Now before you call me a fascist and a racist and a sexist and a whatever-the-fuck-ist, please do consider that the population growth curve is approaching vertical
Actually it's not. History is showing that the trend is quite different. You normally see a baby boom almost identical the the USA's as a country industrializes, as technology allows vast gains in infant survival rates, then the birth rate crashes within a couple generations to sustainment or even below.
First world countries have more trouble with not enough births to maintain population than too many.
I thought the point of using car parts is that they are cheap, easily available, and run on 12 volts
The thousand dollars quote leads me to believe that you're right - it wouldn't be even a grand if you're taking parts out of a junk yard, but who wants to do that? I figure that these incubators are using new parts, just ones from the automotive industry and not the medical industry for economy of scale and robustness. Though, yes, the ability to use junkyard parts for repairs is mentioned.
For example, most headlights are under $20, so that's only $40 for your heat source. 12V cabling is relatively cheap, there's various thermostats you can get. They're vibration resistant, and a AC-DC transformer will provide a good amount of protection from surges, especially if you put a car battery in the circuit to provide backup. Automotive fuses can provide safety and prevent damage. For that matter, car equipment is designed to take anything from like 12V to 14.4-15V, so it's robust from that angle as well.
The article mentions detractors that say that intervention, skilled delivery people, emergency care would be better. I'd argue that those would cost more - importing a western trained doctor is expensive. Low-hanging fruit, people. One step at a time. Incremental improvement.
Heck, if it's good enough, I'd like to see them in our hospitals. Perhaps a fancier, more expensive model, but still cheap compared to current ones. Look at our healthcare costs. How much money would be saved if we could get, instead of a $40k incubator, a $5k incubator instead? Figure a thousand incubators a state per year(50k total), that'd be $1.75 BILLON saved. Not including any maintenance savings, given that current incubators are maintenance hogs per the article, and I didn't see any mention of parts that you'd expect to replace any given year with the auto one. Sure, not much against the trillions we spend now - but as they say, a billion here and there, and suddenly you're looking at real money.;)
True, but when higher technology is actually available, the birth rate drops. This has held true for about 1/3 of the earths population across several cultures.
I'd argue that it's more due to quality of life improvements and education, but technology largely allows them, so it's a bit of a moot point.
your car wasn't worth much anyway it might not be worth it.
There was an article in a paper a while back where the police mentioned they'd confiscated like seven cars from this one dude - he had basically a lifetime revokation for DUIs, and they'd take his car whenever they caught him(usuaully drunk). He'd just go out and buy another cheap sub-$500 car - cheaper than impound fees and such. Part of the article was, of course, outrage over why the guy wasn't in prison.
Personally, call me old fashioned but I think that car confiscations, even/especially for drug stuff should be handled through the courts. Confiscations, period, for that matter.
As others have pointed out, regenerative braking works *much* better when you're feeding back to a capacitor instead of a battery
The difference is much smaller with the Tesla's EV sized batteries compared to the 'just big enough' NiMH cells in most hybrids today though. I'm seeing figures of 99.9% efficiency for charging LiIon, 66% for NiMH. Given the increase in circuit complexity to charge an EEStore capacitor, I don't see it being any more efficient than LiIon. You just need a large pack to abosrb that many amps - and being an electric sports car, the Tesla has the amp capacity available in it's 6,800 cell battery pack.
The lighter weight helps, but I figure the best way for EEStor to win is on the basis of cost - everything else LiIon is competitive on, even if it ends up weighing 3 times as much. Heck, battery pack cost is the #1 killer of electronic vehicles today.
Hmmm... EEStore might be better off targeting the hybrid market for now than the EV market. Smaller, lighter, cheaper electric energy storage.
Given what I've seen of the Volt, there isn't any space to be taken away in the Cabin, really. There are women's purses that won't fit in what the Volt calls the 'trunk'. Then again, you don't expect to see women who carry those purses in a Volt either...
Shaving off nearly 500 pounds will still increase the range of the volt, though. It's not that heavy of a vehicle to begin with. Though it would need increased/more complex charging/discharge logic and equipment, so the savings might be closer to 400 pounds.
Eh... I'd like to see a PHEV light truck with two of them underneath the bed. No need for sandbags in the winter!
Of 'production' vehicles, the Tesla is currently the best of the pack for regenerative braking. But then, like you note, it's pack is designed more around capacity than amperage. The LiIon chemistry is also more efficient at taking a charge than NiMH.
For most hybrids, they're actually designed more for amperage - they want to keep the battery pack small and light, after all they still need to shove an engine in there. Also, because of said engine they can afford to be a bit lax on the regenerative braking. So they generally can't pull all the energy from an 'emergency brake', but can do a moderately hard stop from something like 40mph.
At the size needed to get those amperages, you get at least a few miles of capacity, so until we start shifting to plug in hybrids, they're good enough.
Actually, they weren't putting lead in the gas yet in the model t days. It was after that - when demand was rising and octanes falling that they started putting lead in. By that very wiki article - TEL(Tetra-ethyl lead)'s antiknock properties were discovered in 1921. The Model T ford was produced 1908 to 1927. So almost the entire run of Model T's would have had been designed to run on unleaded gas. Unleaded gas that - up to the '20s or so was actually pretty high octane because they could pick the part of the refinery stack with higher octanes for the gasoline(more t-butanol than n-pentane, for example). Later Model T's were built with lower compression engines due to the falling average octane ratings. On a sidenote - man, talk about making a difference, they say that lead levels in blood have dropped by 75% since they banned TEL.
Modern cars do depend on a much higher octane rating than historical vehicles.
It's actually the opposite - Our cars are normally built to run on pretty low octane ratings today. We have to take a huge swath of the stack for gasoline to satisfy our demand for it, and the result is that our gas, knock wise, is pretty low.
From what I can find, the Model T ran on 93 octane. Not exactly what I'd call a low octane.
Early gas was actually pretty high octane(but tolerances weren't as tight); we didn't actually need all that much of it and it was still competing against Ethanol*, among other fuels. It was only later that gasoline demand started getting high enough that they started running short on the higher octanes, and needed to mix in lesser octane hydrocarbons.
One interesting fact i came across was that the Model T was Ford's original dual fuel vehicle - it featured manual spark advance control and could run on anything from 100% gasoline to 100% ethanol.
*During this time period, everything was competing. There were dozens of electric car companies; steam, ethanol, diesel, gasoline were all competing.
It'd get quite annoying trying to keep track of all my games and all the distributers on steam. Stardock works closer to what you say, but I view it something like talking down a Heroin addict - you can't expect EA to back off 100% immediately.
As far as I know they've fixed most of those issues. Had a bit of problem with steam being a resource hog- but that seemed to be fixed sometime last year.
Not saying that I don't love them as well. What I meant by Second tier is that Steam gets games that get TV advertising and places on the shelf at every store that sells games. Stardock, doesn't.
The current exchange rate isn't helping right now either. Steam games are frequently cheaper at the moment, if you buy them on special. Also, steam games are the prettier visually.;)
But I'm running out of space on D:, so am looking at throwing another drive in and redoing the system a bit. It'd be nice to have the option to NOT have steam and all of the games on the same media.
It is the only authentication system that actually gives you something in return for authenticating your game, and it doesn't bitch about me having virtual drive software.
, but their game list is more second-tier, if you know what I mean.
Steam is winning on number of games in my list right now, mostly due to their offering package deals quite frequently, as well as the 75% off deals. Even though I have a physical CD somewhere, I might just buy their 75% off($5) Stalker deal right now.
But maybe it'll convince EA that at least over restrictive DRM IS an issue - and SECUROM, limited installs, complicated activation schemes and all that is the incorrect method to go about doing DRM.
Or maybe a correct wording would be 'you can't get something for nothing' - you CAN get consumers to accept DRM as long as you offer true advantages to go along with it. I happen to like the idea that even if my house is struck by a meterorite and everything is destroyed I'd be able to play my games again as soon as I got a new computer and an internet connection.
To put it another way, while I've given up the ability(for now) to lend/sell/give away my games, on the other hand I'm also able to play any games in the future with relatively no problems, simply by re-downloading it.
As the success of iTunes and such shows, the ability to redownload your purchases whenever you want often trumps the portability/sellability of physical mediums.
As all my friends are either non-gamers or get the same games anyways, it's not a big deal there, and I like occasionally playing an old game.
No media, no serial numbers, just a single username and password for all my games.
Free unlimited downloads, relatively automatic updates, etc... Though changing the install directory could be good.
I bought Crysis through the EA store download method as an experiment. While I captured the download file that should allow me to reinstall, I'm not sure I'd be able to today. With steam, that wouldn't be a problem.
I have to agree, I like steam. They manage to do online download gaming right.
Oops, my bad. I was remembering the GB amount incorrectly.
Or your pricing information is about 6 months out of date. Somebody needs to keep some statistical information on the price history for SSD drives, and their platter counterparts to make some neat graphs - largest capacity available, cost for said largest capacity, size of 'sweet spot'* drive, cost per GB at the largest, and sweet spot.
Expand it out a bit further - track MLC and SLC seperately, as well as laptop vs desktop drives. Maybe server drives as well.
Hmm... Happened to have some price history from the 24th of november - HD price per gig dropped from $.10 to $.9. At the same time, the 'sweet spot' for SSDs dropped from $2.17 to $1.95. Right now it seems to take on-board RAM cache to get the best performance for writes - but let's be honest - how many HD's DON'T have significant amounts of cache today? The 1.5TB hd that has the best GB/Dollar has 32MB of cache.
*Best price per GB, or maybe where the price per GB flattens out. Eh, keep both. Call the smaller 'smallest economical drive'
Go back far enough and 2 MB would have cost you thousands.;)
But yeah, the bleeding edge has always been expensive - and in multiple ways.
For example - The most of the current line of MLC flash drives suffers from poor write performance due to some flaws. The fix is looking to be some amount of RAM to cache the writes for speed. Many times the 'best', the newest, suffer from prototyping errors, things like overheating, etc...
Same deal with cars - my mechanic relatives have always advised against buying a new car the first year it's introduced or after major revisions.
Still, I haven't noticed anybody else mention this - the largest 2.5" laptop hard drive currently available is 500GB.
That means that, in 2009, flash memory will equal the largest 2.5" HD, the same form factor of SSD flash drives themselves. Maybe not in price - We're still looking at $110 for the hard drive vs $700-1400*. Going further in, I figure that a 20-25GB SSD 'drive' will cost ~$35 sometime next year. Even cheaper in bulk/wholesale. At that price I'd expect to see them start showing up as the sole drive for smallest notebooks - even/especially economy models, built onto the very motherboard, perhaps with an empty drive slot to take an upgrade HD for those who need more storage.
*Going by the current $700 for a 250GB OCZ Solid Series OCZSSD2-1SLD250G 2.5" 250GB SATA II Internal Solid state disk (SSD) - Retail. History has shown that increased memory sizes are generally accompanied with similiar decreases in price per unit such that the top generally stays in the same price range. So we're likely to see closer to double the GB per dollar instead of a doubling in total price.
Actually, I'd peg the requirements for a 'natural' monopoly to be a bit different.
The "desired output" is largely identical for all cutsomers - Agreed
The delivery method doesn't lend itself to parallel installation
The costs of install/setup for providing a service are so high compared to delivering the service that an already established company can economically kill anybody trying to compete
People want different things from their last mile connections. Some people merely want voice or voice+video service. Some people want on-demand content. Some people want hagh transmission capacity. Some people only want data... Etc..
And a single pair of fiber, or a few more pairs of copper can provide all of this.
If the diversity of customer needs is high enough (it should be), the second assumption also falls. If each of the last mile providers can attract a significant portion of the market, they should each be able to achieve a customer base large enough to bring down the costs of the network that would need to be passed on to the individual users.
The other problem is that you don't get significant economies of scale when it comes to last mile installs. It can cost hundreds to tens of thousands to run a data line to a single house. House N+1 still costs thousands. You have to mark and existing utility lines, repair concrete disturbed, install service equipment, run the main line, etc...
Makes me kinda wish we had had a bright idea and installed conduit in our cities. I recognize that it wouldn't work for water/sewer, but you'd be able to run local power, cable, telephone, and other data lines through them relatively cheaply. Access to the shared service boxes(and right to pull cable through them) would be controlled by the local government.
And lastly, proof that this is the case. We currently have a situation where most communities have multiple last mile providers with overlapping services. One or more cable companies, and an incumbent telephone company. Both of which can justify upgrading their last-mile networks to the point where it's essentially a complete rebuild. This would be impossible if the last mile were a natural monopoly.
They're converging as providers, and a lot of the reason there's two of them is due to regulations not allowing cable companies to provide phone service, and technical problems with running cable over twisted pair for the telephone companies.
They're both already established - due to the cost of running a wire infrastructure, both can rather easily crush any competitors who have to duplicate that.
"The government may now take steps to legislate so that Telstra can't build a network that competes with the NBN"
Now, I'll admit that I'm not that familiar with Australian legislative processes, but from my experience with the US one, wouldn't they be able to do this no matter what's going on, assuming they got a big enough bug up their butt?
Going on, it sounds like I should be glad to not have to deal with Telstra. Deutsch Telecom was bad enough when I was in Germany. Now I'm with a local coop, and love them.
Ah, but therein lies the problem. The shuttle estimates weren't particularly egregious, either. F(known cost, unknown cost) = [True cost], and in the shuttles case we now know both the known and unknown costs. However, in the 'ideal' case we're comparing against, we don't know the unknown costs. Sometimes even hindsight isn't 20/20.
Basically, it's two shuttle services to equal the TOTAL cost of the hubble construction(basically they removed the groundside expenses). Going by spare construction and whatnot, about 50% of the cost of the satellite was R&D, not construction. They ended up with a 'construction line' cost of ~$400 million, dropping the more they build, and improvements not being much more. Around $30million to launch it on a rocket.
Today, we should be able to do quite extensive servicing by way of robots and teleoperation - far cheaper than shuttles.
Slashdot Mirror
Now before you call me a fascist and a racist and a sexist and a whatever-the-fuck-ist, please do consider that the population growth curve is approaching vertical
Actually it's not. History is showing that the trend is quite different. You normally see a baby boom almost identical the the USA's as a country industrializes, as technology allows vast gains in infant survival rates, then the birth rate crashes within a couple generations to sustainment or even below.
First world countries have more trouble with not enough births to maintain population than too many.
I thought the point of using car parts is that they are cheap, easily available, and run on 12 volts
The thousand dollars quote leads me to believe that you're right - it wouldn't be even a grand if you're taking parts out of a junk yard, but who wants to do that? I figure that these incubators are using new parts, just ones from the automotive industry and not the medical industry for economy of scale and robustness. Though, yes, the ability to use junkyard parts for repairs is mentioned.
For example, most headlights are under $20, so that's only $40 for your heat source. 12V cabling is relatively cheap, there's various thermostats you can get. They're vibration resistant, and a AC-DC transformer will provide a good amount of protection from surges, especially if you put a car battery in the circuit to provide backup. Automotive fuses can provide safety and prevent damage. For that matter, car equipment is designed to take anything from like 12V to 14.4-15V, so it's robust from that angle as well.
The article mentions detractors that say that intervention, skilled delivery people, emergency care would be better. I'd argue that those would cost more - importing a western trained doctor is expensive. Low-hanging fruit, people. One step at a time. Incremental improvement.
Heck, if it's good enough, I'd like to see them in our hospitals. Perhaps a fancier, more expensive model, but still cheap compared to current ones. Look at our healthcare costs. How much money would be saved if we could get, instead of a $40k incubator, a $5k incubator instead? Figure a thousand incubators a state per year(50k total), that'd be $1.75 BILLON saved. Not including any maintenance savings, given that current incubators are maintenance hogs per the article, and I didn't see any mention of parts that you'd expect to replace any given year with the auto one. Sure, not much against the trillions we spend now - but as they say, a billion here and there, and suddenly you're looking at real money. ;)
True, but when higher technology is actually available, the birth rate drops. This has held true for about 1/3 of the earths population across several cultures.
I'd argue that it's more due to quality of life improvements and education, but technology largely allows them, so it's a bit of a moot point.
your car wasn't worth much anyway it might not be worth it.
There was an article in a paper a while back where the police mentioned they'd confiscated like seven cars from this one dude - he had basically a lifetime revokation for DUIs, and they'd take his car whenever they caught him(usuaully drunk). He'd just go out and buy another cheap sub-$500 car - cheaper than impound fees and such. Part of the article was, of course, outrage over why the guy wasn't in prison.
Personally, call me old fashioned but I think that car confiscations, even/especially for drug stuff should be handled through the courts. Confiscations, period, for that matter.
As others have pointed out, regenerative braking works *much* better when you're feeding back to a capacitor instead of a battery
The difference is much smaller with the Tesla's EV sized batteries compared to the 'just big enough' NiMH cells in most hybrids today though. I'm seeing figures of 99.9% efficiency for charging LiIon, 66% for NiMH. Given the increase in circuit complexity to charge an EEStore capacitor, I don't see it being any more efficient than LiIon. You just need a large pack to abosrb that many amps - and being an electric sports car, the Tesla has the amp capacity available in it's 6,800 cell battery pack.
The lighter weight helps, but I figure the best way for EEStor to win is on the basis of cost - everything else LiIon is competitive on, even if it ends up weighing 3 times as much. Heck, battery pack cost is the #1 killer of electronic vehicles today.
Hmmm... EEStore might be better off targeting the hybrid market for now than the EV market. Smaller, lighter, cheaper electric energy storage.
Given what I've seen of the Volt, there isn't any space to be taken away in the Cabin, really. There are women's purses that won't fit in what the Volt calls the 'trunk'. Then again, you don't expect to see women who carry those purses in a Volt either...
Shaving off nearly 500 pounds will still increase the range of the volt, though. It's not that heavy of a vehicle to begin with. Though it would need increased/more complex charging/discharge logic and equipment, so the savings might be closer to 400 pounds.
Eh... I'd like to see a PHEV light truck with two of them underneath the bed. No need for sandbags in the winter!
Of 'production' vehicles, the Tesla is currently the best of the pack for regenerative braking. But then, like you note, it's pack is designed more around capacity than amperage. The LiIon chemistry is also more efficient at taking a charge than NiMH.
For most hybrids, they're actually designed more for amperage - they want to keep the battery pack small and light, after all they still need to shove an engine in there. Also, because of said engine they can afford to be a bit lax on the regenerative braking. So they generally can't pull all the energy from an 'emergency brake', but can do a moderately hard stop from something like 40mph.
At the size needed to get those amperages, you get at least a few miles of capacity, so until we start shifting to plug in hybrids, they're good enough.
Actually, they weren't putting lead in the gas yet in the model t days. It was after that - when demand was rising and octanes falling that they started putting lead in.
By that very wiki article - TEL(Tetra-ethyl lead)'s antiknock properties were discovered in 1921.
The Model T ford was produced 1908 to 1927. So almost the entire run of Model T's would have had been designed to run on unleaded gas. Unleaded gas that - up to the '20s or so was actually pretty high octane because they could pick the part of the refinery stack with higher octanes for the gasoline(more t-butanol than n-pentane, for example). Later Model T's were built with lower compression engines due to the falling average octane ratings.
On a sidenote - man, talk about making a difference, they say that lead levels in blood have dropped by 75% since they banned TEL.
Modern cars do depend on a much higher octane rating than historical vehicles.
It's actually the opposite - Our cars are normally built to run on pretty low octane ratings today. We have to take a huge swath of the stack for gasoline to satisfy our demand for it, and the result is that our gas, knock wise, is pretty low.
From what I can find, the Model T ran on 93 octane. Not exactly what I'd call a low octane.
Early gas was actually pretty high octane(but tolerances weren't as tight); we didn't actually need all that much of it and it was still competing against Ethanol*, among other fuels. It was only later that gasoline demand started getting high enough that they started running short on the higher octanes, and needed to mix in lesser octane hydrocarbons.
One interesting fact i came across was that the Model T was Ford's original dual fuel vehicle - it featured manual spark advance control and could run on anything from 100% gasoline to 100% ethanol.
*During this time period, everything was competing. There were dozens of electric car companies; steam, ethanol, diesel, gasoline were all competing.
But even Nero has DRM.
It'd get quite annoying trying to keep track of all my games and all the distributers on steam. Stardock works closer to what you say, but I view it something like talking down a Heroin addict - you can't expect EA to back off 100% immediately.
As far as I know they've fixed most of those issues. Had a bit of problem with steam being a resource hog- but that seemed to be fixed sometime last year.
Not saying that I don't love them as well. What I meant by Second tier is that Steam gets games that get TV advertising and places on the shelf at every store that sells games. Stardock, doesn't.
The current exchange rate isn't helping right now either. Steam games are frequently cheaper at the moment, if you buy them on special. Also, steam games are the prettier visually. ;)
Correct. My Steam install is actually on D:.
But I'm running out of space on D:, so am looking at throwing another drive in and redoing the system a bit. It'd be nice to have the option to NOT have steam and all of the games on the same media.
It is the only authentication system that actually gives you something in return for authenticating your game, and it doesn't bitch about me having virtual drive software.
Well, there's also stardock
, but their game list is more second-tier, if you know what I mean.
Steam is winning on number of games in my list right now, mostly due to their offering package deals quite frequently, as well as the 75% off deals. Even though I have a physical CD somewhere, I might just buy their 75% off($5) Stalker deal right now.
Steam is DRM - its better, but still DRM.
But maybe it'll convince EA that at least over restrictive DRM IS an issue - and SECUROM, limited installs, complicated activation schemes and all that is the incorrect method to go about doing DRM.
Or maybe a correct wording would be 'you can't get something for nothing' - you CAN get consumers to accept DRM as long as you offer true advantages to go along with it. I happen to like the idea that even if my house is struck by a meterorite and everything is destroyed I'd be able to play my games again as soon as I got a new computer and an internet connection.
To put it another way, while I've given up the ability(for now) to lend/sell/give away my games, on the other hand I'm also able to play any games in the future with relatively no problems, simply by re-downloading it.
As the success of iTunes and such shows, the ability to redownload your purchases whenever you want often trumps the portability/sellability of physical mediums.
As all my friends are either non-gamers or get the same games anyways, it's not a big deal there, and I like occasionally playing an old game.
No media, no serial numbers, just a single username and password for all my games.
Free unlimited downloads, relatively automatic updates, etc... Though changing the install directory could be good.
I bought Crysis through the EA store download method as an experiment. While I captured the download file that should allow me to reinstall, I'm not sure I'd be able to today. With steam, that wouldn't be a problem.
I have to agree, I like steam. They manage to do online download gaming right.
Who on earth works in base 2?
Every major Operating System on the Planet? RAM manufacturers?
HD/SSD sizes are the odd ball out.
Oops, my bad. I was remembering the GB amount incorrectly.
Or your pricing information is about 6 months out of date. Somebody needs to keep some statistical information on the price history for SSD drives, and their platter counterparts to make some neat graphs - largest capacity available, cost for said largest capacity, size of 'sweet spot'* drive, cost per GB at the largest, and sweet spot.
Expand it out a bit further - track MLC and SLC seperately, as well as laptop vs desktop drives. Maybe server drives as well.
Hmm... Happened to have some price history from the 24th of november - HD price per gig dropped from $.10 to $.9. At the same time, the 'sweet spot' for SSDs dropped from $2.17 to $1.95. Right now it seems to take on-board RAM cache to get the best performance for writes - but let's be honest - how many HD's DON'T have significant amounts of cache today? The 1.5TB hd that has the best GB/Dollar has 32MB of cache.
*Best price per GB, or maybe where the price per GB flattens out. Eh, keep both. Call the smaller 'smallest economical drive'
Go back far enough and 2 MB would have cost you thousands. ;)
But yeah, the bleeding edge has always been expensive - and in multiple ways.
For example - The most of the current line of MLC flash drives suffers from poor write performance due to some flaws. The fix is looking to be some amount of RAM to cache the writes for speed. Many times the 'best', the newest, suffer from prototyping errors, things like overheating, etc...
Same deal with cars - my mechanic relatives have always advised against buying a new car the first year it's introduced or after major revisions.
Still, I haven't noticed anybody else mention this - the largest 2.5" laptop hard drive currently available is 500GB.
That means that, in 2009, flash memory will equal the largest 2.5" HD, the same form factor of SSD flash drives themselves. Maybe not in price - We're still looking at $110 for the hard drive vs $700-1400*. Going further in, I figure that a 20-25GB SSD 'drive' will cost ~$35 sometime next year. Even cheaper in bulk/wholesale. At that price I'd expect to see them start showing up as the sole drive for smallest notebooks - even/especially economy models, built onto the very motherboard, perhaps with an empty drive slot to take an upgrade HD for those who need more storage.
*Going by the current $700 for a 250GB OCZ Solid Series OCZSSD2-1SLD250G 2.5" 250GB SATA II Internal Solid state disk (SSD) - Retail. History has shown that increased memory sizes are generally accompanied with similiar decreases in price per unit such that the top generally stays in the same price range. So we're likely to see closer to double the GB per dollar instead of a doubling in total price.
Actually, I'd peg the requirements for a 'natural' monopoly to be a bit different.
People want different things from their last mile connections. Some people merely want voice or voice+video service. Some people want on-demand content. Some people want hagh transmission capacity. Some people only want data... Etc..
And a single pair of fiber, or a few more pairs of copper can provide all of this.
If the diversity of customer needs is high enough (it should be), the second assumption also falls. If each of the last mile providers can attract a significant portion of the market, they should each be able to achieve a customer base large enough to bring down the costs of the network that would need to be passed on to the individual users.
The other problem is that you don't get significant economies of scale when it comes to last mile installs. It can cost hundreds to tens of thousands to run a data line to a single house. House N+1 still costs thousands. You have to mark and existing utility lines, repair concrete disturbed, install service equipment, run the main line, etc...
Makes me kinda wish we had had a bright idea and installed conduit in our cities. I recognize that it wouldn't work for water/sewer, but you'd be able to run local power, cable, telephone, and other data lines through them relatively cheaply. Access to the shared service boxes(and right to pull cable through them) would be controlled by the local government.
And lastly, proof that this is the case. We currently have a situation where most communities have multiple last mile providers with overlapping services. One or more cable companies, and an incumbent telephone company. Both of which can justify upgrading their last-mile networks to the point where it's essentially a complete rebuild. This would be impossible if the last mile were a natural monopoly.
They're converging as providers, and a lot of the reason there's two of them is due to regulations not allowing cable companies to provide phone service, and technical problems with running cable over twisted pair for the telephone companies.
They're both already established - due to the cost of running a wire infrastructure, both can rather easily crush any competitors who have to duplicate that.
"The government may now take steps to legislate so that Telstra can't build a network that competes with the NBN"
Now, I'll admit that I'm not that familiar with Australian legislative processes, but from my experience with the US one, wouldn't they be able to do this no matter what's going on, assuming they got a big enough bug up their butt?
Going on, it sounds like I should be glad to not have to deal with Telstra. Deutsch Telecom was bad enough when I was in Germany. Now I'm with a local coop, and love them.
Most of those types get the 'free phone with x year contract', and are happy enough - they don't travel.
Ah, but therein lies the problem. The shuttle estimates weren't particularly egregious, either. F(known cost, unknown cost) = [True cost], and in the shuttles case we now know both the known and unknown costs. However, in the 'ideal' case we're comparing against, we don't know the unknown costs. Sometimes even hindsight isn't 20/20.
Basically, it's two shuttle services to equal the TOTAL cost of the hubble construction(basically they removed the groundside expenses). Going by spare construction and whatnot, about 50% of the cost of the satellite was R&D, not construction. They ended up with a 'construction line' cost of ~$400 million, dropping the more they build, and improvements not being much more. Around $30million to launch it on a rocket.
Today, we should be able to do quite extensive servicing by way of robots and teleoperation - far cheaper than shuttles.
Oops... My bad.
For some reason I saw the first and last company, despite reading that line three times.