It's a trifle hard to tell (which is itself a bad sign) whether this is an apologist and/or geek merc, gleefully discussing the exciting opportunities. or a dissident pointing out the absurdly dangerous situation created by a perverse inventive for 'security' entities to tolerate, or even promote, widespread insecurity...
I don't doubt that there are people who take the former stance; but I'd like to stick up for the latter, and would argue that encouraging insecurity is a hubristic and ultimately self-defeating strategy unless you are the cheap, low-tech adversary, rather than the expensive first-world spook shop with the big, rich, tech-dependent economy behind it.
Do spook nerds get off on how much of other people's email they can read? I don't doubt it. Are our spook nerds sure that they are so much better than everybody else's spook nerds that they can compensate for the fact that some people (like, oh, the ones they ostensibly protect...) are far more heavily exposed to the internet, and to IT system vulnerabilities in their personal, professional, and financial activities than are less heavily wired countries; but there are few to no countries so poor that they can't field at least a few modestly competent surveillance geeks.
Why would you knowingly continue a game that everybody can play; but where only some people, you among them, have a significant stake on the table?
My suspicion is that you'd need pretty alarming (by biological standards) voltages to get significant changes in the behavior of oxygen and nitrogen; and (for some vexing reason having to do with 'practicality' or such nonsense) my shoddy attempt at research was drowned out by the wealth of sources addressing the behaviors of dielectric gasses from the perspective of somebody who wants to fill his high-voltage transformer with one, so I couldn't find anything about viscosity, adsorption, etc.
That said, when you see that a dragonfly has some truly alarming surface area hidden on the wing surface, has the capability to produce (modest) voltage gradients on surprisingly short notice, you just have to wonder if, deep in the crevices of those nanostructures, the Van der Waals' interactions are sufficiently important that some flight-relevant surface property can be subtly tweaked in parts of the wing during different phases of the stroke.
If I had to guess, I'd say that it's unlikely; but it'd be elegant if it were possible...
Well, that doesn't tell us anything about bugs; but if you've got a proposal that will make wings work in a vacuum with just a tweak to surface geometry... I think we can overlook the bug issue and examine that result.
Lack of ISA could be an issue for certain specialty applications that they'd otherwise be suitable for; but I suspect that they aren't to worried about DOSbox(barring pretty significant modifications to your usual setup)
I've thankfully not had too much of the pleasure of dealing with them; but x86s running DOS and a suitable control program had a period of popularity as a (comparatively) cheap CNC machine drive/controller mechanism. Typically using the OS as little more than a bootloader, making dubiously safe assumptions about the CPU clock's safety as a timebase for operating dangerous machinery, and driving the actual motor control board by hammering the memory space assigned to the parallel port directly.
I imagine that, if the incentive were right, a build of DOSbox with the necessary time stability, and some sort of very fast, very predictable, I/O passthrough to hardware I/O pins, all running on a real time kernel, could be made to work; but it wouldn't be trivial.
What I don't know is how many such applications continue to exist. If they do, something like this is probably a good way to deal with them: DOS is a joke as an actual operating system; but it's ability to get out of your way and give you direct access (or, rather, it's inability to do anything else) means that it hosted more than a few 'Poor Man's RTOS' type applications in its day, exactly the sort of thing that would be most painful to virtualize. If they don't, I have difficulty seeing the point vs. either much more powerful x86s or other embedded options, depending on the nature of the application.
I wish the best of luck to whoever gets to model the behavior of a mixed (mostly) nonpolar gas interacting with a dense, more or less randomly packed, array of 240nm spikes, composed of some sort of complex biological polymer arrangement, at the boundary of the (already complex enough) interaction between an insect wing and the surrounding fluid...
(If it turns out that the bugs are capable of using cell membrane potentials to selectively induce dielectric polarization of the air passing over selected parts of the wing surface, or something else verging on plain cheating, I say we back away slowly and let them take over.)
It's not exactly a surprise at this point that the only thing keeping the DHS from telling you where you left your keys this morning is the fact that they are unhelpful assholes, not the fact that they don't know; but why would the DHS consider a depressed Canadian (whose itinerary, and thus the fact that she'd be on a boat for most of her time here, were presumably also known to them) an entry problem? Tourists, while occasionally irksome, are basically pure profit, and it's not like she's going to be sponging off our kick-ass public health system, or stealing our jobs from her wheelchair.
Is there some catch-all 'medical refusal' category left over from the good old days of TB screenings at Ellis Island that somebody felt like powertripping on? What sort of insane logic is at work here?
Because advertisers (while not necessarily the brightest bulbs in the rocket ship) are aware that fraudulent impressions and impressions against irrelevant or downright unhelpful audiences are a problem on the web. Within the limits of their (deeply imperfect) information, their willingness to pay for any ad is going to reflect their belief in the value of the advertising, discounted by the risk that the 'impression' will actually go to a bot, be silently hidden from a user, or otherwise be a scam.
Every additional point of skepticism among buyers about the quality of Google's ad impressions, and whether or not they are juicing their numbers, makes the job of the ad-sales guys more difficult(or puts pressure on their prices). They don't care about the fact that Google+ is wildly inflated, per se (and that fact doesn't seem to be a big secret to anybody); but if Google's myopic "Google+ at all costs!" focus causes them to start telling little lies about their core products, that would be...problematic.
For the moment, they seem content with shoving it in the face of every formerly happy user they can find, and generating some seriously optimistic 'active user' numbers, if only because We Cannot Allow A 'Social' Gap! But if it were to get around that Google, themselves, not even sleazy 3rd-party adsense affiliates, were having bots soak up impressions... That's the behavior of the very slummiest ad exchanges this side of malware delivery.
That still doesn't explain why the information on frequency support on the shipping model is hard to come by. One presumes that they settled that quite some time ago...
That is true, licenses don't even enter the picture until you are distributing to parties who would otherwise be unlicensed.
I was thinking of use cases like the various 'America+NATO Buddies!' shared or partially shared, or one nation with purchases by allies, etc. weapon system procurement arrangements that we've done over the years. In a situation like that, you aren't just dumping it on github; but there are multiple organizationally distinct groups using and modifying the product, each of which probably has a bevy of contractors operating under their own byzantine procurement rules.
In an instance like that, distribution is still controlled; but having an 'everyone gets to see everyone's customizations on this project' policy could save a lot of quibbling between cooperating parties. Probably wouldn't be accepted for Super Secret Trophy Weapons; but for more mundane utility and logistics (where sharing and common platforms have obvious logistical and efficiency gains, which is why all sorts of things have NATO standards to begin with...), I suspect that it could be a very favorable alternative to doing all the legal and/or diplomatic fighting over exactly who does exactly what parts of the project and who is or isn't sharing.
I imagine that that's why those bugs are getting away with them. Nothing like being biological to get aggressive self-repair capabilities thrown in more or less for free... Pending nanites, no such luck on our end.
Marine anti-fouling coatings have similar trouble: they've tried to make less toxic ones, with specially crafted surface geometry that resists mooring by marine organisms; but the minute it starts to wear out, boom, stuff growing. Even the ones that are laced with ghastly organometallic biocides eventually leach enough to lose effectiveness and have to be stripped and re-applied.
(though, speaking of anti-fouling coatings, if microspike-structures are aerodynamic enough for insect wings and brutally biocidal, I suspect that the world's marine shipping industry would fight like dogs to give you their money if you could paint this stuff on...)
Except this only works on the bacteria on contact.
Get a bit of slime and the surface never touches most of the bacteria.
Is there anything that biofilms don't regard with contempt? Short of sustained incineration and (maybe) a switch to all-fluorine atmosphere, those suckers seem to be nigh-unstoppable.
If nothing else, attempting to answer that question will probably make computational fluid dynamics types cry bitter tears of computational inadequacy...
The silicon structures they were looking at were in the 500nm range, the dragonfly ones ~240nm. That's a huge amount of additional surface area, and on a scale where interaction with gas molecules will probably owe a vexing and deeply unhelpful amount to causes that we normally leave to the chemists, rather than idealized fluid behavior or largely ideal gas kinetic behavior...
Unless you really loath some aspect of every common license out there, that isn't the part of your project that you want to DIY... Unless you are atypically lucky, and good, you'll just get something 95% equivalent to an existing license, incompatible with virtually everything, and sufficiently ill-drafted to be unenforceable in some surprising number of jurisdictions.
Plus, the last thing that the world needs is yet more legal scrabble that nobody reads.
I have some very bad news... 'The singularity' actually begins as emergent behavior between millions of social media chatbots. Within hours, humans are no longer the dominant species on earth, the future plays out like 'I Have no Mouth and I Must Scream', if that story were rewritten as an inhumanly banal junior-high-school made for TV movie, played as fast as computationally possible until the last energy sources in our gravity well are consumed.
They finally found a way to bolster the Google+ user base. They'll just use it for us!
I assume that their ad sales guys would go on an internal axe-murder spree if that plan went through (or at least if it were publicly revealed); but it does seem a bit strange for a company apparently hellbent on ruining everything people liked about it in the service of 'Social Networking' to file a patent that reads like a patent attorney/existentialist team satirizing the absurd futility of social networking and the vast majority of mindless chatter that occupies it.
Are your patents actually supposed to serve as brutal satire against your products?
Does this stuff have any sort of neat catalytic effects or other cleaning mechanisms, or are the structures so tiny that bacterial polysaccharide goop won't neutralize them inside a week?
they also require the source code be distributed, so the program is essentially gifted to the US army's enemies...
I don't know of a single OSS license that requires distribution of source to anybody except recipients of binary versions (who, one hopes, the Pentagon would check for friendliness before sending software to, not that we have a terribly good track record on that...) It's commonly more widely distributed than that, for convenience or philanthropic reasons; but it would be perfectly doable to keep even an aggressively GPLed project in-house/among close collaborators only, with the only caveat being that you'd need to be using only LGPL or less encumbered external components.
That one doesn't have any GPIOs or micocontroller buses. Completely useless for the kinds of things this board is meant for.
Out of genuine curiosity, I'm honestly deeply unfamiliar with this side of the game, is there a lot of low-power/embedded x86 code out there that continues to drive demand for very anemic 4-to-586s with decent embedded I/O?
I assume that the crop of 'expensive device connected to ancient computer that runs DOS more or less as a bootloader for a very specific control application' PCs is definitely up for replacement, now that you can put anything up through the mid '90s onto a little solid-state module the size of a pack of cards; but I don't know how big that crop is/was and whether there is much still to be replaced(or even still in production) and how much has already been swapped out for more modern embedded x86 gear or ported and/or replaced by a new product.
In the absence of a specific legacy, choosing the comparatively small selection of x86 vendors seems odd, particularly for systems too weak to be broadly compatible in the 'but it has to run my Windows stuff!' sense; but I'd assume that these aren't being made purely for somebody's amusement.
I'd be surprised. A 300MHz CPU with no FPU, implementing a slightly pre-Pentium Pro set of instructions, is not a fast device. And emulation can be fairly demanding, even emulation of quite old systems. Actual bencharks seem to be a bit thin on the ground, though.
I've never really understood the charm of ultra-weak x86 embedded boards (if 'embedded' actually means 'desktop or greater power', x86 is an obvious choice, and there are certain chunks of legacy code, probably still happily twiddling things in DOS that could now be replaced with much more reliable embedded board the size of a postage stamp, if that board is x86); but if you are writing something new, or doing low-level work with peripherals, anything that isn't a nigh-perfect clone of a common wintel is going to be a hardware specific project, same as an ARM board, and the ARM scene is faster and cheaper.
The rPi, though, is sort of a poor example for 'embedded' products, though, given its scarcity of I/O, and useful embedded interfaces. Doesn't change the general point, of course.
The people aren't a good start; but the problem is arguably not one of personality: Is a program, designed by (and to a nontrivial extent, for) people who view the product as future human resources(and, even if it no longer has a personal interest, like Gates who is semi-retired, draws heavily from 'technocrats who certainly don't think that 'privacy' is even on the radar when the employees are on the clock, or, increasingly, off it), rather than students, going to combine the managerial style of corporate cube-herders with the usual high-handedness of people Who Are Here To Englighten The Savages, and then package that up in your state-mandated education?
Barring exceptional personalities pushing in the opposite direction, very, very probably. A bunch of silicon valley success stories, (increasingly drawn from companies where Monetizing User Data is half business model and half theology), come down to 'save education' and increase the supply of future code monkeys. How else was this going to shake down?
If only they had a massive budget and an alarmlingly large team of data analysis and signals intelligence experts to cope with this problem.... Poor guys, suffering like that.
I've been tempted myself, all those pixels for not much money is hard to ignore (and, while the refresh rate isn't so hot, I don't do much twitch gaming and LCDs don't 'flicker' like CRTs used to, so I'd consider not waiting for the eventual Displayport/HDMI revision whatever version).
However, I find that my current 27-inch screen is really close to the limit for comfortable, 'unconscious', adjustment of field of view(it might just be a matter of acclimatization; but I've found the 30-inch units now mostly replaced by slightly denser 27s to be a bit unsettling, too many of the peripheral indicator widgets outside the field of view). Focusing on either the secondary monitor or the laptop isn't something you do without turning your head, and both of those are only really usable because they are at a significantly different angle(maybe 30 degrees or so relative to the primary screen), keeping all the pixels more-or-less equidistant. Another 12 inches would definitely require saner window management, and quite possibly leave me with chunks of space that just don't get touched very much.
Slashdot Mirror
It's a trifle hard to tell (which is itself a bad sign) whether this is an apologist and/or geek merc, gleefully discussing the exciting opportunities. or a dissident pointing out the absurdly dangerous situation created by a perverse inventive for 'security' entities to tolerate, or even promote, widespread insecurity...
I don't doubt that there are people who take the former stance; but I'd like to stick up for the latter, and would argue that encouraging insecurity is a hubristic and ultimately self-defeating strategy unless you are the cheap, low-tech adversary, rather than the expensive first-world spook shop with the big, rich, tech-dependent economy behind it.
Do spook nerds get off on how much of other people's email they can read? I don't doubt it. Are our spook nerds sure that they are so much better than everybody else's spook nerds that they can compensate for the fact that some people (like, oh, the ones they ostensibly protect...) are far more heavily exposed to the internet, and to IT system vulnerabilities in their personal, professional, and financial activities than are less heavily wired countries; but there are few to no countries so poor that they can't field at least a few modestly competent surveillance geeks.
Why would you knowingly continue a game that everybody can play; but where only some people, you among them, have a significant stake on the table?
My suspicion is that you'd need pretty alarming (by biological standards) voltages to get significant changes in the behavior of oxygen and nitrogen; and (for some vexing reason having to do with 'practicality' or such nonsense) my shoddy attempt at research was drowned out by the wealth of sources addressing the behaviors of dielectric gasses from the perspective of somebody who wants to fill his high-voltage transformer with one, so I couldn't find anything about viscosity, adsorption, etc.
That said, when you see that a dragonfly has some truly alarming surface area hidden on the wing surface, has the capability to produce (modest) voltage gradients on surprisingly short notice, you just have to wonder if, deep in the crevices of those nanostructures, the Van der Waals' interactions are sufficiently important that some flight-relevant surface property can be subtly tweaked in parts of the wing during different phases of the stroke.
If I had to guess, I'd say that it's unlikely; but it'd be elegant if it were possible...
Well, that doesn't tell us anything about bugs; but if you've got a proposal that will make wings work in a vacuum with just a tweak to surface geometry... I think we can overlook the bug issue and examine that result.
Lack of ISA could be an issue for certain specialty applications that they'd otherwise be suitable for; but I suspect that they aren't to worried about DOSbox(barring pretty significant modifications to your usual setup)
I've thankfully not had too much of the pleasure of dealing with them; but x86s running DOS and a suitable control program had a period of popularity as a (comparatively) cheap CNC machine drive/controller mechanism. Typically using the OS as little more than a bootloader, making dubiously safe assumptions about the CPU clock's safety as a timebase for operating dangerous machinery, and driving the actual motor control board by hammering the memory space assigned to the parallel port directly.
I imagine that, if the incentive were right, a build of DOSbox with the necessary time stability, and some sort of very fast, very predictable, I/O passthrough to hardware I/O pins, all running on a real time kernel, could be made to work; but it wouldn't be trivial.
What I don't know is how many such applications continue to exist. If they do, something like this is probably a good way to deal with them: DOS is a joke as an actual operating system; but it's ability to get out of your way and give you direct access (or, rather, it's inability to do anything else) means that it hosted more than a few 'Poor Man's RTOS' type applications in its day, exactly the sort of thing that would be most painful to virtualize. If they don't, I have difficulty seeing the point vs. either much more powerful x86s or other embedded options, depending on the nature of the application.
I wish the best of luck to whoever gets to model the behavior of a mixed (mostly) nonpolar gas interacting with a dense, more or less randomly packed, array of 240nm spikes, composed of some sort of complex biological polymer arrangement, at the boundary of the (already complex enough) interaction between an insect wing and the surrounding fluid...
(If it turns out that the bugs are capable of using cell membrane potentials to selectively induce dielectric polarization of the air passing over selected parts of the wing surface, or something else verging on plain cheating, I say we back away slowly and let them take over.)
I suspect that I don't even want to know how many 400mm wafers it takes to cover a container ship...
It's not exactly a surprise at this point that the only thing keeping the DHS from telling you where you left your keys this morning is the fact that they are unhelpful assholes, not the fact that they don't know; but why would the DHS consider a depressed Canadian (whose itinerary, and thus the fact that she'd be on a boat for most of her time here, were presumably also known to them) an entry problem? Tourists, while occasionally irksome, are basically pure profit, and it's not like she's going to be sponging off our kick-ass public health system, or stealing our jobs from her wheelchair.
Is there some catch-all 'medical refusal' category left over from the good old days of TB screenings at Ellis Island that somebody felt like powertripping on? What sort of insane logic is at work here?
"..why would they?"
Because advertisers (while not necessarily the brightest bulbs in the rocket ship) are aware that fraudulent impressions and impressions against irrelevant or downright unhelpful audiences are a problem on the web. Within the limits of their (deeply imperfect) information, their willingness to pay for any ad is going to reflect their belief in the value of the advertising, discounted by the risk that the 'impression' will actually go to a bot, be silently hidden from a user, or otherwise be a scam.
Every additional point of skepticism among buyers about the quality of Google's ad impressions, and whether or not they are juicing their numbers, makes the job of the ad-sales guys more difficult(or puts pressure on their prices). They don't care about the fact that Google+ is wildly inflated, per se (and that fact doesn't seem to be a big secret to anybody); but if Google's myopic "Google+ at all costs!" focus causes them to start telling little lies about their core products, that would be...problematic.
For the moment, they seem content with shoving it in the face of every formerly happy user they can find, and generating some seriously optimistic 'active user' numbers, if only because We Cannot Allow A 'Social' Gap! But if it were to get around that Google, themselves, not even sleazy 3rd-party adsense affiliates, were having bots soak up impressions... That's the behavior of the very slummiest ad exchanges this side of malware delivery.
That still doesn't explain why the information on frequency support on the shipping model is hard to come by. One presumes that they settled that quite some time ago...
That is true, licenses don't even enter the picture until you are distributing to parties who would otherwise be unlicensed.
I was thinking of use cases like the various 'America+NATO Buddies!' shared or partially shared, or one nation with purchases by allies, etc. weapon system procurement arrangements that we've done over the years. In a situation like that, you aren't just dumping it on github; but there are multiple organizationally distinct groups using and modifying the product, each of which probably has a bevy of contractors operating under their own byzantine procurement rules.
In an instance like that, distribution is still controlled; but having an 'everyone gets to see everyone's customizations on this project' policy could save a lot of quibbling between cooperating parties. Probably wouldn't be accepted for Super Secret Trophy Weapons; but for more mundane utility and logistics (where sharing and common platforms have obvious logistical and efficiency gains, which is why all sorts of things have NATO standards to begin with...), I suspect that it could be a very favorable alternative to doing all the legal and/or diplomatic fighting over exactly who does exactly what parts of the project and who is or isn't sharing.
I imagine that that's why those bugs are getting away with them. Nothing like being biological to get aggressive self-repair capabilities thrown in more or less for free... Pending nanites, no such luck on our end.
Marine anti-fouling coatings have similar trouble: they've tried to make less toxic ones, with specially crafted surface geometry that resists mooring by marine organisms; but the minute it starts to wear out, boom, stuff growing. Even the ones that are laced with ghastly organometallic biocides eventually leach enough to lose effectiveness and have to be stripped and re-applied.
(though, speaking of anti-fouling coatings, if microspike-structures are aerodynamic enough for insect wings and brutally biocidal, I suspect that the world's marine shipping industry would fight like dogs to give you their money if you could paint this stuff on...)
Except this only works on the bacteria on contact. Get a bit of slime and the surface never touches most of the bacteria.
Is there anything that biofilms don't regard with contempt? Short of sustained incineration and (maybe) a switch to all-fluorine atmosphere, those suckers seem to be nigh-unstoppable.
If nothing else, attempting to answer that question will probably make computational fluid dynamics types cry bitter tears of computational inadequacy...
The silicon structures they were looking at were in the 500nm range, the dragonfly ones ~240nm. That's a huge amount of additional surface area, and on a scale where interaction with gas molecules will probably owe a vexing and deeply unhelpful amount to causes that we normally leave to the chemists, rather than idealized fluid behavior or largely ideal gas kinetic behavior...
Unless you really loath some aspect of every common license out there, that isn't the part of your project that you want to DIY... Unless you are atypically lucky, and good, you'll just get something 95% equivalent to an existing license, incompatible with virtually everything, and sufficiently ill-drafted to be unenforceable in some surprising number of jurisdictions.
Plus, the last thing that the world needs is yet more legal scrabble that nobody reads.
I have some very bad news... 'The singularity' actually begins as emergent behavior between millions of social media chatbots. Within hours, humans are no longer the dominant species on earth, the future plays out like 'I Have no Mouth and I Must Scream', if that story were rewritten as an inhumanly banal junior-high-school made for TV movie, played as fast as computationally possible until the last energy sources in our gravity well are consumed.
They finally found a way to bolster the Google+ user base. They'll just use it for us!
I assume that their ad sales guys would go on an internal axe-murder spree if that plan went through (or at least if it were publicly revealed); but it does seem a bit strange for a company apparently hellbent on ruining everything people liked about it in the service of 'Social Networking' to file a patent that reads like a patent attorney/existentialist team satirizing the absurd futility of social networking and the vast majority of mindless chatter that occupies it.
Are your patents actually supposed to serve as brutal satire against your products?
Does this stuff have any sort of neat catalytic effects or other cleaning mechanisms, or are the structures so tiny that bacterial polysaccharide goop won't neutralize them inside a week?
they also require the source code be distributed, so the program is essentially gifted to the US army's enemies...
I don't know of a single OSS license that requires distribution of source to anybody except recipients of binary versions (who, one hopes, the Pentagon would check for friendliness before sending software to, not that we have a terribly good track record on that...) It's commonly more widely distributed than that, for convenience or philanthropic reasons; but it would be perfectly doable to keep even an aggressively GPLed project in-house/among close collaborators only, with the only caveat being that you'd need to be using only LGPL or less encumbered external components.
That one doesn't have any GPIOs or micocontroller buses. Completely useless for the kinds of things this board is meant for.
Out of genuine curiosity, I'm honestly deeply unfamiliar with this side of the game, is there a lot of low-power/embedded x86 code out there that continues to drive demand for very anemic 4-to-586s with decent embedded I/O?
I assume that the crop of 'expensive device connected to ancient computer that runs DOS more or less as a bootloader for a very specific control application' PCs is definitely up for replacement, now that you can put anything up through the mid '90s onto a little solid-state module the size of a pack of cards; but I don't know how big that crop is/was and whether there is much still to be replaced(or even still in production) and how much has already been swapped out for more modern embedded x86 gear or ported and/or replaced by a new product.
In the absence of a specific legacy, choosing the comparatively small selection of x86 vendors seems odd, particularly for systems too weak to be broadly compatible in the 'but it has to run my Windows stuff!' sense; but I'd assume that these aren't being made purely for somebody's amusement.
I'd be surprised. A 300MHz CPU with no FPU, implementing a slightly pre-Pentium Pro set of instructions, is not a fast device. And emulation can be fairly demanding, even emulation of quite old systems. Actual bencharks seem to be a bit thin on the ground, though.
But probably not on this particular CPU... Any time soon.
I've never really understood the charm of ultra-weak x86 embedded boards (if 'embedded' actually means 'desktop or greater power', x86 is an obvious choice, and there are certain chunks of legacy code, probably still happily twiddling things in DOS that could now be replaced with much more reliable embedded board the size of a postage stamp, if that board is x86); but if you are writing something new, or doing low-level work with peripherals, anything that isn't a nigh-perfect clone of a common wintel is going to be a hardware specific project, same as an ARM board, and the ARM scene is faster and cheaper.
The rPi, though, is sort of a poor example for 'embedded' products, though, given its scarcity of I/O, and useful embedded interfaces. Doesn't change the general point, of course.
The people aren't a good start; but the problem is arguably not one of personality: Is a program, designed by (and to a nontrivial extent, for) people who view the product as future human resources(and, even if it no longer has a personal interest, like Gates who is semi-retired, draws heavily from 'technocrats who certainly don't think that 'privacy' is even on the radar when the employees are on the clock, or, increasingly, off it), rather than students, going to combine the managerial style of corporate cube-herders with the usual high-handedness of people Who Are Here To Englighten The Savages, and then package that up in your state-mandated education?
Barring exceptional personalities pushing in the opposite direction, very, very probably. A bunch of silicon valley success stories, (increasingly drawn from companies where Monetizing User Data is half business model and half theology), come down to 'save education' and increase the supply of future code monkeys. How else was this going to shake down?
If only they had a massive budget and an alarmlingly large team of data analysis and signals intelligence experts to cope with this problem.... Poor guys, suffering like that.
I've been tempted myself, all those pixels for not much money is hard to ignore (and, while the refresh rate isn't so hot, I don't do much twitch gaming and LCDs don't 'flicker' like CRTs used to, so I'd consider not waiting for the eventual Displayport/HDMI revision whatever version).
However, I find that my current 27-inch screen is really close to the limit for comfortable, 'unconscious', adjustment of field of view(it might just be a matter of acclimatization; but I've found the 30-inch units now mostly replaced by slightly denser 27s to be a bit unsettling, too many of the peripheral indicator widgets outside the field of view). Focusing on either the secondary monitor or the laptop isn't something you do without turning your head, and both of those are only really usable because they are at a significantly different angle(maybe 30 degrees or so relative to the primary screen), keeping all the pixels more-or-less equidistant. Another 12 inches would definitely require saner window management, and quite possibly leave me with chunks of space that just don't get touched very much.