The one interesting quirk about being directly based on a video codec, though, is that a variety of devices (eg. SoCs with fixed function decode blocks, computers with OS vendors that pay the appropriate tithes, etc.) might be 'automatically' licensed for the special-case use of treating a single HEVC frame as a still image by virtue of being suitably licensed for HEVC playback. It'd be amusing to see whether that argument is accepted, or whether 'single frame playback' will mysteriously require a different set of licenses.
Not a patent minefield I want to wade into, though, though I do very much appreciate the work of those who ensure that we have access to solid implementations of these standards, even if it is far beyond their power to untangle the patent nightmare.
Apropos of that review, it would be interesting to see how it stacks up against NAS4free at this point. I'm currently running NAS4Free, after a period on FreeNAS, and my impression largely mirrored Arstechnica's.
NAS4Free isn't terribly shiny, and lacks some features that would be kind of neat to have; but (as one might hope for a NAS) its file server stuff is about as solid, predictable, and simple as it reasonably can be (which isn't always simple, should you delve into the more exiting waters of heavy AD integration, ZFS-Fu, or both at the same time; but it's dead simple to set up a basic NAS and is thankfully free of the "magic GUI promises to make it easy, utterly trashes things!" issue.
If FreeNAS is improving on some of its eccentricities with each release it will likely overtake NAS4Free at some point, since it'll still have all the features and the reliability; but NAS4Free has also had a few little bits grafted on over time (mostly additional quasi-NAS protocol support, DLNA, whatever iTunes uses, etc.)
CDNs also have that 'comparative advantage' thing going for them. Damned if I know how they do it; but the relevant specialists can keep a given chunk of data online for preposterously tiny amounts of money, especially, compared to what it'd cost me to do relatively low-volume hosting myself. At sufficiently large scales, farming out something like that is increasingly likely to be a false economy; but you need to be slinging a nontrivial amount of data(or tolerant of severely dubious backup and redundancy) before you can beat the people who do nothing but that for a living.
That, and potentially somebody popping out of the woodwork with a patent claim, however dodgy.
For a free format without any sponsors(either altruistic or, as with WebM, probably doing it as a tactical move to improve their negotiating position if the MPEG LA ever decided to shake down youtube for more than token fees for H264), it doesn't take a terribly plausible patent claim to be plausible enough that it'll be hard to find somebody to go slug it out in court.
As much as certain formats are...getting rather elderly...it certainly is handy to be assured through sheer age that any patents that might apply to a given format have expired.
With 'mobile' there are really two considerations: One is the fact that 'mobile' (even if the fault is, in fact, with shitty backhaul) is going to be fairly slow in emerging markets. Two, relevant even in wealthy developed nations without asshole oligopoly telcos, is the fact that mobile devices are brutally power constrained, and RF chatter isn't exactly cheap in energy terms. The faster you get the data you need and shut down, or move to a slower, lower power mode, the radios, the happier your battery will be.
With mains power it matters less (electricity isn't free; but a few extra dollars a month is far less annoying than having your battery keel over dead at a bad time); but barring exciting breakthroughs in battery chemistry or design, basically all the savings are going to have to come from the device side.
Sometimes you can 'fix' a format (maybe I'm just dating myself; but MP3s made impressive strides in apparent quality at a given bitrate in the years after their introduction); but you run into a major constraint:
If a spec-compliant; but not otherwise updated, decoder chokes on your 'improvement', you don't really have an improvement, you have a new format derived from the old format. If there is room to improve the encoder, while still producing something that existing decoders consider valid, great, adoption will likely be easy; but if decoder behavior prevents you from doing what you want to make some improvement or other, sometimes you just need a new format.
JPEG is (barring the possibility of some lossless mode that looks very little like JPEG except for a few metadata fields; but is technically part of the spec, not sure if we have one of those) indeed compressed; but it's lossy compression and lossy compression is an area where there is actually a reasonable amount of ongoing development.
This isn't to say that lossless compression is a trivial problem, or that there have been absolutely no improvements; but the 'by definition, it isn't lossless unless applying the decompression operation to the result of the compression operation produces something identical to the input' criterion makes it much easier to let the mathematicians and computer scientists work out the limits of the possible.
With lossy compression, there aren't any formal limits, which leaves the field much more open to solutions that rely on following the strong and weak points of human perception(visual in this case, auditory in other cases, visual/motion related in others), which leads to much greater complexity and diversity.
Took me five years as a teenager to master the Sargon II chess game for the Commodore 64 on the hardest difficulty level. I'll like to see a four-year-old do that in less time.
I haven't played it myself; but they say that Robot Odyssey will either break your pitiful hominid brain like reject before The Monolith, turn you into a hardcore programmer geek for life, or turn you against any computer game that isn't Medal of Halo Gears of Assault 3.
Protocol IV of the 1980 Convention on Certain Conventional Weapons is unlikely to be a problem.
Per Article 1, weapons specifically designed as their sole combat function, or one of their functions, to cause permanent blindness are Not OK.
However, Per Article 3, "Blinding as an incidental or collateral effect of the legitimate military employment of laser systems, including laser systems used against optical equipment, is not covered by the prohibition of this Protocol." Just aim for a legitimate target and stock up on braille sympathy cards.
Problem solved.
As for making war with fire, light isn't fire, and conventional explosives(never mind thermobarics and incendiaries) are markedly more strongly associated with fire. Lasers have that novelty thing going against them; but anybody who actually cares about the letter of the law probably has hangups about tracers, attacks on fuel dumps, and other routine stuff. As soon as the novelty wears off lasers will recede into the background.
What remains to be seen is whether jets and missiles can shrug off (either through brute-force thickening or more sophisticated ablative armor or actively deployed particulates that effectively scatter incoming light) the relatively tepid amounts of energy that lasers (especially anything that dodges the rather nasty requirements of chemical lasers) are good for, particularly at range, under optically sub-optimal conditions (never have those at sea!).
Even under the ideal and closely controlled conditions of industrial laser cutters, lasers are abundantly unsafe for ocular exposure; but by no means the speediest remover of bulk material. In an atmosphere, range is going to be constrained by thermal distortion if nothing else, so the ease of keeping photons on target won't be quite as dramatic as it would be in space, and against close-in non-aircraft, there'll be a lot of cheap 'n nasty (but probably embarrassingly effective) countermeasures involving coating things with mud, spraying them with seawater, and generally making a 3rd world nuisance of yourself.
(By way of comparison, assuming that this 30Kw laser delivers 100% of energy to target, it'll take 2/3 of a second of continuous exposure to deliver the same number of joules to the target as a.50 BMG. Now, if you can't put a bullet on target, that's irrelevant; but in terms of expected stopping power the finest in combat laser technology is...distinctly middling... compared to guns that date back to the period between the world wars. Obviously the fire control system has evolved out of sight; but given how long it'll have to stay on target, you'd hope so.)
It doesn't help that, even on the good days when they aren't running malware, ad 'networks' long ago stopped just farming out ads and started gathering impression data, trying to correlate those data across multiple sites that used the same(or related by business in some way) networks, and generally watching you.
The TV-style bullshit that just leaps for your eyes and blares at you is insufferable; but at least it only watches you at the level of granularity provided by the Neilson lab rats.
If the online advertisers aren't watching you every step of the way, it's certainly not for lack of trying. Fuck 'em.
I'm never quite sure if it's mere hyperbole or raging entitlement that causes people to breathlessly use the language of actual oppressive activity to describe people who are being 'punished' in ways that most of us will never have the pleasure of being treated.
"Blacklisting" is, from time to time, an actual labor market practice, and it stings a lot more than being quietly retired to a cushy honorary position. As in, y'know, literally never working in this town (or larger area) again, prolonged unemployment, penury, that sort of fun stuff. Do people just not know or care about what the real thing looks like, or do they think themselves deserving of so much coddling that the slightest inconvenience is identical to the direst oppression?
If you achieve Step 2 while pulling $350k/year, you may not stay out of penury all that long after Step 4, so hopefully 'an acceptable time' is relatively short.
It's probably worth noting that, aside from both being set up and funded by Alfred Nobel, there isn't necessarily much connection between a nobel in one field and one in another. Different judges, different selection processes, different external pressures, etc. Not really any reason to think differently of one because you think the laureates for another are badly chosen.
Also, the history of the nobel peace prize is, unfortunately, littered with 'statesmen', which frequently means that winners have more blood on their hands than your average SuperMax inmate. If anything, it would have been more appropriate to award Obama after he'd made clear that a unilateral campaign of drone strikes and a 'war on terror' of unlimited scope and duration were just peachy keen with him, rather than before he'd had a chance to really get into stride.
As for Watson, he seems to have mistaken being treated like a slightly embarrassing curmudgeon whose most productive years as a scientist are well behind him with being actively persecuted. I can't really blame him for acting a bit disappointed at the feeling of having come down in the world; but it's pretty hard to imagine how absurdly high-powered his later career would have had to have been to not be a letdown after his early work, and it wasn't particularly. His old-age PR does seem somewhat worse than some other scientific luminaries who have just quietly faded away; but even the ones that age into avuncular and eventually grandfatherly figures still tend to be remembered for the science they did when they were substantially younger, with their later output varying between 'respectable' and 'more or less retirement'.
Even if they don't provide for it, it'll take something fairly exotic, verging on either imaginary perfect DRM or some sort of verging-on-sci-fi handwaving, to make something that runs on a computer impossible to make copies of.
Even if it isn't a nice, well-behaved, pure-software implementation (say, with loads of FPGA-like hardware elements involved that are modified during execution) it would still be rather striking if it weren't possible, with sufficient trouble, to cut the power and dump all the relevant state one FPGA header at a time and then load it into another suitably similar piece of hardware and start it back up.
AI will do what it is programming to do and follow the rules we lay out for it to follow.
It'll be kind of a shitty 'AI' in that case, no?
This seems to be partially a semantic game(a variant of the one played when devising tests for 'what is AI?'):
People are not, in general, suggesting that expert systems will exterminate humanity (or, if they do, it'll be because some asshole told them to, not because it was their idea, or even an idea at all); but when people say 'AI', they usually have in mind something different, and usually less predictable, than 'an expert system with atypically good tolerance for sloppy input'.
So long as you work on the assumption that 'AI' is, in fact, a discipline concerned with building better expert systems, then yeah, sure, we already know that those are all kinds of useful and, barring deliberate malice or considerable folly, unlikely to go off the rails. Doing, that, though, involves making several substantial claims either about the scope of the possible('AI' is the discipline of expert systems because 'real AI' simply isn't possible) or human stupidity, malice, and shortsightedness(Nobody would ever bother to build an expert system for enhancing expert systems rather than for achieving some direct goal, nor would anybody build an expert system with malice in mind because that'd just be naughty, nor would the expert systems that people will build to kill one another more effectively suffer from IFF issues, potentially on a dramatic scale.)
Personally, given the amazing advances that 'real AI' researchers have largely failed to make, the advances that expert systems have made, and the number of people killed by other people through malice, apathy, and worse, I'm not terribly concerned about the AIpocalypse myself; but it seems like semantic quibbling at best, and begging the question at worst, to say that "AI wouldn't do that; because I've just redefined AI as something that is pretty good at doing what I tell it to but totally lacking in agency."
Aside from the groups that are merely clueless or for sale, the major argument involves a little careful wordplay about what 'internet access' will actually mean as well as some careful dodging of inconvenient questions concerning prices.
Given the present cost of providing service and accepted margin, there are a variety of areas and people who don't end up with internet access, mostly urban poor and rural. The theory is that, without net neutrality, various exciting new business models (mostly variations on shaking down existing internet activity for more money; but never you mind that...) will be possible, which will make it economic to provide service to the currently underserved. Don't, of course, ask why this change would lead to more access in shitty areas, rather than continued non-access in those areas and higher prices elsewhere, or inquire as to whether an internet built on shaking down businesses looking to reach customers might be wildly in favor of incumbents...
It is also quite likely that, in terms of understanding and the people involved, there's a lot of holdover from the assorted minority-interest/minority-owned radio and TV battles. With the exception of the people who are simply too young to have been involved, those are the areas were people concerned with lack of minority access to culturally relevant communications systems are likely to be coming from; but, inconveniently, those areas really lead to a number of nasty misunderstandings: "Internet" isn't especially similar to broadcast media. People who come at the problem from a background in scrabbling over broadcast media seem to fall fairly readily into one of several traps.
Most obviously, the temptation seems to be to fall into notion that 'internet access' is more or less a binary thing, possibly with some understanding of 'broadband' vs. 'not broadband'. In this case, the ISP strategy is to promise some additional coverage of uneconomic customers at cheap rates(often with a raft of fine print limitations and for a limited time, because it's a screwjob rather than a good faith offer) if they are allowed to get their way with customers who have enough money to be worth squeezing. If your background is campaigning for access on behalf of those without access, this looks fairly attractive. Unfortunately, you end up being the least-valued customers of an even worse oligopoly ISP at an overall cost that is likely to be higher than just outright subsidizing the additional households that actually ended up getting service.
Aside from the fairly obvious point that Tesla is going to stay relatively niche if they can't deliver something cheaper, how does any of this make sense?
So, Tesla is the niche-elitist-greenies car because it costs too much; but it doesn't want to 'play by the rules' because those rules would involve an expensive physical buildout and/or sharing profits with a lot of middlemen.
At best, these are two unrelated issues (Tesla's product line vs. their distribution model). At worst, this is internally contradictory nonsense: Tesla is too expensive; but doomed to be niche because they won't embrace a sales model that would make their product even more expensive? How does that work? If the product comes off the line too expensive, any additional friction on the way to the customer isn't going to help.
I wouldn't bet on it, at least not in the near to medium term. Car dealers have a surprising amount of local and state clout and skew heavily republican.
They aren't terribly popular, nor is catering to them a matter of principle(aside from the principle of scratching the backs of those who scratch yours); but crossing them is something that will be rather uncomfortable unless the pressure becomes inevitable enough that they can be deserted by more or less all their allies, all at once.
The idea that things would be fundamentally different because the internet was always rather silly(if any technology can claim to have fundamentally changed language it would probably be writing; but aside from that pickings are somewhat slim); but some less extreme variants are more plausible: the internet certainly has changed who it is cheap and easy to speak to relatively frequently, though not as much as either its biggest friends or its biggest detractors may have expected.
You do realize that the coinage of new words(indeed, more than a few new languages) predates whatever goofy persecution complex you've worked yourself into by millenia, right?
I suspect that the details differ(probably enough that what I'm about to say will cause any subject-matter-expert to want to punch me); but the third rail contact is essentially a consumable, like a heroic-scale version of the 'brushes' used in electric motors(except the brushless ones, obviously).
Slashdot Mirror
HEVC, ouch, that'll sting pretty brutally.
The one interesting quirk about being directly based on a video codec, though, is that a variety of devices (eg. SoCs with fixed function decode blocks, computers with OS vendors that pay the appropriate tithes, etc.) might be 'automatically' licensed for the special-case use of treating a single HEVC frame as a still image by virtue of being suitably licensed for HEVC playback. It'd be amusing to see whether that argument is accepted, or whether 'single frame playback' will mysteriously require a different set of licenses.
Not a patent minefield I want to wade into, though, though I do very much appreciate the work of those who ensure that we have access to solid implementations of these standards, even if it is far beyond their power to untangle the patent nightmare.
Apropos of that review, it would be interesting to see how it stacks up against NAS4free at this point. I'm currently running NAS4Free, after a period on FreeNAS, and my impression largely mirrored Arstechnica's.
NAS4Free isn't terribly shiny, and lacks some features that would be kind of neat to have; but (as one might hope for a NAS) its file server stuff is about as solid, predictable, and simple as it reasonably can be (which isn't always simple, should you delve into the more exiting waters of heavy AD integration, ZFS-Fu, or both at the same time; but it's dead simple to set up a basic NAS and is thankfully free of the "magic GUI promises to make it easy, utterly trashes things!" issue.
If FreeNAS is improving on some of its eccentricities with each release it will likely overtake NAS4Free at some point, since it'll still have all the features and the reliability; but NAS4Free has also had a few little bits grafted on over time (mostly additional quasi-NAS protocol support, DLNA, whatever iTunes uses, etc.)
CDNs also have that 'comparative advantage' thing going for them. Damned if I know how they do it; but the relevant specialists can keep a given chunk of data online for preposterously tiny amounts of money, especially, compared to what it'd cost me to do relatively low-volume hosting myself. At sufficiently large scales, farming out something like that is increasingly likely to be a false economy; but you need to be slinging a nontrivial amount of data(or tolerant of severely dubious backup and redundancy) before you can beat the people who do nothing but that for a living.
That, and potentially somebody popping out of the woodwork with a patent claim, however dodgy.
For a free format without any sponsors(either altruistic or, as with WebM, probably doing it as a tactical move to improve their negotiating position if the MPEG LA ever decided to shake down youtube for more than token fees for H264), it doesn't take a terribly plausible patent claim to be plausible enough that it'll be hard to find somebody to go slug it out in court.
As much as certain formats are...getting rather elderly...it certainly is handy to be assured through sheer age that any patents that might apply to a given format have expired.
My main problem with the name is that every time I see "BPG" I just mentally autocorrect it to "BGP" and then end up somewhat confused.
With 'mobile' there are really two considerations: One is the fact that 'mobile' (even if the fault is, in fact, with shitty backhaul) is going to be fairly slow in emerging markets. Two, relevant even in wealthy developed nations without asshole oligopoly telcos, is the fact that mobile devices are brutally power constrained, and RF chatter isn't exactly cheap in energy terms. The faster you get the data you need and shut down, or move to a slower, lower power mode, the radios, the happier your battery will be.
With mains power it matters less (electricity isn't free; but a few extra dollars a month is far less annoying than having your battery keel over dead at a bad time); but barring exciting breakthroughs in battery chemistry or design, basically all the savings are going to have to come from the device side.
Sometimes you can 'fix' a format (maybe I'm just dating myself; but MP3s made impressive strides in apparent quality at a given bitrate in the years after their introduction); but you run into a major constraint:
If a spec-compliant; but not otherwise updated, decoder chokes on your 'improvement', you don't really have an improvement, you have a new format derived from the old format. If there is room to improve the encoder, while still producing something that existing decoders consider valid, great, adoption will likely be easy; but if decoder behavior prevents you from doing what you want to make some improvement or other, sometimes you just need a new format.
JPEG is (barring the possibility of some lossless mode that looks very little like JPEG except for a few metadata fields; but is technically part of the spec, not sure if we have one of those) indeed compressed; but it's lossy compression and lossy compression is an area where there is actually a reasonable amount of ongoing development.
This isn't to say that lossless compression is a trivial problem, or that there have been absolutely no improvements; but the 'by definition, it isn't lossless unless applying the decompression operation to the result of the compression operation produces something identical to the input' criterion makes it much easier to let the mathematicians and computer scientists work out the limits of the possible.
With lossy compression, there aren't any formal limits, which leaves the field much more open to solutions that rely on following the strong and weak points of human perception(visual in this case, auditory in other cases, visual/motion related in others), which leads to much greater complexity and diversity.
The only trick is that deionized water turns into boring old water with some enthusiasm, 'universal solvent' and all that.
If there were a market for pre-leached hardware suitable for various mediums this would presumably be solved; but it hasn't really yet/
Took me five years as a teenager to master the Sargon II chess game for the Commodore 64 on the hardest difficulty level. I'll like to see a four-year-old do that in less time.
I haven't played it myself; but they say that Robot Odyssey will either break your pitiful hominid brain like reject before The Monolith, turn you into a hardcore programmer geek for life, or turn you against any computer game that isn't Medal of Halo Gears of Assault 3.
Protocol IV of the 1980 Convention on Certain Conventional Weapons is unlikely to be a problem.
Per Article 1, weapons specifically designed as their sole combat function, or one of their functions, to cause permanent blindness are Not OK.
However, Per Article 3, "Blinding as an incidental or collateral effect of the legitimate military employment of laser systems, including laser systems used against optical equipment, is not covered by the prohibition of this Protocol." Just aim for a legitimate target and stock up on braille sympathy cards.
Problem solved.
As for making war with fire, light isn't fire, and conventional explosives(never mind thermobarics and incendiaries) are markedly more strongly associated with fire. Lasers have that novelty thing going against them; but anybody who actually cares about the letter of the law probably has hangups about tracers, attacks on fuel dumps, and other routine stuff. As soon as the novelty wears off lasers will recede into the background.
I'm sure that the crew includes a contractor to wipe the lens before firing and then invoice Uncle Sam for precision optical services.
As for mirrors, purely reflective countermeasures will be a waste of time; but ablative ones or water cooled open cell foams might be an issue.
What remains to be seen is whether jets and missiles can shrug off (either through brute-force thickening or more sophisticated ablative armor or actively deployed particulates that effectively scatter incoming light) the relatively tepid amounts of energy that lasers (especially anything that dodges the rather nasty requirements of chemical lasers) are good for, particularly at range, under optically sub-optimal conditions (never have those at sea!).
.50 BMG. Now, if you can't put a bullet on target, that's irrelevant; but in terms of expected stopping power the finest in combat laser technology is...distinctly middling... compared to guns that date back to the period between the world wars. Obviously the fire control system has evolved out of sight; but given how long it'll have to stay on target, you'd hope so.)
Even under the ideal and closely controlled conditions of industrial laser cutters, lasers are abundantly unsafe for ocular exposure; but by no means the speediest remover of bulk material. In an atmosphere, range is going to be constrained by thermal distortion if nothing else, so the ease of keeping photons on target won't be quite as dramatic as it would be in space, and against close-in non-aircraft, there'll be a lot of cheap 'n nasty (but probably embarrassingly effective) countermeasures involving coating things with mud, spraying them with seawater, and generally making a 3rd world nuisance of yourself.
(By way of comparison, assuming that this 30Kw laser delivers 100% of energy to target, it'll take 2/3 of a second of continuous exposure to deliver the same number of joules to the target as a
It doesn't help that, even on the good days when they aren't running malware, ad 'networks' long ago stopped just farming out ads and started gathering impression data, trying to correlate those data across multiple sites that used the same(or related by business in some way) networks, and generally watching you.
The TV-style bullshit that just leaps for your eyes and blares at you is insufferable; but at least it only watches you at the level of granularity provided by the Neilson lab rats.
If the online advertisers aren't watching you every step of the way, it's certainly not for lack of trying. Fuck 'em.
I'm never quite sure if it's mere hyperbole or raging entitlement that causes people to breathlessly use the language of actual oppressive activity to describe people who are being 'punished' in ways that most of us will never have the pleasure of being treated.
"Blacklisting" is, from time to time, an actual labor market practice, and it stings a lot more than being quietly retired to a cushy honorary position. As in, y'know, literally never working in this town (or larger area) again, prolonged unemployment, penury, that sort of fun stuff. Do people just not know or care about what the real thing looks like, or do they think themselves deserving of so much coddling that the slightest inconvenience is identical to the direst oppression?
If you achieve Step 2 while pulling $350k/year, you may not stay out of penury all that long after Step 4, so hopefully 'an acceptable time' is relatively short.
It's probably worth noting that, aside from both being set up and funded by Alfred Nobel, there isn't necessarily much connection between a nobel in one field and one in another. Different judges, different selection processes, different external pressures, etc. Not really any reason to think differently of one because you think the laureates for another are badly chosen.
Also, the history of the nobel peace prize is, unfortunately, littered with 'statesmen', which frequently means that winners have more blood on their hands than your average SuperMax inmate. If anything, it would have been more appropriate to award Obama after he'd made clear that a unilateral campaign of drone strikes and a 'war on terror' of unlimited scope and duration were just peachy keen with him, rather than before he'd had a chance to really get into stride.
As for Watson, he seems to have mistaken being treated like a slightly embarrassing curmudgeon whose most productive years as a scientist are well behind him with being actively persecuted. I can't really blame him for acting a bit disappointed at the feeling of having come down in the world; but it's pretty hard to imagine how absurdly high-powered his later career would have had to have been to not be a letdown after his early work, and it wasn't particularly. His old-age PR does seem somewhat worse than some other scientific luminaries who have just quietly faded away; but even the ones that age into avuncular and eventually grandfatherly figures still tend to be remembered for the science they did when they were substantially younger, with their later output varying between 'respectable' and 'more or less retirement'.
Even if they don't provide for it, it'll take something fairly exotic, verging on either imaginary perfect DRM or some sort of verging-on-sci-fi handwaving, to make something that runs on a computer impossible to make copies of.
Even if it isn't a nice, well-behaved, pure-software implementation (say, with loads of FPGA-like hardware elements involved that are modified during execution) it would still be rather striking if it weren't possible, with sufficient trouble, to cut the power and dump all the relevant state one FPGA header at a time and then load it into another suitably similar piece of hardware and start it back up.
AI will do what it is programming to do and follow the rules we lay out for it to follow.
It'll be kind of a shitty 'AI' in that case, no?
This seems to be partially a semantic game(a variant of the one played when devising tests for 'what is AI?'):
People are not, in general, suggesting that expert systems will exterminate humanity (or, if they do, it'll be because some asshole told them to, not because it was their idea, or even an idea at all); but when people say 'AI', they usually have in mind something different, and usually less predictable, than 'an expert system with atypically good tolerance for sloppy input'.
So long as you work on the assumption that 'AI' is, in fact, a discipline concerned with building better expert systems, then yeah, sure, we already know that those are all kinds of useful and, barring deliberate malice or considerable folly, unlikely to go off the rails. Doing, that, though, involves making several substantial claims either about the scope of the possible('AI' is the discipline of expert systems because 'real AI' simply isn't possible) or human stupidity, malice, and shortsightedness(Nobody would ever bother to build an expert system for enhancing expert systems rather than for achieving some direct goal, nor would anybody build an expert system with malice in mind because that'd just be naughty, nor would the expert systems that people will build to kill one another more effectively suffer from IFF issues, potentially on a dramatic scale.)
Personally, given the amazing advances that 'real AI' researchers have largely failed to make, the advances that expert systems have made, and the number of people killed by other people through malice, apathy, and worse, I'm not terribly concerned about the AIpocalypse myself; but it seems like semantic quibbling at best, and begging the question at worst, to say that "AI wouldn't do that; because I've just redefined AI as something that is pretty good at doing what I tell it to but totally lacking in agency."
Aside from the groups that are merely clueless or for sale, the major argument involves a little careful wordplay about what 'internet access' will actually mean as well as some careful dodging of inconvenient questions concerning prices.
Given the present cost of providing service and accepted margin, there are a variety of areas and people who don't end up with internet access, mostly urban poor and rural. The theory is that, without net neutrality, various exciting new business models (mostly variations on shaking down existing internet activity for more money; but never you mind that...) will be possible, which will make it economic to provide service to the currently underserved. Don't, of course, ask why this change would lead to more access in shitty areas, rather than continued non-access in those areas and higher prices elsewhere, or inquire as to whether an internet built on shaking down businesses looking to reach customers might be wildly in favor of incumbents...
It is also quite likely that, in terms of understanding and the people involved, there's a lot of holdover from the assorted minority-interest/minority-owned radio and TV battles. With the exception of the people who are simply too young to have been involved, those are the areas were people concerned with lack of minority access to culturally relevant communications systems are likely to be coming from; but, inconveniently, those areas really lead to a number of nasty misunderstandings: "Internet" isn't especially similar to broadcast media. People who come at the problem from a background in scrabbling over broadcast media seem to fall fairly readily into one of several traps.
Most obviously, the temptation seems to be to fall into notion that 'internet access' is more or less a binary thing, possibly with some understanding of 'broadband' vs. 'not broadband'. In this case, the ISP strategy is to promise some additional coverage of uneconomic customers at cheap rates(often with a raft of fine print limitations and for a limited time, because it's a screwjob rather than a good faith offer) if they are allowed to get their way with customers who have enough money to be worth squeezing. If your background is campaigning for access on behalf of those without access, this looks fairly attractive. Unfortunately, you end up being the least-valued customers of an even worse oligopoly ISP at an overall cost that is likely to be higher than just outright subsidizing the additional households that actually ended up getting service.
Aside from the fairly obvious point that Tesla is going to stay relatively niche if they can't deliver something cheaper, how does any of this make sense?
So, Tesla is the niche-elitist-greenies car because it costs too much; but it doesn't want to 'play by the rules' because those rules would involve an expensive physical buildout and/or sharing profits with a lot of middlemen.
At best, these are two unrelated issues (Tesla's product line vs. their distribution model). At worst, this is internally contradictory nonsense: Tesla is too expensive; but doomed to be niche because they won't embrace a sales model that would make their product even more expensive? How does that work? If the product comes off the line too expensive, any additional friction on the way to the customer isn't going to help.
I wouldn't bet on it, at least not in the near to medium term. Car dealers have a surprising amount of local and state clout and skew heavily republican.
They aren't terribly popular, nor is catering to them a matter of principle(aside from the principle of scratching the backs of those who scratch yours); but crossing them is something that will be rather uncomfortable unless the pressure becomes inevitable enough that they can be deserted by more or less all their allies, all at once.
The idea that things would be fundamentally different because the internet was always rather silly(if any technology can claim to have fundamentally changed language it would probably be writing; but aside from that pickings are somewhat slim); but some less extreme variants are more plausible: the internet certainly has changed who it is cheap and easy to speak to relatively frequently, though not as much as either its biggest friends or its biggest detractors may have expected.
You do realize that the coinage of new words(indeed, more than a few new languages) predates whatever goofy persecution complex you've worked yourself into by millenia, right?
I suspect that the details differ(probably enough that what I'm about to say will cause any subject-matter-expert to want to punch me); but the third rail contact is essentially a consumable, like a heroic-scale version of the 'brushes' used in electric motors(except the brushless ones, obviously).