate post is that its examples encourage you look for predictors that are unexpectedly reliable, rather than those that first jump to mind.
That I haven’t heard of many plane crashes in the past decade, this sounds like something I might hear on a post arguing an opposite point. “Sure, you’ve haven’t heard of plane crashes in a decade, but why suspect that reliable predictors are in the neighborhood of your daily activities rather than your knowledge about the world? And now I will eye you knowingly until you learn something of your cognitive biases!”
Although, perhaps there aren’t any estimated-statistics approaches that would be any good here, that wouldn’t rely on other more incidental bits of information you happen to possess. Sure, you could try to list the top causes of flight disasters (human error and mechanical malfunction?) and estimate the likelihood these things occur, and also estimate how many of these flights result in large scale deaths. But there may be too many variables; either that, or I have ways to go in making Fermi estimates. In any case, it would be hard to incorporate the time variation of flight disaster outcomes. For all I know, flight safety has skyrocketed in the past decade due to widespread process improvement resulting from studying past disasters. And how could I ever predict that effectiveness, or predict how long it would take to come about?
Please see my sibling comment on why the availability heuristic delivers in spades about jumbo jet crashes.
I can’t think of many car crashes either. That just means that I don’t hear about them.
If every car crash received headline coverage due to hundreds of people dying simultaneously horribly, followed by sporadic articles as the investigation of that car crash went on and profiles of people who died in the crash, then not being able to think of any recent car crashes probably means that there weren’t any, not that the entire media industry collectively decided to stop covering them.
So, you don’t commute much? I can think of a few specific ones I’ve driven past in the last few months, and I know that I’m not even remembering them all. Plus hearing traffic reports, etc.
Sometimes they’re easier to look up, and it’s almost always more reliable.
Of course it’s more reliable to look them up, but the point of doing a Fermi estimate here is to figure out what else you might be mistaken about. For example, if you do a Fermi estimate about how much a rocket costs and it turns out to be really inaccurate, you might learn that that’s because you have a really inaccurate picture of how much engineers get paid, or how many engineers it takes to build a rocket, or how much the materials for something like a rocket cost, or…
I don’t see how I could have Fermi-estimated the relative costs of the rocket and the fuel...
Anna and I actually did a Fermi estimate of the fuel for the Apollo 11 mission over dinner last week, and we were off by a factor of two. Some of the available inputs:
A crude estimate of the potential energy of mass lifted from Earth’s surface to a distance of many times its radius
The heights reached by jet aircraft using fuel amounting to only a very small portion of their mass
A crude estimate of the energy content of gasoline (one approximation is to energy content of food, and/or the energy output of humans), with adjustment for the need to carry oxygen into space
Images of rockets launching, which show that the fuel tanks are much bigger than payload, but not thousands or millions of times bigger
Knowledge of the price of consumer gasoline, or the price of oil
The existence of science fiction writers with physics backgrounds, SpaceX, the L-5 societies, and other groups seriously pushing for advancements to slash cost-to-orbit
Rough knowledge of NASA’s budget, either directly or by bounding it relative to known US budget items
Knowledge of the enormous cost of producing military aircraft and naval vessels, which can be in the hundreds of millions to billions of dollars
The existence of an ecology of NASA contractors condemned for their enormous costs (these would be trivial if fuel was the major cost)
Estimate how many people you would have needed to employ and what their salaries might be, estimate the cost of materials…
But those aren’t the main reasons for the costs. The main reasons are that rockets have not benefited from reduced costs through mass production—something that would be very hard to estimate if you didn’t know or guess that.
Right, and attempting to do a Fermi estimate and then checking it would help you see that that was something you didn’t know. (Maybe I’m not getting my point across well here. My point wasn’t that such a Fermi estimate would have been in any way accurate. My point is that identifying what made it inaccurate would have helped you update more.)
Ah, I see. I’ll try and bear that in mind next time I go looking for info like that!