There is another transmissible cancer, in dogs that is an STD. Luckily for the dogs, it is substantially less aggressive than the Tasmanian Devil cancer, and while it spreads reasonably readily(estimates of the length of the cell line vary; but it is definitely the oldest living dog in the world, for certain values of 'dog') the host's immune system typically controls it well enough that it causes only minor symptoms or is asymptomatic.
There have been a few once-off transmissions of cancer in humans; but no (known) ones under 'natural' conditions. An improperly screened donor organ, followed by immunosuppressants? Sure. Surgeon accidentally cutting himself and tumor cells from the patient getting into the wound? I believe that that has been documented; but no known in-the-wild transmission of actual cancer cells.
The details are easier to come by for the oldest gear(since they've had the most time to diffuse and in some cases have been formally declassified because they are obsolete); but there are a lot of spy satellite launches.
KH-1's had 10 launches, KH-2 7, KH-3 9, KH-4 26, KH-4A 56, KH-4B 17, KH-5 12, KH-6 3, KH-7 38, KH-8 54, KH-9 20. KH-10 didn't go anywhere; but kindly donated 6 72-inch optical mirrors to the Multiple Mirror Telescope. KH-11 superceded KH-10 and saw 16 launches. Information about 'Misty' and 'Enhanced Imaging System' is new enough to be... spotty. And those are just the ones with some substantial optical component that makes them a crossover with astronomy.
One thing to note, in fairness, is that pre-KH-11, these satellites actually used film, with various re-entry pods to send it back to earth for processing, so their missions were of necessarily limited duration. Only the more recent digital imaging satellites even have the option of long term operation, with mission lifespans depending on what orbit you want, whether anything goes wrong, fuel supply, and so on.
Wasn't the case of the Hubble a slightly less severe version of the same thing? If memory serves, it was substantially derived from the KH-11; and its spook origins ended up being the reason that the PerkinElmer got the job to produce the mirror, having done so for the KH-9s; and ended up beating out Kodak's(actually correctly shaped) mirror.
The dense storage mechanism proposed in the paper is significantly different, since it depends on manipulating birefringence; but the proposed 'just write some directions into the storage disk' part of the plan would be more or less identical, just on a sturdier medium.
Man, I kind of miss microfilm. Clunky; but you still run across items available on microfische and not in digital formats from time to time.
I don't know how much of a barrier there will be, I just wanted to be careful to emphasize that my post only covered the problem of dealing with legacy storage media, not linguistic issues.
My wild guess is that (barring some cataclysm that makes prediction basically futile) whether or not they'll call the most English-like thing spoken in 4016 'English' or not will depend more on political continuity than on its exact properties(nation states love to have a pet language, so even minor regional differences can be dressed up into 'languages' and even serious barriers to mutual intelligibility can be handwaved as 'dialects' depending on where the borders are); but given the turnover of English vocabulary, a layman might well find something written in today's English familiar but not all that readable. It would be a surprise if you couldn't find a suitable specialist, though, much as it's not a huge challenge to find a classicist today.
There are some undeciphered texts out there; but some of the main causes of difficulty are scarcity and lack of context. It's not clear that you can really resurrect 'fluency' in a language that has been solidly killed off with only a written record to go on; but the larger the supply of texts and, ideally, the larger the supply of texts including multiple languages, pictures, accounting systems, periodic tables, etc. the better off you are.
Hard to say how much the future will care; but they'll have a much easier time of it if digging up one box of these things provides them with several hundred thousand pages of text, with translations into multiple contemporary languages where available, than we do when looking at something like Linear A, for which we have under 2,000 known specimens of any length(most of them short). Team Linguistics can get fairly clever when they have something to work with; but the archaeological record really, really, sucks in some cases.
It's considered cheating; but you can also ensure that the backups always outlast the users by 'retiring' any user whose backup media are starting to show signs of flakiness. The side benefit is the steep reduction in the number of people asking you to pull something from backups for them.
Unless something I'm not thinking of forbids this, I'd imagine that having the ability to produce durable structures small enough to be useful for bulk data storage would also allow you to build larger structures that are visible to the naked eye or under various levels of magnification, at the expense of data density.
This doesn't solve the rather nasty tech-writer challenge of trying to compose an instruction manual for a reader-of-the-language-in-use-2000-years-from-now; but it would allow you to provide multiple 'stages' of readable data with various trade-offs between storage capacity and intelligibility. Text large enough to be obvious and readable with the naked eye would be inefficient; but hard to miss. Text large enough to require modest magnification to actually read; but look patterned enough to be worth investigating to the naked eye could easily crunch several paragraphs into a reasonably modest space(microfilm/microfiche scale, say). Text invisible to the naked eye; but readable without any fancy polarization tricks and just an optical microscope could be denser still; and finally the technique described could be used for bulk data storage.
Doesn't solve the language barrier; but it would allow you to do some amount of self-documenting of the format, starting with a visible 'README', and proceeding down through one or more layers of less densely packed data describing how to interpret the more densely packed layer beneath, and finally the data area.(which we would presumably encrypt and tie to a DRM system that was nuked to ashes millenia ago; because what's a good technological advance without some self defeating stupidity?)
The guy is absolute ruler and head of a cult of personality in one of the most insular walled, er, 'gardens' on earth; so why wouldn't he be a Mac user?
How can this feature exist when 'important tweet' is somewhere between 'theoretical' and 'logically impossible', and definitely not easy to find in the wild?
I suspect that Instagram simply realizes that 'different accounts' is a security/visibility-control model that people find easier and more familiar than various sorts of filters/tags/groups/'friends only'/etc. It's not as though they will have much difficulty correlating a user's accounts(even if the app doesn't explicitly send them 'all usernames on this device', seeing logins to certain accounts from a specific device is a pretty big clue); but switching accounts is easier than futzing with security settings when trying to maintain distinct audiences.
In Apple's defense, it does seem reasonably plausible that the biometric sensor widget built into the 'home' button(and quite possibly the cable connecting the home button to the logic board) is a 'trusted' element of the system, in the 'the integrity of the system depends on this part performing as expected and not being malicious' sense of 'trusted'. So, I can see why it would be impossible or prohibitively difficult to keep the biometric authentication feature secure while also allowing random people to swap random hardware in to that part of the system.
However, what is a lot less clear is why(especially when many iDevices, including current-model ones, simply lack this feature entirely) 'security' demands that the entire phone be bricked, rather than just the biometric features flushing any private storage associated with them and leaving the phone usable as though it were a model without that feature. This might involve wiping all locally stored data, if the device encryption keys are tangled up with the biometric authentication feature's private storage; but it should still be able to function as though you had just restored it to defaults.
This also raises the question of whether, with the correct incentives, it is possible to induce authorized repair services to introduce malicious components when doing these repairs, and whether doing so would allow you to extract highly sensitive information. Since Apple-blessed repairs can apparently fix home buttons without destroying the handset, and since Apple's line is that tampering threatens the integrity of the authentication system, this seems like a natural place to try to get your malicious part introduced: much more likely that an authorized repair outfit exists in your jurisdiction than that Apple Inc. does; many more low-level techs you could potentially lean on; and home button repairs are a pretty common service request...
Depends on when you start the scenario: it's actually a very morning-person assumption that the day will begin when you wake up; rather than everything being over by the time yesterday's nigh owls have decided to call it a day and go to sleep.
Among the little people and the petty criminals; 'invoice fraud' is a classic. You just pump out a whole bunch of reasonably plausible looking invoices for suitably generic goods or services, and hope that some of the recipients pay without checking too closely. Illegal, of course. Exactly how much 'unfair billing' and how many 'errors'(mysteriously in your favor much more often than not) do you have to accrue before people stop cringing and call your practices what they are, when not pulled by giant oligopolies?
They could(and, as best I can tell, they already do to some degree when planning buildouts and upgrades, at least in areas that aren't so oligopolistic that market forces simply don't apply). It just wouldn't have much effect on whether or not they are adhering to 'net neutrality' while doing so.
If you are willing to make a longer term commitment to buying some given allotment of bandwidth every month, you usually pay somewhat less per unit for it than if you prefer the flexibility of a pay as you go/no ETF/no contract arrangement. If a telco judges a given area to be a likely-reliable buyer of service, they are more likely to build out there; while if you want a remote facility or some unusual arrangement set up they may refuse or have you eat more or less the entire install cost to run a line out there.
The 'neutrality' isn't in treating different customers identically; but in exploiting your ability, as man in the middle, to distort things in your favor by billing differently depending on what they are doing with the bandwidth(or in this case, the electricity) you sell them. If they billed 'HPC kilowatt hours' differently from 'bitcoin hashing kilowatt hours'; that would be distinctly non-neutral. If their observation is "Very high density customers pay more; because they have the nasty habit of sometimes demanding enough to require expensive buildouts; and sometimes going more or less entirely dark", that's not unlike prepaid users with nonexistent credit scores paying more per minute than people on 2 year contracts.
The issue is demand volatility: when you incur a large capital cost to build a generating unit, you need to set the price such that you cover operating expenses and recover the capital cost before the end-of-life of the unit.
If your customers are 100% predictable, there is room for squabbling about how much profit you get(and added complexity because the time value of money may change depending on conditions in other markets); but it is relatively simple to set a price that meets this goal.
If there is a nontrivial risk that a source of demand may arrive, require a new build-out, and then vanish relatively quickly; you'll lose most of your initial investment unless you set rates to recover that investment over a shorter timespan.
Consider the two (largely hypothetical, but convenient) limit cases: if you want to buy a new power plant, nobody will sell for less than the amount of money it costs to build it. If you are buying power from a plant with perfectly stable demand and an unlimited lifespan, your rate would closely approach the cost of production as the initial investment can easily be recovered.
In real life, obviously, no source of demand is 100% risk free; and utility customers are not asked to pay 100% of the price of the infrastructure up front; but different sorts of customers are more and less risky(both in that they, individually, will leave unexpectedly; and more importantly that they and everyone like them might experience a highly correlated change in demand and leave all at once without replacement).
For not terribly shocking reasons, this utility suspects that bitcoin miners are (a)risky and (b) likely to enter or exit the market in large groups, unpredictably. Depending on what the price of bitcoins does, miners can either demand as much electricity as you can deliver to them, or potentially shut down everything but the emergency lights in a matter of minutes to hours if mining becomes uneconomic.
It's not that they care what you use the electricity for, it's that they care how likely you are to be a predictable customer. It's like why getting a hotel room for a night is more expensive, per hour, than getting an equivalent apartment for a year: it's not that the sellers care what you are doing with the room; but they do care about the odds that they'll have a paying customer for it on any given day.
Last I checked, the only thing more popular than crying when your subsidy gets cut is coming up with rationalizations for why your subsidy is morally justified and the other guy's is just parasitic laziness.
This seems like an issue of how you want to allocate the costs of risk, not a terribly uncommon problem: Building the additional capacity will cost the utility a nontrivial amount of money, and if the demand that originally justified the buildout dries up, they won't exactly be able to return it for a refund(and, if they can't operate it profitably, its resale value is unlikely to be very exciting).
Unless one simply wishes to deny that, and pretend that this sort of capital investment is risk free, which is silly; the question is really just how the cost of the risk is paid: If you want the utility to bear the risk, giving you the ability to purchase or not purchase power from month to month as you see fit; they'll want to make up the cost of the risk by increasing the price. If you offer to take on the risk; but making a long-term commitment to purchasing a given amount of power, I'm sure they'd be happy to offer you a suitably lower rate.
This is only 'discriminatory' if, in fact, 'bitcoin businesses' are not a more volatile and hard to predict customer base than other electricity users; but the utility is just treating them as though they are. If they are in fact more unpredictable, it is only reasonable that the utility would want them to pay more: the rate you pay is basically their operating costs, plus the cost of the initial investment in building the generating capacity. If you are highly predictable, they'll be content to be paid back for that over the long term. If you might be gone in six months without a replacement, they need to be repaid faster. Not fundamentally different from paying more for credit if you are considered a lousy repayment risk.
I don't know how awful the situation has to get before Microsoft has an incentive to step in and write a device driver; but I would (perhaps naively) think that they would take a very, very, hard line on allowing anyone to use Windows Update to distribute drivers that make the Windows user experience look worse, especially if they are doing it intentionally, rather than being not-quite-careful-enough with some monstrously complex GPU driver or something.
FTDI can do whatever they think they can get away with on FTDI.com; but WU is something that MS operates to make its OS more appealing and pleasant for users, not to save OEMs from having to provide support pages, so if an OEM is being a bad actor, I would have expected them to get the shove.
What I'd be curious to know is how FTDI managed to pull this again. I would have imagined that Microsoft would have been less than pleased with them after their last attempt and either watching them more carefully or only letting them back with some sort of stern warning. One would certainly think that it would hurt FTDI more than it hurts Microsoft if FTDI chips become 'those ones you have to manually download drivers for'.
I'm all for charitably providing access, given the trivial cost. What rubs me very, very, much the wrong way is being shoved(with varying degrees of force) into providing uncompensated location and power for my friendly local ISP oligopoly.
Aside from being as little a matter of choice as they can make it, you'll notice that these secondary hotspots aren't being run as a public service; but with captive portals and subscriber sign-ins.
I have, and do, offer an open wifi channel(QoS ranked below anything I want to do, obviously); but I'll be damned if I get to pay for infrastructure that my ISP is too cheap to build out themselves.
Slashdot Mirror
There is another transmissible cancer, in dogs that is an STD. Luckily for the dogs, it is substantially less aggressive than the Tasmanian Devil cancer, and while it spreads reasonably readily(estimates of the length of the cell line vary; but it is definitely the oldest living dog in the world, for certain values of 'dog') the host's immune system typically controls it well enough that it causes only minor symptoms or is asymptomatic.
There have been a few once-off transmissions of cancer in humans; but no (known) ones under 'natural' conditions. An improperly screened donor organ, followed by immunosuppressants? Sure. Surgeon accidentally cutting himself and tumor cells from the patient getting into the wound? I believe that that has been documented; but no known in-the-wild transmission of actual cancer cells.
The details are easier to come by for the oldest gear(since they've had the most time to diffuse and in some cases have been formally declassified because they are obsolete); but there are a lot of spy satellite launches.
KH-1's had 10 launches, KH-2 7, KH-3 9, KH-4 26, KH-4A 56, KH-4B 17, KH-5 12, KH-6 3, KH-7 38, KH-8 54, KH-9 20. KH-10 didn't go anywhere; but kindly donated 6 72-inch optical mirrors to the Multiple Mirror Telescope. KH-11 superceded KH-10 and saw 16 launches. Information about 'Misty' and 'Enhanced Imaging System' is new enough to be... spotty. And those are just the ones with some substantial optical component that makes them a crossover with astronomy.
One thing to note, in fairness, is that pre-KH-11, these satellites actually used film, with various re-entry pods to send it back to earth for processing, so their missions were of necessarily limited duration. Only the more recent digital imaging satellites even have the option of long term operation, with mission lifespans depending on what orbit you want, whether anything goes wrong, fuel supply, and so on.
No need for that; aliens are only a concern if they get too close to our border...
Wasn't the case of the Hubble a slightly less severe version of the same thing? If memory serves, it was substantially derived from the KH-11; and its spook origins ended up being the reason that the PerkinElmer got the job to produce the mirror, having done so for the KH-9s; and ended up beating out Kodak's(actually correctly shaped) mirror.
The dense storage mechanism proposed in the paper is significantly different, since it depends on manipulating birefringence; but the proposed 'just write some directions into the storage disk' part of the plan would be more or less identical, just on a sturdier medium.
Man, I kind of miss microfilm. Clunky; but you still run across items available on microfische and not in digital formats from time to time.
I don't know how much of a barrier there will be, I just wanted to be careful to emphasize that my post only covered the problem of dealing with legacy storage media, not linguistic issues.
My wild guess is that (barring some cataclysm that makes prediction basically futile) whether or not they'll call the most English-like thing spoken in 4016 'English' or not will depend more on political continuity than on its exact properties(nation states love to have a pet language, so even minor regional differences can be dressed up into 'languages' and even serious barriers to mutual intelligibility can be handwaved as 'dialects' depending on where the borders are); but given the turnover of English vocabulary, a layman might well find something written in today's English familiar but not all that readable. It would be a surprise if you couldn't find a suitable specialist, though, much as it's not a huge challenge to find a classicist today.
There are some undeciphered texts out there; but some of the main causes of difficulty are scarcity and lack of context. It's not clear that you can really resurrect 'fluency' in a language that has been solidly killed off with only a written record to go on; but the larger the supply of texts and, ideally, the larger the supply of texts including multiple languages, pictures, accounting systems, periodic tables, etc. the better off you are.
Hard to say how much the future will care; but they'll have a much easier time of it if digging up one box of these things provides them with several hundred thousand pages of text, with translations into multiple contemporary languages where available, than we do when looking at something like Linear A, for which we have under 2,000 known specimens of any length(most of them short). Team Linguistics can get fairly clever when they have something to work with; but the archaeological record really, really, sucks in some cases.
It's considered cheating; but you can also ensure that the backups always outlast the users by 'retiring' any user whose backup media are starting to show signs of flakiness. The side benefit is the steep reduction in the number of people asking you to pull something from backups for them.
Unless something I'm not thinking of forbids this, I'd imagine that having the ability to produce durable structures small enough to be useful for bulk data storage would also allow you to build larger structures that are visible to the naked eye or under various levels of magnification, at the expense of data density.
This doesn't solve the rather nasty tech-writer challenge of trying to compose an instruction manual for a reader-of-the-language-in-use-2000-years-from-now; but it would allow you to provide multiple 'stages' of readable data with various trade-offs between storage capacity and intelligibility. Text large enough to be obvious and readable with the naked eye would be inefficient; but hard to miss. Text large enough to require modest magnification to actually read; but look patterned enough to be worth investigating to the naked eye could easily crunch several paragraphs into a reasonably modest space(microfilm/microfiche scale, say). Text invisible to the naked eye; but readable without any fancy polarization tricks and just an optical microscope could be denser still; and finally the technique described could be used for bulk data storage.
Doesn't solve the language barrier; but it would allow you to do some amount of self-documenting of the format, starting with a visible 'README', and proceeding down through one or more layers of less densely packed data describing how to interpret the more densely packed layer beneath, and finally the data area.(which we would presumably encrypt and tie to a DRM system that was nuked to ashes millenia ago; because what's a good technological advance without some self defeating stupidity?)
The guy is absolute ruler and head of a cult of personality in one of the most insular walled, er, 'gardens' on earth; so why wouldn't he be a Mac user?
How can this feature exist when 'important tweet' is somewhere between 'theoretical' and 'logically impossible', and definitely not easy to find in the wild?
I suspect that Instagram simply realizes that 'different accounts' is a security/visibility-control model that people find easier and more familiar than various sorts of filters/tags/groups/'friends only'/etc. It's not as though they will have much difficulty correlating a user's accounts(even if the app doesn't explicitly send them 'all usernames on this device', seeing logins to certain accounts from a specific device is a pretty big clue); but switching accounts is easier than futzing with security settings when trying to maintain distinct audiences.
In 2016 it is 'news' when a client application supports storing more than one set of credentials?
Truly, what miracles 'mobile' hath wrought. Wherever would we be without such innovators?
In Apple's defense, it does seem reasonably plausible that the biometric sensor widget built into the 'home' button(and quite possibly the cable connecting the home button to the logic board) is a 'trusted' element of the system, in the 'the integrity of the system depends on this part performing as expected and not being malicious' sense of 'trusted'. So, I can see why it would be impossible or prohibitively difficult to keep the biometric authentication feature secure while also allowing random people to swap random hardware in to that part of the system.
However, what is a lot less clear is why(especially when many iDevices, including current-model ones, simply lack this feature entirely) 'security' demands that the entire phone be bricked, rather than just the biometric features flushing any private storage associated with them and leaving the phone usable as though it were a model without that feature. This might involve wiping all locally stored data, if the device encryption keys are tangled up with the biometric authentication feature's private storage; but it should still be able to function as though you had just restored it to defaults.
This also raises the question of whether, with the correct incentives, it is possible to induce authorized repair services to introduce malicious components when doing these repairs, and whether doing so would allow you to extract highly sensitive information. Since Apple-blessed repairs can apparently fix home buttons without destroying the handset, and since Apple's line is that tampering threatens the integrity of the authentication system, this seems like a natural place to try to get your malicious part introduced: much more likely that an authorized repair outfit exists in your jurisdiction than that Apple Inc. does; many more low-level techs you could potentially lean on; and home button repairs are a pretty common service request...
Depends on when you start the scenario: it's actually a very morning-person assumption that the day will begin when you wake up; rather than everything being over by the time yesterday's nigh owls have decided to call it a day and go to sleep.
If it's so genetic, I'll agree to drop the hatred and adopt an attitude of dispassionate eugenics. Happy now, Mr. Sunshine?
This should be quite a bloodbath; but the satisfying kind of bloodbath, where the guilty are cut down in swaths.
Among the little people and the petty criminals; 'invoice fraud' is a classic. You just pump out a whole bunch of reasonably plausible looking invoices for suitably generic goods or services, and hope that some of the recipients pay without checking too closely. Illegal, of course. Exactly how much 'unfair billing' and how many 'errors'(mysteriously in your favor much more often than not) do you have to accrue before people stop cringing and call your practices what they are, when not pulled by giant oligopolies?
They could(and, as best I can tell, they already do to some degree when planning buildouts and upgrades, at least in areas that aren't so oligopolistic that market forces simply don't apply). It just wouldn't have much effect on whether or not they are adhering to 'net neutrality' while doing so.
If you are willing to make a longer term commitment to buying some given allotment of bandwidth every month, you usually pay somewhat less per unit for it than if you prefer the flexibility of a pay as you go/no ETF/no contract arrangement. If a telco judges a given area to be a likely-reliable buyer of service, they are more likely to build out there; while if you want a remote facility or some unusual arrangement set up they may refuse or have you eat more or less the entire install cost to run a line out there.
The 'neutrality' isn't in treating different customers identically; but in exploiting your ability, as man in the middle, to distort things in your favor by billing differently depending on what they are doing with the bandwidth(or in this case, the electricity) you sell them. If they billed 'HPC kilowatt hours' differently from 'bitcoin hashing kilowatt hours'; that would be distinctly non-neutral. If their observation is "Very high density customers pay more; because they have the nasty habit of sometimes demanding enough to require expensive buildouts; and sometimes going more or less entirely dark", that's not unlike prepaid users with nonexistent credit scores paying more per minute than people on 2 year contracts.
The issue is demand volatility: when you incur a large capital cost to build a generating unit, you need to set the price such that you cover operating expenses and recover the capital cost before the end-of-life of the unit.
If your customers are 100% predictable, there is room for squabbling about how much profit you get(and added complexity because the time value of money may change depending on conditions in other markets); but it is relatively simple to set a price that meets this goal.
If there is a nontrivial risk that a source of demand may arrive, require a new build-out, and then vanish relatively quickly; you'll lose most of your initial investment unless you set rates to recover that investment over a shorter timespan.
Consider the two (largely hypothetical, but convenient) limit cases: if you want to buy a new power plant, nobody will sell for less than the amount of money it costs to build it. If you are buying power from a plant with perfectly stable demand and an unlimited lifespan, your rate would closely approach the cost of production as the initial investment can easily be recovered.
In real life, obviously, no source of demand is 100% risk free; and utility customers are not asked to pay 100% of the price of the infrastructure up front; but different sorts of customers are more and less risky(both in that they, individually, will leave unexpectedly; and more importantly that they and everyone like them might experience a highly correlated change in demand and leave all at once without replacement).
For not terribly shocking reasons, this utility suspects that bitcoin miners are (a)risky and (b) likely to enter or exit the market in large groups, unpredictably. Depending on what the price of bitcoins does, miners can either demand as much electricity as you can deliver to them, or potentially shut down everything but the emergency lights in a matter of minutes to hours if mining becomes uneconomic.
It's not that they care what you use the electricity for, it's that they care how likely you are to be a predictable customer. It's like why getting a hotel room for a night is more expensive, per hour, than getting an equivalent apartment for a year: it's not that the sellers care what you are doing with the room; but they do care about the odds that they'll have a paying customer for it on any given day.
Last I checked, the only thing more popular than crying when your subsidy gets cut is coming up with rationalizations for why your subsidy is morally justified and the other guy's is just parasitic laziness.
This seems like an issue of how you want to allocate the costs of risk, not a terribly uncommon problem: Building the additional capacity will cost the utility a nontrivial amount of money, and if the demand that originally justified the buildout dries up, they won't exactly be able to return it for a refund(and, if they can't operate it profitably, its resale value is unlikely to be very exciting).
Unless one simply wishes to deny that, and pretend that this sort of capital investment is risk free, which is silly; the question is really just how the cost of the risk is paid: If you want the utility to bear the risk, giving you the ability to purchase or not purchase power from month to month as you see fit; they'll want to make up the cost of the risk by increasing the price. If you offer to take on the risk; but making a long-term commitment to purchasing a given amount of power, I'm sure they'd be happy to offer you a suitably lower rate.
This is only 'discriminatory' if, in fact, 'bitcoin businesses' are not a more volatile and hard to predict customer base than other electricity users; but the utility is just treating them as though they are. If they are in fact more unpredictable, it is only reasonable that the utility would want them to pay more: the rate you pay is basically their operating costs, plus the cost of the initial investment in building the generating capacity. If you are highly predictable, they'll be content to be paid back for that over the long term. If you might be gone in six months without a replacement, they need to be repaid faster. Not fundamentally different from paying more for credit if you are considered a lousy repayment risk.
I don't know how awful the situation has to get before Microsoft has an incentive to step in and write a device driver; but I would (perhaps naively) think that they would take a very, very, hard line on allowing anyone to use Windows Update to distribute drivers that make the Windows user experience look worse, especially if they are doing it intentionally, rather than being not-quite-careful-enough with some monstrously complex GPU driver or something.
FTDI can do whatever they think they can get away with on FTDI.com; but WU is something that MS operates to make its OS more appealing and pleasant for users, not to save OEMs from having to provide support pages, so if an OEM is being a bad actor, I would have expected them to get the shove.
What I'd be curious to know is how FTDI managed to pull this again. I would have imagined that Microsoft would have been less than pleased with them after their last attempt and either watching them more carefully or only letting them back with some sort of stern warning. One would certainly think that it would hurt FTDI more than it hurts Microsoft if FTDI chips become 'those ones you have to manually download drivers for'.
I'm all for charitably providing access, given the trivial cost. What rubs me very, very, much the wrong way is being shoved(with varying degrees of force) into providing uncompensated location and power for my friendly local ISP oligopoly.
Aside from being as little a matter of choice as they can make it, you'll notice that these secondary hotspots aren't being run as a public service; but with captive portals and subscriber sign-ins.
I have, and do, offer an open wifi channel(QoS ranked below anything I want to do, obviously); but I'll be damned if I get to pay for infrastructure that my ISP is too cheap to build out themselves.