Yes, the 2% of the desktop/laptop world is the "only" relevant question here./sarcasm
Depends on what you mean by "linux". If you count Chromebooks, nobody is quite sure how many are being sold, but it's clearly not so insignificant.
But of course by "linux" we usually don't mean the linux kernel surrounded by a proprietary userland and tied to proprietary network services; we mean a proper distro. But I know *I* have installed Linux on machines that are only used for browsing just to avoid the annoyance of Windows.
You may be right, but if you've ever looked at scientific papers in nutrition most of the studies are modest, and there's a lot of arguing over what looks like not enough data. More nutrition research could wipe out a lot of common-wisdom extrapolations from limited data and replace them with solid, evidence-based results.
The US spends about 1.6 billion in nutrition research annually, which sounds like a lot, but it's quite evident that a really good nutrition study (large enough sample size, long enough duration, adequate experimental controls) is extraordinarily expensive. And to put 1.6 billion in perspective, we spend 3.2 trillion on health care annually, a lot of it to treat conditions (diabetes, obesity, hypertension) that we know are connected to nutrition but in ways we don't fully understand. I don't think there's anyone who doesn't believe we could achieve significant improvements in longevity and quality of life with better nutrition, but catch is we don't know how.
We're funding nutrition like "small science", when it should by any reasonable measure be big, "moon shot" level science.
It's surprising what scientists learn when they actually look, which is why it's important to fund research.
The gold standard in calorie consumption measurement is something called "doubly labelled water" -- basically a scheme for measuring energy use using water tagged with uncommon isotopes of hydrogen and oxygen. The incorporation of these isotopes into metabolic by-products provides an indirect means for measuring metabolic rate in real world conditions.
So researchers did a doubly labeled water study on the Hadza of Tanzania, one of the few remaining hunter-gather societies in the world, and discovered that there was no significant correlation between activity level (measured by GPS) and calorie consumption. While they did exhibit more energy usage than sedentary westerners, it was only on the order of 200 calories/day -- which admittedly over a year is a lot of calories.
Now it's a mistake to extrapolate from the Hadza to people living in industrialized economies. The Hadza are smaller, leaner, much more active and eat a much different diet which varies in calorie content every day. But the most important thing to take away is the unreliability of the naive "scientific" model of the human body as being like, to a first approximation, an insulated calorimeter that can only shed energy by exercise.
Or it could be a non-representative sample. Or the difference in rates might not be statistically significant. Or the sample used might be too heterogeneous in terms of content or subject to precisely compare rates. Or women could submit code more frequently and have the same acceptance rate. Or things might look different if you control for submission size.
It's nearly impossible to tell what's going on with a single aggregate figure like this without access to the underlying data, if not the code in question. Anyone can construct any scenario they want because there is not enough information to draw any conclusions.
A robot is a machine that can be programmed to perform a variety of complex sequences of actions (e.g. an industrial robot in a car factory). This is in contrast to a machine performs a complex action for which it is mechanically specialized (e.g. a bottling machine at a brewery).
Naturally there is no perfectly sharp dividing line between the two. For example an industrial robot may have specialized attachments which allow it to weld, or to inspect welds for that matter. A bottling machine may be controlled by an industrial computer and might have other modules of a general purpose nature. They are in many ways more alike than different.
Which means there's no sharp dividing line between robots and computerized machines in general. Which is true of a lot of things: child and adult, night and day, etc. The law necessarily draws sharp lines between fuzzy sets. You're an adult at eighteen, but it could just as well have been 17 years 11 months or 18 years 1 month; 18 is just a conveniently round number.
Yes, I know what carbon intensity is, but the linked article confuses the issue and summary confuses it more.
Emissions intensity is emissions relative to the volume of an activity, so I was trying to point out that if you measure an activity relative to energy consumption, you would not expect much change in the transport sector because people are for the most part powering their vehicles the way they did in 1975. For practical purposes almost nobody is running their car on hydrogen, so the carbon intensity per unit of energy used is for practical purposes determined by the chemistry of alkane combustion.
If people were switching over from coal-powered cars, it would be a different story. Coal has a lower energy density by weight and by volume than gasoline. It has a lower energy density by weight than natural gas too (although higher by volume). The reason is that in coal the carbon atoms are bound up in large sheets of polycyclic rings. The net energy you get from oxidizing the carbon in coal is less per carbon atom because you have to account for the cost of breaking up those aromatic rings.
I suspect we're suffering from a misunderstanding of terminology here.
Thermodynamic efficiency has nothing to do with carbon emitted by an engine per unit of fuel; that's determined by stoichiometry. Basically every carbon atom that goes into your tank comes out the tailpipe as soot, CO2, and a very smalll quantity (these days) of unburned alkanes.
Since that means that the CO2 emitted per unit consumed by the engine is approximately constant, I suspect they're talking about the entire supply chain including extraction, refinement, and transport.
Suppose the supply chain consumes a gallon of oil to get one gallon of gasoline from the ground to your car tank. Then every gallon of gasoline you burn in the car will represent 40 pounds of CO2: 20 coming out the tailpipe, 20 emitted somewhere else before the gas even got to your car.
For centuries the intellectual basis for conservatism has been set, not by Jesus, or Adam Smith, but by Edmund Burke, whose philosophy could be summed up this way: if it ain't broke, don't fix it.
Burke was the kind of man who could defend the monarchy while despising monarchists: he thought the notion that monarchy was an ideal form of government was fatuous twaddle. But he thought all grand, all-encompassing theories were foolish, so he wasn't any more enthusiastic about pure democracy. Burke preferred a monarchy restrained by a democratically elected parliament not because it was the best system, but because it worked, experience showed that men could be tolerably free and prosperous under such a system.
So the notion that we need to "fix" an innovative segment of the economy to be more like what our theory of what an innovative industry should look like is about as un-conservative as you can get. It is, in fact, radicalism of the sort Burke detested.
Back in my day we had the pill and there were no STDs that couldn't be cured with a week or two of antibiotics, with predictable results. So yes, in some ways things were better.
It doesn't mean everything was better. Python is better than rolling my own control structures with Fortran IV's computed goto.
When you get older you'll realize "you win some, you lose some" is pretty much what life amounts to.
... the gods themselves contend in vain, Schiller said.
But maybe not. I just finished Hanah Arendt's famous Eichmann in Jerusalem: A Report on the Banality of Evil, about the abduction and trial of the former SS officer who was in charge of "evacuating" Jews to the death camps. Eichmann claimed -- probably truthfully -- to be horrified and distressed when he saw what was happening in the extermination camps. But his horror was greatly mollified at the a conference in the Berlin suburb of Wannsee in which many important and respectable people discussed the Final Solution frankly and unabashedly, as if it were no big deal.
Arendt also points out something interesting about Denmark, a country which was under total military domination by the Third Reich but in which society from the King down resisted the expulsion of Jews. Not only were the Germans unable to expel even stateless Jews from Denmark, confirmed SS officers posted to Denmark would suddenly become unreliable on the Jewish Question.
This suggests to me that when you feel like you're powerless against stupid or even evil ideas, there is always something you can do that can be very powerful: you can set an example.
"Transforming" the government is something a bunch of Silicon Valley executives could no doubt accomplish. The problem is that "transformed" isn't necessary "better".
Facebook has transformed peoples' lives, to the point it's regarded indispensable for many. It hasn't really made their lives better.
"Transformation" is a goal without a point of view; and blank screen on which anyone is free to imagine whatever colors he likes. Not like "cheaper". Cheaper is quite concrete. "More responsive" would be harder to measure than cheaper, but at least it represents something.
And in fact, it *does* power the planet. It's only our narrow focus on things that are literally "accounted for" in our economy that makes it appear otherwise. If we had to replace all the things that sunlight does for us with our non-renewable energy sources they wouldn't even come close, but that all happens off the books so it's invisible.
Now I worked for environmental organizations in the years of transition from crying indian environmentalism to "sustainability" based environmentalism, and I always had a problem with the new framing: non-sustainability is by definition a self-correcting problem.
So if we survive as a society, that society will eventually be powered by (and limited by) solar energy. The problem isn't non-sustainability per se, but the predictable costs of running unprepared into the limits of the processes we depend on.
You might be the most powerful man on Earth, but your problems are proportionately bigger.
Any reasonably thoughtful person would look at the presidency, and conclude that being president means feeling frustrated by not having enough power most of the time. And it doesn't mean getting respect -- even respect people think is due to the office, not the occupant. Trump of all people should have realized that.
You can't bomb ISIS away, or shock-and-awe Syria into submission, or intimidate North Korea. Everything you might try to accomplish is accompanied by a perplexing calculus of risk/reward. Manned space exploration is no exception. If you want to do it, you need to make sacrifices.
Sure, if Trump's ego gets us to Mars, we should all be for it, but then a falsehood implies anything. I could just as truly say that if Trump's ego gets us to Mars, the crew vehicle will be powered by unicorn farts.
Well, to make things as simplistic as possible, Mars at its closest is over at thousand times farther away.
Now I don't have much of a throwing arm -- you might say I'm starting with nothing -- but I'm pretty certain if you gave me a few years I could manage to throw a football twenty yards. It doesn't follow that given a few more years I could somehow throw a football twenty kilometers.
PDFs are inconvenient in e-readers because PDFs are page-oriented. That makes them inconvenient on smaller-screen devices. That said, PDFs represent that page with a high degree of fidelity, which is PDF's biggest strength.
The problem with math ebooks published in AZW format is that they either render equations incorrectly (making them useless), or render them as bitmaps (making them less useful than they should be, and sometimes illegible).
In most cases I'd take an EPUB or AZW over a PDF for reading on a small device, but for math I'd take the PDF, despite its inconveniences. Or better yet, a physical book.
I'm in my 50s and my house is literally full of physical books. Every room is lined with bookcases most of them stacked two deep, and I've literally had to put jackposts in my basement to keep the floors from sagging.
Buying new books as ebooks means I don't have to get rid of my old books. It's also nice being able to travel with a generous selection of reading material.
Overall I find the reading experience to be about a wash, but that's a highly personal thing. For pure reading a physical book is better except in low-light conditions, but the search and note taking functions on an ebook are a big plus.
The biggest drawback for ebooks for me is the terrible mathematics typesetting, which is obviously a niche concern; but it's beyond bad; it renders many math ebooks unusable. Often the equations are rendered as low-resolution bitmaps that are close to unreadable, or in other cases I've seen equation terms randomly spread hither-and-yon across the page. For scientific and technical books I would much prefer a larger, higher resolution device. It's too bad nothing really fits the bill because I hate throwing out cases of obsolete technical books every year.
If I had to choose just one format, I'd choose paper. But I find ebooks have their uses.
We had a lot of odd minicomputers in my high school, but the one I used most in school was a Digital Equipment PDP-8. You loaded the bootstrap from a paper tape reader, and you loaded the paper tape reader program by switches on the front panel which allowed you to set memory address contents word by word and set the program counter to a particular octal address. Input/output was through a teletype that printed on a roll of paper.
I have to say that this primitive hardware was as satisfying in its way to work on as the latest core i7 laptop I'm writing this on -- despite the actual core memory's unreliability in our building which was next to a busy subway track. I suppose I did have positive feelings toward DEC, until I got to college and worked in a lab that stored its research data on RK05 disc packs.
In my experience -- which as you can probably tell is by now extensive -- there are two kinds of people, those that adapt readily to new stuff, and those who stubbornly stick with whatever they already know. And as you look at successively older cohorts, the greater the proportion of stick-what-you-knowers there will be.
So the idea that you'll imprint *kids* on your technology is dubious. Yes you will imprint them, but it won't prevent them from switching to something else.
The human population is composed of experts, with divisions of labor. It is not unreasonable for AI programs to have areas of expertise.
In fact this is not true. The human population is composed of experts, some of whom have required in addition specialized skills due to division of labor.
I once read a book by an early aviator on techniques of navigating by landmark from the air. He recounts a number of feats of navigation by what were then called "primitive people", including one account of preparations for raid by a group of 19th Century teenage Apaches on an enemy village. None of the boys had ever been there, and so they sought out an elderly man who'd been there once when he was a boy. He described all the landmarks along the way, e..g. turn south at the hill that looks like such and so, a process took almost two days because the village in question was almost five hundred miles distant.
Now if a 19th C Apache had devised an intelligence test, chances are you or I would score retarded. There's no way I could give turn-by-turn directions to a place I'd visited just once thirty years ago. And if I could the chance you could just hear them and then go there without any difficulty is nil. We are simply too unfamiliar and unpracticed a task that is second nature to them.
Slashdot Mirror
Yes, the 2% of the desktop/laptop world is the "only" relevant question here. /sarcasm
Depends on what you mean by "linux". If you count Chromebooks, nobody is quite sure how many are being sold, but it's clearly not so insignificant.
But of course by "linux" we usually don't mean the linux kernel surrounded by a proprietary userland and tied to proprietary network services; we mean a proper distro. But I know *I* have installed Linux on machines that are only used for browsing just to avoid the annoyance of Windows.
You may be right, but if you've ever looked at scientific papers in nutrition most of the studies are modest, and there's a lot of arguing over what looks like not enough data. More nutrition research could wipe out a lot of common-wisdom extrapolations from limited data and replace them with solid, evidence-based results.
The US spends about 1.6 billion in nutrition research annually, which sounds like a lot, but it's quite evident that a really good nutrition study (large enough sample size, long enough duration, adequate experimental controls) is extraordinarily expensive. And to put 1.6 billion in perspective, we spend 3.2 trillion on health care annually, a lot of it to treat conditions (diabetes, obesity, hypertension) that we know are connected to nutrition but in ways we don't fully understand. I don't think there's anyone who doesn't believe we could achieve significant improvements in longevity and quality of life with better nutrition, but catch is we don't know how.
We're funding nutrition like "small science", when it should by any reasonable measure be big, "moon shot" level science.
It's surprising what scientists learn when they actually look, which is why it's important to fund research.
The gold standard in calorie consumption measurement is something called "doubly labelled water" -- basically a scheme for measuring energy use using water tagged with uncommon isotopes of hydrogen and oxygen. The incorporation of these isotopes into metabolic by-products provides an indirect means for measuring metabolic rate in real world conditions.
So researchers did a doubly labeled water study on the Hadza of Tanzania, one of the few remaining hunter-gather societies in the world, and discovered that there was no significant correlation between activity level (measured by GPS) and calorie consumption. While they did exhibit more energy usage than sedentary westerners, it was only on the order of 200 calories/day -- which admittedly over a year is a lot of calories.
Now it's a mistake to extrapolate from the Hadza to people living in industrialized economies. The Hadza are smaller, leaner, much more active and eat a much different diet which varies in calorie content every day. But the most important thing to take away is the unreliability of the naive "scientific" model of the human body as being like, to a first approximation, an insulated calorimeter that can only shed energy by exercise.
Or it could be a non-representative sample. Or the difference in rates might not be statistically significant. Or the sample used might be too heterogeneous in terms of content or subject to precisely compare rates. Or women could submit code more frequently and have the same acceptance rate. Or things might look different if you control for submission size.
It's nearly impossible to tell what's going on with a single aggregate figure like this without access to the underlying data, if not the code in question. Anyone can construct any scenario they want because there is not enough information to draw any conclusions.
If it could only perform a specific sequences of welding operations, I wouldn't count it as robot.
I meant physical actions. A robot has to be able to physically manipulate the world.
A robot is a machine that can be programmed to perform a variety of complex sequences of actions (e.g. an industrial robot in a car factory). This is in contrast to a machine performs a complex action for which it is mechanically specialized (e.g. a bottling machine at a brewery).
Naturally there is no perfectly sharp dividing line between the two. For example an industrial robot may have specialized attachments which allow it to weld, or to inspect welds for that matter. A bottling machine may be controlled by an industrial computer and might have other modules of a general purpose nature. They are in many ways more alike than different.
Which means there's no sharp dividing line between robots and computerized machines in general. Which is true of a lot of things: child and adult, night and day, etc. The law necessarily draws sharp lines between fuzzy sets. You're an adult at eighteen, but it could just as well have been 17 years 11 months or 18 years 1 month; 18 is just a conveniently round number.
Yes, I know what carbon intensity is, but the linked article confuses the issue and summary confuses it more.
Emissions intensity is emissions relative to the volume of an activity, so I was trying to point out that if you measure an activity relative to energy consumption, you would not expect much change in the transport sector because people are for the most part powering their vehicles the way they did in 1975. For practical purposes almost nobody is running their car on hydrogen, so the carbon intensity per unit of energy used is for practical purposes determined by the chemistry of alkane combustion.
If people were switching over from coal-powered cars, it would be a different story. Coal has a lower energy density by weight and by volume than gasoline. It has a lower energy density by weight than natural gas too (although higher by volume). The reason is that in coal the carbon atoms are bound up in large sheets of polycyclic rings. The net energy you get from oxidizing the carbon in coal is less per carbon atom because you have to account for the cost of breaking up those aromatic rings.
I suspect we're suffering from a misunderstanding of terminology here.
Thermodynamic efficiency has nothing to do with carbon emitted by an engine per unit of fuel; that's determined by stoichiometry. Basically every carbon atom that goes into your tank comes out the tailpipe as soot, CO2, and a very smalll quantity (these days) of unburned alkanes.
Since that means that the CO2 emitted per unit consumed by the engine is approximately constant, I suspect they're talking about the entire supply chain including extraction, refinement, and transport.
Suppose the supply chain consumes a gallon of oil to get one gallon of gasoline from the ground to your car tank. Then every gallon of gasoline you burn in the car will represent 40 pounds of CO2: 20 coming out the tailpipe, 20 emitted somewhere else before the gas even got to your car.
For centuries the intellectual basis for conservatism has been set, not by Jesus, or Adam Smith, but by Edmund Burke, whose philosophy could be summed up this way: if it ain't broke, don't fix it.
Burke was the kind of man who could defend the monarchy while despising monarchists: he thought the notion that monarchy was an ideal form of government was fatuous twaddle. But he thought all grand, all-encompassing theories were foolish, so he wasn't any more enthusiastic about pure democracy. Burke preferred a monarchy restrained by a democratically elected parliament not because it was the best system, but because it worked, experience showed that men could be tolerably free and prosperous under such a system.
So the notion that we need to "fix" an innovative segment of the economy to be more like what our theory of what an innovative industry should look like is about as un-conservative as you can get. It is, in fact, radicalism of the sort Burke detested.
Unless in that time you developed a catapult to throw it for you. :)
Do the math. To achieve a range of 20km, the catapult would have to launch the football at 990 miles per hour -- and that's in a vacuum.
>back in MY day, hurdy durdy doo
Back in my day we had the pill and there were no STDs that couldn't be cured with a week or two of antibiotics, with predictable results. So yes, in some ways things were better.
It doesn't mean everything was better. Python is better than rolling my own control structures with Fortran IV's computed goto.
When you get older you'll realize "you win some, you lose some" is pretty much what life amounts to.
... the gods themselves contend in vain, Schiller said.
But maybe not. I just finished Hanah Arendt's famous Eichmann in Jerusalem: A Report on the Banality of Evil, about the abduction and trial of the former SS officer who was in charge of "evacuating" Jews to the death camps. Eichmann claimed -- probably truthfully -- to be horrified and distressed when he saw what was happening in the extermination camps. But his horror was greatly mollified at the a conference in the Berlin suburb of Wannsee in which many important and respectable people discussed the Final Solution frankly and unabashedly, as if it were no big deal.
Arendt also points out something interesting about Denmark, a country which was under total military domination by the Third Reich but in which society from the King down resisted the expulsion of Jews. Not only were the Germans unable to expel even stateless Jews from Denmark, confirmed SS officers posted to Denmark would suddenly become unreliable on the Jewish Question.
This suggests to me that when you feel like you're powerless against stupid or even evil ideas, there is always something you can do that can be very powerful: you can set an example.
I think of my RSS feed as my morning newspaper.
Well, there's your problem. The idea of starting your day with a cup of coffee and a broad sample of current events has gone the way of the dodo.
"Transforming" the government is something a bunch of Silicon Valley executives could no doubt accomplish. The problem is that "transformed" isn't necessary "better".
Facebook has transformed peoples' lives, to the point it's regarded indispensable for many. It hasn't really made their lives better.
"Transformation" is a goal without a point of view; and blank screen on which anyone is free to imagine whatever colors he likes. Not like "cheaper". Cheaper is quite concrete. "More responsive" would be harder to measure than cheaper, but at least it represents something.
And in fact, it *does* power the planet. It's only our narrow focus on things that are literally "accounted for" in our economy that makes it appear otherwise. If we had to replace all the things that sunlight does for us with our non-renewable energy sources they wouldn't even come close, but that all happens off the books so it's invisible.
Now I worked for environmental organizations in the years of transition from crying indian environmentalism to "sustainability" based environmentalism, and I always had a problem with the new framing: non-sustainability is by definition a self-correcting problem.
So if we survive as a society, that society will eventually be powered by (and limited by) solar energy. The problem isn't non-sustainability per se, but the predictable costs of running unprepared into the limits of the processes we depend on.
You might be the most powerful man on Earth, but your problems are proportionately bigger.
Any reasonably thoughtful person would look at the presidency, and conclude that being president means feeling frustrated by not having enough power most of the time. And it doesn't mean getting respect -- even respect people think is due to the office, not the occupant. Trump of all people should have realized that.
You can't bomb ISIS away, or shock-and-awe Syria into submission, or intimidate North Korea. Everything you might try to accomplish is accompanied by a perplexing calculus of risk/reward. Manned space exploration is no exception. If you want to do it, you need to make sacrifices.
Sure, if Trump's ego gets us to Mars, we should all be for it, but then a falsehood implies anything. I could just as truly say that if Trump's ego gets us to Mars, the crew vehicle will be powered by unicorn farts.
Well, to make things as simplistic as possible, Mars at its closest is over at thousand times farther away.
Now I don't have much of a throwing arm -- you might say I'm starting with nothing -- but I'm pretty certain if you gave me a few years I could manage to throw a football twenty yards. It doesn't follow that given a few more years I could somehow throw a football twenty kilometers.
PDFs are inconvenient in e-readers because PDFs are page-oriented. That makes them inconvenient on smaller-screen devices. That said, PDFs represent that page with a high degree of fidelity, which is PDF's biggest strength.
The problem with math ebooks published in AZW format is that they either render equations incorrectly (making them useless), or render them as bitmaps (making them less useful than they should be, and sometimes illegible).
In most cases I'd take an EPUB or AZW over a PDF for reading on a small device, but for math I'd take the PDF, despite its inconveniences. Or better yet, a physical book.
I'm in my 50s and my house is literally full of physical books. Every room is lined with bookcases most of them stacked two deep, and I've literally had to put jackposts in my basement to keep the floors from sagging.
Buying new books as ebooks means I don't have to get rid of my old books. It's also nice being able to travel with a generous selection of reading material.
Overall I find the reading experience to be about a wash, but that's a highly personal thing. For pure reading a physical book is better except in low-light conditions, but the search and note taking functions on an ebook are a big plus.
The biggest drawback for ebooks for me is the terrible mathematics typesetting, which is obviously a niche concern; but it's beyond bad; it renders many math ebooks unusable. Often the equations are rendered as low-resolution bitmaps that are close to unreadable, or in other cases I've seen equation terms randomly spread hither-and-yon across the page. For scientific and technical books I would much prefer a larger, higher resolution device. It's too bad nothing really fits the bill because I hate throwing out cases of obsolete technical books every year.
If I had to choose just one format, I'd choose paper. But I find ebooks have their uses.
We had a lot of odd minicomputers in my high school, but the one I used most in school was a Digital Equipment PDP-8. You loaded the bootstrap from a paper tape reader, and you loaded the paper tape reader program by switches on the front panel which allowed you to set memory address contents word by word and set the program counter to a particular octal address. Input/output was through a teletype that printed on a roll of paper.
I have to say that this primitive hardware was as satisfying in its way to work on as the latest core i7 laptop I'm writing this on -- despite the actual core memory's unreliability in our building which was next to a busy subway track. I suppose I did have positive feelings toward DEC, until I got to college and worked in a lab that stored its research data on RK05 disc packs.
In my experience -- which as you can probably tell is by now extensive -- there are two kinds of people, those that adapt readily to new stuff, and those who stubbornly stick with whatever they already know. And as you look at successively older cohorts, the greater the proportion of stick-what-you-knowers there will be.
So the idea that you'll imprint *kids* on your technology is dubious. Yes you will imprint them, but it won't prevent them from switching to something else.
Actually I do have nice hair. Or did, but I buzz it because I can't be bothered.
The rest is airy-fairy drippy-hippySJW nonsense.
Appeal to the stone.
The human population is composed of experts, with divisions of labor. It is not unreasonable for AI programs to have areas of expertise.
In fact this is not true. The human population is composed of experts, some of whom have required in addition specialized skills due to division of labor.
Oh, and here's another example I just thought of.
I once read a book by an early aviator on techniques of navigating by landmark from the air. He recounts a number of feats of navigation by what were then called "primitive people", including one account of preparations for raid by a group of 19th Century teenage Apaches on an enemy village. None of the boys had ever been there, and so they sought out an elderly man who'd been there once when he was a boy. He described all the landmarks along the way, e..g. turn south at the hill that looks like such and so, a process took almost two days because the village in question was almost five hundred miles distant.
Now if a 19th C Apache had devised an intelligence test, chances are you or I would score retarded. There's no way I could give turn-by-turn directions to a place I'd visited just once thirty years ago. And if I could the chance you could just hear them and then go there without any difficulty is nil. We are simply too unfamiliar and unpracticed a task that is second nature to them.