There's no constitutional right to drive a car, so I don't think that prior restraint enters into it. Most places, cars need registration, plates, insurance, inspection, and a licensed driver. Lack any of those, and you don't get to drive.
The problem with your incentives-based approach to this is that when you are drunk, incentives don't work so well. It lets us establish a moral system whereby we can determine after the fact that people are bad and smite them, but it does not undo the harm that they did, or prevent it in the first place.
For your purposes, no, an electric car would definitely require a lunch+charge stop. Twice:-). You also said that $5 gas would make the trip unaffordable, which is what did not make sense to me.
What doesn't quite make sense to me about the e-cars, is that they seem to require a Goldilocks commute. Too long, and the range becomes an issue. Too short, and you don't save enough gasoline to justify the incremental cost. And a commute shorter than five miles (mine is ten), depending on the weather and route, you can easily do on a bicycle, or a scooter, or a motorcycle -- these are all variously efficient, and generally cheaper than any of these cars. (And those of us who already bike to work often, are wondering where our big subsidy is.)
I'm unconvinced. There's lots of cars that get 40mpg, it's 280 miles to San Antonio. Round trip, 14 gallons of gas, costs $42 at $3, or $70 at $5. Cars need insurance and repairs, this is not going to break the bank.
It may be the case that you own some huge car that gets much worse gas mileage than that, but if you're so tight for money, why would you own such a huge car?
No particular reason you can't run your logistics through a rail network, assuming (ahem) it is not run by assholes. Bad weather, and the rails can be somewhat more reliable (based on what keeps running up near Boston, when it snows badly).
Damage to a road is proportional (so I read) to the 4th power of per-wheel weight. Apparently (Michigan, what Google gave me) a truck can have 18,000 lbs per axle, or 4500lbs/wheel. Versus my car (Honda Civic), at about 600lbs/wheel. 7.5x the weight per wheel, or 3100x the road wear, per wheel, and the truck's got more wheels (not all of them loaded that much). For every dollar towards road construction or repair that I pay, a heavy truck should pay several thousand dollars. And it is true, there are cars out there much heavier than mine -- and they should also pay more. I see no reason why I should be subsidizing other people's lifestyle choices like this.
The problem, not immediately obvious, is that if you shrink the grid size in a finite-elements simulation (which describes very very many of them), you must also shrink the time step, because you are modeling changes in the physical world, and it takes less time for change to propagate across a smaller element. And at each time step, everyone must chat with their "neighbors" about the new state of the world. The chatting is what supercomputers do well, compared to a city full of gaming rigs with GPUs.
The constraints of faster chatting also drive up power density.
Another issue with COTS hardware (or any other) is that if you use enough of it, failure of some part becomes a certainty. This means you need redundancy and/or checkpointing, and changes your tolerance for hardware failure (COTS may not be tuned to that design point). Full redundancy uses twice the resources, but means you (almost) never wait; otherwise, checkpoint and recover eats into your wall time.
Tunnelling is a fabulous suggestion, but it's not intuitively obvious (I figure it out when I need it, and retain the info just long enough to get the shell script working). If you can provide the incantation, that would be very helpful.
Maybe. If you look at some of the issues listed, arguably they are fall out from being a little half-hearted in the pursuit of non-evil. Schmidt's remarks on privacy were not helpful for the company image. Size they can do little about (I mean, come on, "Google", it's a huge number, for a huge company), but privacy and copyright, absolutely, they could do better than they did, especially privacy.
The other advantage of "don't be evil" is that it removes a huge number of the choices that you might otherwise consider, giving you a much smaller search space for what-to-do-next.
Open container laws in many states (don't know offhand about the one I am in) do not allow any open beer at all in a car. You have to focus on what's important:-).
Yes, but -- that's not carbon neutral. You get perhaps double the output from each unit of CO2 pollution (one unit from burning the coal, one from burning the output of the oil-bacteria, then it escapes to the atmosphere).
And last time I went to NY, it was just me, I was able to work on the train, it was lovely. And I could buy a beer, and drink it. Can your SUV do that?:-)
The one thing I've measured pretty accurately over the years, is code translation speed. I can translate about 1000 lines per day of code from (for example) C to Java, or C interfaced with vN of some other language, to vN+1.
Code translation is far from rocket science, so I think that's my limit; for gigantic projects I might write a tool, but in practice it's just busywork and emacs macros.
Also, the intellectual content of the 100loc/day work was pretty high; we did new stuff. If we had been in the research business, we'd have gotten several papers out of what we did.
I think the point is that I was using my hours rather differently; because I could get up and be distracted (productively, knocking chores off the list) when I was thwarted by a problem, I could avoid getting tense about the problem, AND also have time left over at the end of the day, because all the housework was mysteriously done. If you're in an office and burn 11 hours, you go home, and there's still more to do.
I've done a little bit of pair programming since then, and there's a sort of level of problem that it seems to be best for, and I am not 100% sure it is not its own kind of social pressure. There were things that we pair programmed over the phone and it worked well, and there were problems we hashed out "pair-hiking", but often enough, it was more a matter of waiting for the answer to appear, once I had stared at the problem for a little while.
What's interesting, and a little disappointing, is that nowadays a "work from home" day is mostly shot to hell. I'm not sure if it's result of new and improved distractions, or if it is because it is not part of the routine, and instead of running a little ahead of the game at home, I am running a little behind. I've been consistently able to get certain kinds of work done in noisy public places, to the point that when I have papers to review, I just budget mornings to go sit in Peet's. Years ago in college, I would do homework in the pub (back when 18-year-olds were responsible enough to drink).
It depends upon how you count your 10 hours. I worked for quite some time in a startup (distributed, in multiple homes), that ultimately failed, and astonishingly we all remained friends (we started friends). I looked at some logs of when we did commits and when we exchanged emails, and we worked ridiculous hours, for a long time, and generated code at a sustained rate (years) of something like 100 lines per day per person, not too picky about definition of "line", not too buggy (it was a brand-X JVM. We could run Swing apps. We could run Websphere.), including overhead for a fair amount of the crap that goes with running a business. This also included learning how to do stuff that was completely new to us, like emulating Sparc FP on an Intel, or translating gdtoa (FP-to-string) from C to Java, and debugging it.
However, we were working at home, and I know that those "work hours" had holes in them, sometimes lots of holes in them. If I got stuck on a problem, I would do laundry, wash dishes, rake the yard, anything else that needed doing, and usually a solution would occur to me while I was doing something else. I think if I had been in an office, "forced" (by social pressure, if nothing else) to "look productive" for those hours, there is no way I could have done it.
And yes, we all got non-trivial amounts of stock. Those of us who were getting paid a salary, obviously got less.
There's nothing special about winter. Ever heard of snow tires? They make them, for bikes. I own some, I rode on a lot of ice, to work, yesterday (ran into a guy on a bike, also with snow tires, same advisor as a guy that I work with, had a nice chat, took an offroad route to work. So not just ice, but lumpy, sculpted ice, mixed with packed and loose snow). In dense areas, most people don't have door-to-door freeways, so cars are at best 2x faster, if that, plus cars have occasional outrageous traffic delays that bikes do not.
You're also single-counting time spent on the bike, and not giving it credit for the exercise -- that's time you would otherwise spend jogging, or at the gym, or spinning pedals in your basement (people, "all over the US", don't get enough exercise -- it is a medical necessity, and far, far deadlier than the usual risk of riding a bicycle).
Sometimes true, sometimes not. Blind makes it better than not-blind, and playing fair (not Googling for the TR corresponding to the submission) helps too.
Remember, it's a human process, so imperfection is pretty much taken for granted.
But, for the specific case of Boston, how do you really think this would work? You cannot just rip up the streets -- emergency vehicles and delivery vehicles still need to get through. There is not necessarily room overhead (you may recall, we just spent 2-3 Iraq war-months of money burying an overhead freeway). The bicycle example is instructive -- any time a street is restriped to take FIVE WHOLE FEET for a bike lane, there is no end to the pissing and moaning about the horrible that will make the traffic jams.
You've not yet convinced me (at all) that this stuff would have the capacity to replace busses on the busiest lines; right now, the busses converge on Harvard Square, so that one line that I mentioned, where you would need one every 30 seconds -- well, there's several other high-traffic lines, converging on the same place (#71 and #77, in particular, but there are others).
And subways? That's a huge amount of traffic, crammed into a little tiny place. You might be able to replace cars, but if you need special tracks, you're SOL.
I guess my real problem with "concept" solutions is that there is a certain slipperiness to them. If I have the option of starting from scratch and giving the culture a mild whack on the head, well, hell, there's the bicycle "concept". So for example, I ride a cargo bike. I *could* make it an electric, aerodynamically faired cargo bike, but then I would add $2000 to the cost, and push its weight up past 100 lbs -- and I start to compromise some of the advantages of it being a "bike". Even now, I don't fling my bike over my shoulder and trot up stairs with it. If I am forced to choose one bicycle to ride (I own four), I do not get the "llght-weight" + "cargo carrying" + "low cost" option; I must choose, and there are limits. Elevated tracks, that cannot be cheap. There has to be a maximum speed, and I'll bet that interacts with a lot of other stuff (like cost). At the same time, you are competing with anything else that is similar, yet has a cheaper cost to deploy. A crowd-sourced carpooling app, will probably eat your lunch. If smart cars actually get smart enough to be self-driving, that seems like it would nicely subsume your people carriers, anyway -- a car seats four, it could just drive to pick you and everyone else up, and then drop you off, too, returning to its owner in the end, or maybe it is owned by the local transit authority. As near as I can tell, the only thing that "tracks" solve (in this near future), is a problem of arguing over who's responsible when there is a crash. A legal problem, NOT an engineering problem.
Bicycles ought to eat your lunch (and the electric cars too), if nothing else because though no single bicycle can be all things to all people, you can park a half-dozen different bikes in the space used by one car, and the whole half-dozen probably costs less, and bikes really are in use all over the world all the time. And, in the same places where electric cars win best, and where mass transit wins best, and where these PRTs win best, bicycles also win, but at lower cost, comparable speed, superior flexibility, superior load ability (*), and superior efficiency. Their main problem is that they don't much keep the weather off (I find that clothing is a big help for that.) Also, you need the exercise -- that risk, is much higher than the crash risk.
(*) Bicycles get to cheat in ways that nothing else does -- you can get off the loaded bike, and wheel it in to exactly where you need the load to be. My car can carry more, but who carries it from parking lot to destination?
As the other guy said, don't let one bad apple ruin the whole bunch. There's plenty we know about psychology, that we really do know. The Lake Wobegon Effect for one -- if you look for it, you find it. The Fundamental Attribution Error is another. People make money (and politics) by exploiting these effects.
Add to this -- reviews for conferences (in my field) are often blind -- no idea who the author is, and we're perfectly willing to admit a paper (that's done well) that might start an argument.
How is a "concept" different from "vaporware" or "demoware"? We've got concept solutions out the kazoo.
I have a strong preference for stuff that works, proven in wide real-world use, and I'd like to see the math, on how many of these PRTs would be required to replace (for example) the #73 MBTA bus line, or the Red Line, especially at peak rush.
My preferred solution is bicycles -- already deployed many places, proven fast enough in urban areas, proven flexible, proven efficient, proven compact, proven carrying capacity. Extremely low infrastructure cost, low impact (bike paths wear out when tree roots tear them up, or when floods wash them away). If weather is an issue, we might splurge on the infrastructure, and put some sort of a roof over "bicycle highways" to reduce the rain, snow, and sun.
Consider the proof set for bicycles. We can see that they are very low cost, because there are countries where they are the most widely-used form of transit for very poor people (e.g., China). Are there any examples like this for PRTs? We can see that they work for prosperous people (e.g., the Dutch, the Danes, other northern Europeans). Are there similar examples for PRTs? I can see, personally, that they work for me (cargo bike, about half of my ten-mile commutes). Their efficiency is (conservatively) estimated at about 500 miles per gallon of gasoline-equivalent (e.g., vegetable oil), though of course you must derate that by the cost of producing the fuel (low for peanuts, high for steak).
Given our history and the demographics, for Harlem I expect the invisible hand had some "help". See also, the bridge design in the counties north of NY, intended to keep busses full of "those people" out of the suburbs.
But, now, what keeps poor people from moving to NYC? It's bleeping expensive, that's what. The semi/sub-urbs are a better example of market interference, where snob zoning makes housing expensive by requiring a minimum lot size. (And again, there's sad history of more blunt exclusionary rules, boatloads of it in some places. But now, mostly, it's all done with money.)
My datapoint is the red line repair over the longfellow bridge in Cambridge/Boston, where it was explained that they plan to take a lane of traffic temporarily, because the subway carries more people.
I think your "Personal Rapid Transit" is probably not going to be widely adopted, for the following reasons (I'm assuming you're talking about what I read on Wikipedia).
"There's no room." Ask any cycling-path advocate what objections they hear, this is high on the list. Where would we put your PRT?"
It's not that fast. 12mph, that's a lackadaisical biking speed. That's a 50-year-old man on a cargo bike riding-in-traffic speed. I can ride ALL DAY at that speed.
Not very flexible. It goes where you run it, and nowhere else. This is what keeps people off subways, and keeps them off busses. This is why people drive cars into Boston (among other reasons). It's nowhere near the flexibility of a bicycle, which can easily go off-road, on sidewalks, through narrow spaces, or be wheeled through/around puddles, snowbanks, etc.
Not likely to be that energy efficient, compared to the alternatives. Suppose "Smart cars" are especially good at carpolling (for all I know, there is already an iPhone app for that) and finding easy time/space matches for your commute. Poof!, you've just doubled or tripled the energy efficiency of your car. PRT might be competitive, but consider the ease/expense of deploying an iPhone app, versus PRT. No comparison at all.
Bicycles are another competitor that is far more efficient, cheap to deploy, and flexible.
No carrying capacity. People come and go on foot, they can only carry, what they can carry. I happily haul 50-100 lbs on my (cargo) bike, and can go as high as 200. If you have a 200lb load to carry on foot, you're immobile, but I'm rolling. Other people use cars for that.
Probably not a high enough capacity for the "best" (dense-enough) areas. Semi-urban busses and subways carry an astonishing number of people. The bus line near where I live, at rush hour, runs full, and runs something like once every 5 minutes. At 50 per bus, that's 10 people per minute, or a PRT every 30 seconds -- on the existing bus route, not counting the additional load that you would hope to get from something better that replaces cars (which also pretty much at capacity into Cambridge/Boston). In addition, you've got to beat a bus at speed and efficiency.
Seems like you'd have an unpleasant amount of maintenance to do, with that many units, versus a bus or subway. Automobiles, the maintenance is outsourced via their owners, bicycles, anyone who's vaguely handy with tools can maintain themselves, even on the road.
I think you are right, but this is also mentioned in the discussion of the graph -- there's a two-mile unpopulated donut around the town, it looks like to me. Note, also, that the town noted on the graph with most similar density -- Lexington -- includes two family farms, some serious highway right-of-way, and several office parks. We play that trick, too.
And that density pattern cuts both ways -- if your destination is not in town, then the minimum distance is 2 miles, because you've got to cross the donut. My understanding is that kids (11 and older) from surrounding villages ride in to school in Assen, and it's 5 miles.
For me, the more telling comparison, that I discovered after writing that post, is Groningen, versus Cambridge-Somerville-neighboring. Similar weather, similar student populations, relatively flat, greater density here, yet they have the 57% ride share.
Slashdot Mirror
There's no constitutional right to drive a car, so I don't think that prior restraint enters into it. Most places, cars need registration, plates, insurance, inspection, and a licensed driver. Lack any of those, and you don't get to drive.
The problem with your incentives-based approach to this is that when you are drunk, incentives don't work so well. It lets us establish a moral system whereby we can determine after the fact that people are bad and smite them, but it does not undo the harm that they did, or prevent it in the first place.
WHY ARE THEY BEING ALLOWED TO DRIVE AT ALL?
Because when they're sober, they're good drivers? Or maybe, probation? You have to decide where to set the punishment-vs-prevention knob.
For your purposes, no, an electric car would definitely require a lunch+charge stop. Twice :-). You also said that $5 gas would make the trip unaffordable, which is what did not make sense to me.
What doesn't quite make sense to me about the e-cars, is that they seem to require a Goldilocks commute. Too long, and the range becomes an issue. Too short, and you don't save enough gasoline to justify the incremental cost. And a commute shorter than five miles (mine is ten), depending on the weather and route, you can easily do on a bicycle, or a scooter, or a motorcycle -- these are all variously efficient, and generally cheaper than any of these cars. (And those of us who already bike to work often, are wondering where our big subsidy is.)
Presumably you cut some regressive tax at the same time; this need not be at the expense of the poor.
I'm unconvinced. There's lots of cars that get 40mpg, it's 280 miles to San Antonio. Round trip, 14 gallons of gas, costs $42 at $3, or $70 at $5. Cars need insurance and repairs, this is not going to break the bank.
It may be the case that you own some huge car that gets much worse gas mileage than that, but if you're so tight for money, why would you own such a huge car?
No particular reason you can't run your logistics through a rail network, assuming (ahem) it is not run by assholes. Bad weather, and the rails can be somewhat more reliable (based on what keeps running up near Boston, when it snows badly).
Damage to a road is proportional (so I read) to the 4th power of per-wheel weight. Apparently (Michigan, what Google gave me) a truck can have 18,000 lbs per axle, or 4500lbs/wheel. Versus my car (Honda Civic), at about 600lbs/wheel. 7.5x the weight per wheel, or 3100x the road wear, per wheel, and the truck's got more wheels (not all of them loaded that much). For every dollar towards road construction or repair that I pay, a heavy truck should pay several thousand dollars. And it is true, there are cars out there much heavier than mine -- and they should also pay more. I see no reason why I should be subsidizing other people's lifestyle choices like this.
The problem, not immediately obvious, is that if you shrink the grid size in a finite-elements simulation (which describes very very many of them), you must also shrink the time step, because you are modeling changes in the physical world, and it takes less time for change to propagate across a smaller element. And at each time step, everyone must chat with their "neighbors" about the new state of the world. The chatting is what supercomputers do well, compared to a city full of gaming rigs with GPUs.
The constraints of faster chatting also drive up power density.
Another issue with COTS hardware (or any other) is that if you use enough of it, failure of some part becomes a certainty. This means you need redundancy and/or checkpointing, and changes your tolerance for hardware failure (COTS may not be tuned to that design point). Full redundancy uses twice the resources, but means you (almost) never wait; otherwise, checkpoint and recover eats into your wall time.
Tunnelling is a fabulous suggestion, but it's not intuitively obvious (I figure it out when I need it, and retain the info just long enough to get the shell script working). If you can provide the incantation, that would be very helpful.
I think it would be ok if it said, "Hello, I am Eliza."
Maybe. If you look at some of the issues listed, arguably they are fall out from being a little half-hearted in the pursuit of non-evil. Schmidt's remarks on privacy were not helpful for the company image. Size they can do little about (I mean, come on, "Google", it's a huge number, for a huge company), but privacy and copyright, absolutely, they could do better than they did, especially privacy.
The other advantage of "don't be evil" is that it removes a huge number of the choices that you might otherwise consider, giving you a much smaller search space for what-to-do-next.
Open container laws in many states (don't know offhand about the one I am in) do not allow any open beer at all in a car. You have to focus on what's important :-).
Yes, but -- that's not carbon neutral. You get perhaps double the output from each unit of CO2 pollution (one unit from burning the coal, one from burning the output of the oil-bacteria, then it escapes to the atmosphere).
Don't forget the cost of parking that SUV in NY.
:-)
And last time I went to NY, it was just me, I was able to work on the train, it was lovely. And I could buy a beer, and drink it. Can your SUV do that?
Code translation is far from rocket science, so I think that's my limit; for gigantic projects I might write a tool, but in practice it's just busywork and emacs macros.
Also, the intellectual content of the 100loc/day work was pretty high; we did new stuff. If we had been in the research business, we'd have gotten several papers out of what we did.
I've done a little bit of pair programming since then, and there's a sort of level of problem that it seems to be best for, and I am not 100% sure it is not its own kind of social pressure. There were things that we pair programmed over the phone and it worked well, and there were problems we hashed out "pair-hiking", but often enough, it was more a matter of waiting for the answer to appear, once I had stared at the problem for a little while.
What's interesting, and a little disappointing, is that nowadays a "work from home" day is mostly shot to hell. I'm not sure if it's result of new and improved distractions, or if it is because it is not part of the routine, and instead of running a little ahead of the game at home, I am running a little behind. I've been consistently able to get certain kinds of work done in noisy public places, to the point that when I have papers to review, I just budget mornings to go sit in Peet's. Years ago in college, I would do homework in the pub (back when 18-year-olds were responsible enough to drink).
However, we were working at home, and I know that those "work hours" had holes in them, sometimes lots of holes in them. If I got stuck on a problem, I would do laundry, wash dishes, rake the yard, anything else that needed doing, and usually a solution would occur to me while I was doing something else. I think if I had been in an office, "forced" (by social pressure, if nothing else) to "look productive" for those hours, there is no way I could have done it.
And yes, we all got non-trivial amounts of stock. Those of us who were getting paid a salary, obviously got less.
There's nothing special about winter. Ever heard of snow tires? They make them, for bikes. I own some, I rode on a lot of ice, to work, yesterday (ran into a guy on a bike, also with snow tires, same advisor as a guy that I work with, had a nice chat, took an offroad route to work. So not just ice, but lumpy, sculpted ice, mixed with packed and loose snow). In dense areas, most people don't have door-to-door freeways, so cars are at best 2x faster, if that, plus cars have occasional outrageous traffic delays that bikes do not.
You're also single-counting time spent on the bike, and not giving it credit for the exercise -- that's time you would otherwise spend jogging, or at the gym, or spinning pedals in your basement (people, "all over the US", don't get enough exercise -- it is a medical necessity, and far, far deadlier than the usual risk of riding a bicycle).
Sometimes true, sometimes not. Blind makes it better than not-blind, and playing fair (not Googling for the TR corresponding to the submission) helps too.
Remember, it's a human process, so imperfection is pretty much taken for granted.
But, for the specific case of Boston, how do you really think this would work? You cannot just rip up the streets -- emergency vehicles and delivery vehicles still need to get through. There is not necessarily room overhead (you may recall, we just spent 2-3 Iraq war-months of money burying an overhead freeway). The bicycle example is instructive -- any time a street is restriped to take FIVE WHOLE FEET for a bike lane, there is no end to the pissing and moaning about the horrible that will make the traffic jams.
You've not yet convinced me (at all) that this stuff would have the capacity to replace busses on the busiest lines; right now, the busses converge on Harvard Square, so that one line that I mentioned, where you would need one every 30 seconds -- well, there's several other high-traffic lines, converging on the same place (#71 and #77, in particular, but there are others). And subways? That's a huge amount of traffic, crammed into a little tiny place. You might be able to replace cars, but if you need special tracks, you're SOL.
I guess my real problem with "concept" solutions is that there is a certain slipperiness to them. If I have the option of starting from scratch and giving the culture a mild whack on the head, well, hell, there's the bicycle "concept". So for example, I ride a cargo bike. I *could* make it an electric, aerodynamically faired cargo bike, but then I would add $2000 to the cost, and push its weight up past 100 lbs -- and I start to compromise some of the advantages of it being a "bike". Even now, I don't fling my bike over my shoulder and trot up stairs with it. If I am forced to choose one bicycle to ride (I own four), I do not get the "llght-weight" + "cargo carrying" + "low cost" option; I must choose, and there are limits. Elevated tracks, that cannot be cheap. There has to be a maximum speed, and I'll bet that interacts with a lot of other stuff (like cost). At the same time, you are competing with anything else that is similar, yet has a cheaper cost to deploy. A crowd-sourced carpooling app, will probably eat your lunch. If smart cars actually get smart enough to be self-driving, that seems like it would nicely subsume your people carriers, anyway -- a car seats four, it could just drive to pick you and everyone else up, and then drop you off, too, returning to its owner in the end, or maybe it is owned by the local transit authority. As near as I can tell, the only thing that "tracks" solve (in this near future), is a problem of arguing over who's responsible when there is a crash. A legal problem, NOT an engineering problem.
Bicycles ought to eat your lunch (and the electric cars too), if nothing else because though no single bicycle can be all things to all people, you can park a half-dozen different bikes in the space used by one car, and the whole half-dozen probably costs less, and bikes really are in use all over the world all the time. And, in the same places where electric cars win best, and where mass transit wins best, and where these PRTs win best, bicycles also win, but at lower cost, comparable speed, superior flexibility, superior load ability (*), and superior efficiency. Their main problem is that they don't much keep the weather off (I find that clothing is a big help for that.) Also, you need the exercise -- that risk, is much higher than the crash risk.
(*) Bicycles get to cheat in ways that nothing else does -- you can get off the loaded bike, and wheel it in to exactly where you need the load to be. My car can carry more, but who carries it from parking lot to destination?
As the other guy said, don't let one bad apple ruin the whole bunch. There's plenty we know about psychology, that we really do know. The Lake Wobegon Effect for one -- if you look for it, you find it. The Fundamental Attribution Error is another. People make money (and politics) by exploiting these effects.
Add to this -- reviews for conferences (in my field) are often blind -- no idea who the author is, and we're perfectly willing to admit a paper (that's done well) that might start an argument.
Perhaps "unwitting dupe".
How is a "concept" different from "vaporware" or "demoware"? We've got concept solutions out the kazoo.
I have a strong preference for stuff that works, proven in wide real-world use, and I'd like to see the math, on how many of these PRTs would be required to replace (for example) the #73 MBTA bus line, or the Red Line, especially at peak rush.
My preferred solution is bicycles -- already deployed many places, proven fast enough in urban areas, proven flexible, proven efficient, proven compact, proven carrying capacity. Extremely low infrastructure cost, low impact (bike paths wear out when tree roots tear them up, or when floods wash them away). If weather is an issue, we might splurge on the infrastructure, and put some sort of a roof over "bicycle highways" to reduce the rain, snow, and sun.
Consider the proof set for bicycles. We can see that they are very low cost, because there are countries where they are the most widely-used form of transit for very poor people (e.g., China). Are there any examples like this for PRTs? We can see that they work for prosperous people (e.g., the Dutch, the Danes, other northern Europeans). Are there similar examples for PRTs? I can see, personally, that they work for me (cargo bike, about half of my ten-mile commutes). Their efficiency is (conservatively) estimated at about 500 miles per gallon of gasoline-equivalent (e.g., vegetable oil), though of course you must derate that by the cost of producing the fuel (low for peanuts, high for steak).
Given our history and the demographics, for Harlem I expect the invisible hand had some "help". See also, the bridge design in the counties north of NY, intended to keep busses full of "those people" out of the suburbs.
But, now, what keeps poor people from moving to NYC? It's bleeping expensive, that's what. The semi/sub-urbs are a better example of market interference, where snob zoning makes housing expensive by requiring a minimum lot size. (And again, there's sad history of more blunt exclusionary rules, boatloads of it in some places. But now, mostly, it's all done with money.)
I think your "Personal Rapid Transit" is probably not going to be widely adopted, for the following reasons (I'm assuming you're talking about what I read on Wikipedia).
Bicycles are another competitor that is far more efficient, cheap to deploy, and flexible.
I think you are right, but this is also mentioned in the discussion of the graph -- there's a two-mile unpopulated donut around the town, it looks like to me. Note, also, that the town noted on the graph with most similar density -- Lexington -- includes two family farms, some serious highway right-of-way, and several office parks. We play that trick, too.
And that density pattern cuts both ways -- if your destination is not in town, then the minimum distance is 2 miles, because you've got to cross the donut. My understanding is that kids (11 and older) from surrounding villages ride in to school in Assen, and it's 5 miles.
For me, the more telling comparison, that I discovered after writing that post, is Groningen, versus Cambridge-Somerville-neighboring. Similar weather, similar student populations, relatively flat, greater density here, yet they have the 57% ride share.