This device seems totally unsuited to detecting 'fry your gadgets' failure modes. If you hit one of the corner cases where USB power delivery goes for overkill; it'll be over in moments; so having a few numbers displayed during the frying won't help you much.
These sorts of widgets can come in quite handy(nothing you couldn't do with a decent multimeter and some socket bodging; but socket bodging is annoying and tedious): I used to use them a lot when dealing with 'Smartboards' that used a (vendor supplied) overlength USB cable; but depended on bus power, and could be increasingly glitchy if they weren't getting enough of it. Having an easy way to know which computers used the 'meh, connect USB power to the 5v rail, maybe with some kind of fuse' method, and were good for plenty more than 500ma; which ones took a '500ma is by the book; if you don't like it, go cry to the USB SIG' stance; and which ones(mostly laptops) were spotty about being able to provide as much bus power as standards demanded.
Also handy for getting a look at whether your cheapo portable battery pack droops atrociously under load; testing the various devices that use a min-USB connector for power to see how much the really draw, etc. but not a piece of safety equipment.
It is really off-putting to see this sort of mislabeling. The functions this thing is actually capable of(assuming the vendor didn't screw it up) are quite handy to have in your tech-widget drawer; but it's blatantly dishonest to imply that it has much chance of saving your expensive gadget in the event of a nasty power delivery failure.
Ultimately, pretty much any neurology work is doomed to sound like phrenology; because(while pitifully crude and more or less hopelessly wrong); phrenology was an early attempt at generating data from the hypothesis that "'mind' is ultimately something that the brain does"; so unless you decide to head off and look for Cartesian immaterial thinking substances or souls or something; you end up mounting a more technically sophisticated(and hopefully accurate) attempt to chase down the details of how brain ends up getting translated into mind; or at least what correlations exist between certain brain features and certain mind features.
Haven't the Apple forums and the Apple Store reviews [i]always[/i] been the place to discover what issues Apple is actually worried about by seeing whether or not your post about them silently disappears?
If it's just some oddball problem with your configuration; your post will typically stay up. If it's a bug for which a patch doesn't exist; or a recall in the making, it'll be shoved into the memory hole in fairly short order. A trifle oblique; but actually fairly informative.
Which is, of course, the second(and perhaps larger) problem in this case:
A VPN is a wonderful thing in terms of keeping undesirables out of the traffic between the endpoint device and the VPN provider(with some limited exceptions involving faulty implementation, obsolete protocols, or sneaky traffic analysis of unpadded VPN links); but whoever is terminating the VPN for you is a very, very, trusted party.
If your provider is so sleazy that there is malware in the client you are definitely screwed; but even if the client is clean, they unavoidably see all the traffic sent over the VPN link; and you usually only bother with a VPN because either some of your applications don't encrypt their traffic properly; or because you don't want to reveal to the local wifi hotspot operator what hosts you are communicating with. If the VPN operator is shady; all you've done is add some latency and computational overhead in order to allow a different malicious party to watch your network traffic(and potentially modify it). Even better, while wifi hotspots are managed by zillions of different people and companies, making it somewhat harder to aggregate tracking data for a given user across all the APs they use; you voluntarily connect to your VPN provider; so they get all your traffic no matter where you are.
Honestly, given the numerous alarming things you can do when you are a man-in-the-middle; I'm a bit surprised that adding local malware(and thus substantially increasing the risk of detection) was seen as a good strategy. If I were running an evil VPN I'd want my client(which pesky AV companies or security researchers might well download and inspect) to be squeaky clean and standard; basically just an idiot-proof wrapper around the system provided VPN protocols; and instead load up the malice, ad/malware injection, etc. on the server side.
Luckily, we have a fair number of historical examples drawn from the lower-tech world of competition for the water of a given river between multiple parties that live along it.
Atmospheric moisture flows require more sophistication to track and exploit; but conflict patters should be pretty similar. Good thing land wars in Asia never go poorly!
The trouble is that, while that's a good question; it applies all to readily to using custom ROMs or using stock ones.
Should you trust some random dude on the internet who totally got AOSP+CM tweaks working on a newer kernel by aggregating blobs from 4 different devices? No, probably not. He may well be acting in good faith; but you have zero assurance of that; or much reason to trust that he hasn't made some potentially serious error in the process of making it work.
Should you trust your handset vendor/(and telco, if it's a phone that they've had a hand in)/Google? No, very probably not. The vendors do seem to care slightly more about bugs that might cause customer support calls or returns; and a lot less about security patches or providing vaguely recent versions of anything; but aside from those somewhat different technical priorities they aren't markedly more trustworthy than some random person on the XDA forums.
They don't appear to have abandoned the product line; but it's been ages since I've seen a VIA x86 in the wild. HP used to build thin clients around them, after Transmeta died horribly; and prior to Atoms they showed up reasonably frequently on embedded boards(slow; but markedly cheaper than a Pentium M and markedly smaller and cooler than P4); but they don't seem to have done well recently. They were always pretty slow, and ran pretty warm unless clocked quite low, plus their GPU offering is a descendant of the old S3 'Chrome' designs which is...not good...when it comes to software support.
Between Atoms and the AMD G-series SoCs, it was a bit of a slaughter.
It probably helps that the techniques for neutralizing locks and cameras, while typically not legal if used during a burglary, aren't all that interesting to a potential jury; while the techniques for neutralizing dogs are either rather unreliable or deeply unsympathetic. Some dogs will roll right over for a charm offensive and a treat; but you can't rely on that; and if you kill a dog you've probably made yourself less popular than at least half of the actual murders on the docket, which isn't a good plan for a relatively petty property crime.
If this bill's author has had the temerity to claim to be in favor of 'freedom' or 'free markets'; and then pushes this nonsense, somebody needs to feed him to a wood chipper.
Intel has parts that would work(albeit a bit light on GPIO); I've got a dreadful little tablet here based on the Z3735G, and they packed that CPU, a gig of RAM, 16GB of flash, a 1024x600 touchscreen, some sort of BT and wifi, and a battery together for under $50.
If they hadn't also burdened the device with some of the more agonizing firmware I've had the pleasure of dealing with(AMI's dysfunctional take on 32-bit UEFI, no compatibility support module, on a 64-bit platform? Sign me up!); it'd actually be a decent little Linux toy, since that Atom is supported by intel GPU drivers, not the freaky PowerVR stuff.
As best I can tell, though, the Z-series Atoms didn't attract all that much interest(they are comparable to ARM devices aimed as similar price performance niches; but not particularly superior); and vendors weren't exactly clamoring to buy the chips; and Intel doesn't really like selling parts that cheap. They'd much rather try to sell you on a Core M or the like.
There isn't a whole lot of reason to do it; or apparent interest; but it could be done.
I can understand why the Secret Service might be the ones who actually break the bad news and hand over the handset; but does anyone know if they actually have any role in 'approving' it; or do they just provide whatever the Information Assurance Directorate tells them is the correct answer?
Even aside from their recent history of embarrassing scandal, you don't really think of them as one of the techie outfits.
I'm not quite sure why the iRiver IHP-120/140s didn't do FLAC out of the box. They supported some other specialty goodies(line level and optical in and out) that required more hardware and are probably even more esoteric; and they had ogg vorbis support, so it's not like they were MP3 only or wedded to whatever Microsoft was pushing at the time(the 300 series, though, leaned dangerously in that direction).
Luckily rockbox support is quite good on those models, which takes most of the pain away. LCD isn't good enough to do Doom justice, however.
If memory serves, the original logic behind the existence of this thing was dissatisfaction with Twitter jerking around 3rd party client developers in order to ensure that their freeloading peasants were exposed to enough advertising and had suitably limited control over layout, presentation, etc.
This service was going to be the one where developers came first and you were the customer, not the product. As far as I know that part of the vision was delivered; it just turns out that demand for "Like twitter, except basically empty" isn't all that robust, no matter how nice the service is.
I don't disagree that Apple makes good hardware; my point was that (presumably because they care more about iDevices on the low end; and just don't care on the high end; and because, if only because MS and Intel have been cluebatting them as hard as they can for several years, PC OEMs have stepped up their game a little bit) Apple's offerings have gotten comparatively less exciting. They are still very good, unless you are one of the customers they decided they don't care about anymore; but the difference is not as dramatic as it once was.
Back in the bad old days, getting a genuinely thin and light PC laptop was downright hard. Sony and Fujitsu had some slightly eccentric offerings for moderately alarming amounts of money, some of the X-series Thinkpads were pretty good; but ibooks and powerbooks were often actually cheaper once you ignored the janky plastic crap and barely portable stuff in the bargain bin. That situation eased a bit once Intel dropped the pitiful farce that "Pentium 4M" was actually a mobile CPU and accepted that Pentium M parts were going to have to be available across the board, not just as a high end price-gouge product; but even once suitably low power CPUs were available, atrocious screens, shit build quality, and assorted other sins remained the rule.
On the desktop side, the minis were actually pretty aggressive(you could usually 'beat' them with some mini-tower eMachine that managed to be noisier despite having 10-20 times the volume to put a cooling system in; but that wasn't very impressive); The iMacs compared less well in a straight spec-fight; but good all-in-ones were practically nonexistent elsewhere; and the workstation hardware tended to get gimped GPUs; but was otherwise a pretty solid competitor among its peers.
All of this just isn't as true anymore. You can't get a screen that isn't something of an embarrassment for less than ~$1400(there is the macbook air; but 1440x900, in 2017, for $1000?); and once you move north of a thousand bucks; PC laptops suck far less than they used to. The macbook pros are nice; but more 'nice' than 'pro'. iMacs are still pretty good as AIO options; but the less said about the 'Mac Pro' the better.
I have no interest in arguing that what Apple is doing is bad business, they certainly make enough money on it; but it is pretty hard to be surprised that it isn't doing OSX's market share any favors.
PC laptop screens went through some dark, dark times. The cheap crap still has lousy screens; but there was some time where it was hard to find anything decent, at any price(especially after the harrowing of the 4:3 panels and the massacre of what few 19:10s existed). At least now you can get decent panels again, if you stay out of the bargain basement.
While they continue to pull defeat from the jaws of victory with baffling regularity(eg. needlessly atrocious touchpads for no obvious reason); it's amazing how much less-bad your average PC laptop is today, when compared to the race-to-the-bottom and "Yeah, it's a 15in low-res screen and 2 inches thick" era. Models that can go directly head-to-head with Apple's finest are rarer; but you can often save enough money, vs. the really classy Apple gear, that a few minor sins can be overlooked. Combine that with Apple's more or less total neglect of anything desktop/workstation, which is a boring segment but moves a lot of hardware; and the fair success of Chromebooks as practically-disposable cheap 'n portable options; and you have a few reasons why OSX marketshare might not be doing as well outside of the truly devoted.
Back in the day, an ibook/macbook was both good and actually one of the cheaper options if you needed something small and light; mac minis stacked up reasonably favorably against all but the most atrocious cheapy towers; and Mac Pros were pretty respectably priced workstation offerings. I remember, back when they were still doing the intel-based 'cheese grater' case Pros; we were a Dell shop but when we priced out the Pros vs. equivalent Precisions our Dell rep turned a slightly unhealthy color and had to cut us a deal to make it worth going with those rather than just bootcamping the macs. That...isn't exactly...how the world works anymore.
I certainly wouldn't bet against that; I just don't think that they need to kill cmd.exe to do it.
It's just shell, and not even a terribly good one; and the shell is only as powerful as the programs and commands you can use it to invoke. Going pure GUI tends to involve some loss of control/dumbing down, just because you can't realistically cram everything a CLI can do into a GUI that any sane person would want to look at; but if the OS vendor doesn't want you to do something, making it impossible via CLI isn't a particularly different problem than making it impossible via GUI.
No, Nintendo. Their(crude by modern standards; but quite clear in intention) CIC/10NES lockout chips were in full production well before Sony even had a console in the race; and back when 'Microsoft' meant 'MS-DOS 2.0'; and they have been enthusiastically litigating against vendors and distributors of flash carts and assorted unauthorized accessories for ages.
Sony and Microsoft are also control freaks; and quite possibly better at it than Nintendo(they've made mistakes of their own, like the hilarious PS3 LV0 key leak; or the original Xbox's naive assumption that fast busses were enough to keep low end adversaries at bay even though FPGAs exist); but Nintendo has been at least attempting to keep things locked up nice and tight since before Sony and MS had even entered the market.
It looks like Nintendo did their own, slightly quirky, thing in terms of how the ROMs are stored; but the procedure otherwise uses the same tools you use to manipulate Allwinner SoCs over USB. Since this console is just a cut-down Allwinner board, that isn't a surprise; but (as we know from dealing with cellphones and some tablets from the more obnoxious vendors) the ability to lock the bootloader so that it flatly refuses to do anything with an unsigned payload is a pretty standard feature. Some vendors don't turn it on; or allow it to be turned off; but the hardware is generally capable of it.
Given Nintendo's historical opposition to basically anything they don't explicitly allow happening on their consoles, it seems like a real surprise that this one cheerfully accepts being reflashed with a modified system image. Does Nintendo just not care in this case? Are they doing console lockdown almost as retro as the games being emulated?
Microsoft seems to have been getting less reliable in terms of not killing things for novelty's sake(Ballmer may have been a jerkass; but he understood what the job of a OS company is better than the ipad-envy faction); but it seems hard to imagine killing cmd.exe About a zillion legacy customers depend on it; and, because it's a legacy dependency, they actively don't want it to change.
It's pretty obvious that all of Microsoft's future love and attention are going toward Powershell; but what would they gain by not shipping cmd.exe until hell freezes over?
There's also the minor fact (how do you get called an 'analyst' while screwing up something this simple?) that Samsung is pretty much the vendor who does stylus support in phone-size devices, and the 'Galaxy Note 7', as its name suggests, was one of their stylus equipped models.
Even if you aren't especially wedded to Android, Apple simply doesn't make a comparable device(apparently voluntarily; their 'Apple pencil' thing suggests that they could do a stylus supporting phone if they felt like it). If your best explanation for why the people who spent nontrivial amounts of money to get this device would want a similar replacement is 'committed loyalists'; your analysis is fairly pitiful.
Yeah, it seems to be down to individual preference. I was fine with text based stuff, and most video isn't more than 30fps anyway; but I kept running into a situation where I would move the mouse a bit too fast and the cursor disappearing from where it was in one frame and reappearing where it was for the next frame would show up as a 'blink'/'jump' effect that tweaked my visual response to sudden movement and just drove me nuts.
Saves you some trouble if you aren't bothered, though, things are way better than they used to be; but 4k at 60Hz is still off the table for a lot of real world hardware combinations.
This case highlights a more general problem with most(not quite all, Nexus devices and a few others aren't affected) Android hardware:
Vendors just don't supply system images. If they are in a good mood, you might get some OTA updates; and there will be some key combo that allows you to initiate a 'system restore', which may do the trick if nothing has tampered with or corrupted the 'system' side of things and just wiping the user-writeable data is good enough; but if you want to reflash the entire device? Haha, good luck with that.
Doing this with iDevices generally requires installing the weeping pustule that is iTunes; but if you are willing to do that it's pretty trivial: click, click, new system image. ADB is a trifle clunkier and definitely not intended for general consumer use; but for many models the vendor simply does not provide a system image, period.
It's like the bad old days when Wintel OEMs treated Windows install CDs(rather than 'restore partitions') as practically a controlled substance; except that there is no such thing as a 'generic' Android install, so you either get vendor cooperation; hope for a usable 3rd party build; or get nothing.
I really don't understand why this is the case. If you are willing to distribute the ROM written to handsets; you know full well that anyone who really cares will be able to inspect it without too much trouble; and there are a variety of situations where being able to just reflash everything above the bootloader is really convenient(and, even if you are a control freak, you don't sacrifice any control by providing a signed image with a locked bootloader); but a substantial portion of Android devices just don't have system images available, expect perhaps as unofficial extracted versions pulled from devices. Why not?
Slashdot Mirror
This device seems totally unsuited to detecting 'fry your gadgets' failure modes. If you hit one of the corner cases where USB power delivery goes for overkill; it'll be over in moments; so having a few numbers displayed during the frying won't help you much.
These sorts of widgets can come in quite handy(nothing you couldn't do with a decent multimeter and some socket bodging; but socket bodging is annoying and tedious): I used to use them a lot when dealing with 'Smartboards' that used a (vendor supplied) overlength USB cable; but depended on bus power, and could be increasingly glitchy if they weren't getting enough of it. Having an easy way to know which computers used the 'meh, connect USB power to the 5v rail, maybe with some kind of fuse' method, and were good for plenty more than 500ma; which ones took a '500ma is by the book; if you don't like it, go cry to the USB SIG' stance; and which ones(mostly laptops) were spotty about being able to provide as much bus power as standards demanded.
Also handy for getting a look at whether your cheapo portable battery pack droops atrociously under load; testing the various devices that use a min-USB connector for power to see how much the really draw, etc. but not a piece of safety equipment.
It is really off-putting to see this sort of mislabeling. The functions this thing is actually capable of(assuming the vendor didn't screw it up) are quite handy to have in your tech-widget drawer; but it's blatantly dishonest to imply that it has much chance of saving your expensive gadget in the event of a nasty power delivery failure.
Ultimately, pretty much any neurology work is doomed to sound like phrenology; because(while pitifully crude and more or less hopelessly wrong); phrenology was an early attempt at generating data from the hypothesis that "'mind' is ultimately something that the brain does"; so unless you decide to head off and look for Cartesian immaterial thinking substances or souls or something; you end up mounting a more technically sophisticated(and hopefully accurate) attempt to chase down the details of how brain ends up getting translated into mind; or at least what correlations exist between certain brain features and certain mind features.
Haven't the Apple forums and the Apple Store reviews [i]always[/i] been the place to discover what issues Apple is actually worried about by seeing whether or not your post about them silently disappears?
If it's just some oddball problem with your configuration; your post will typically stay up. If it's a bug for which a patch doesn't exist; or a recall in the making, it'll be shoved into the memory hole in fairly short order. A trifle oblique; but actually fairly informative.
Which is, of course, the second(and perhaps larger) problem in this case:
A VPN is a wonderful thing in terms of keeping undesirables out of the traffic between the endpoint device and the VPN provider(with some limited exceptions involving faulty implementation, obsolete protocols, or sneaky traffic analysis of unpadded VPN links); but whoever is terminating the VPN for you is a very, very, trusted party.
If your provider is so sleazy that there is malware in the client you are definitely screwed; but even if the client is clean, they unavoidably see all the traffic sent over the VPN link; and you usually only bother with a VPN because either some of your applications don't encrypt their traffic properly; or because you don't want to reveal to the local wifi hotspot operator what hosts you are communicating with. If the VPN operator is shady; all you've done is add some latency and computational overhead in order to allow a different malicious party to watch your network traffic(and potentially modify it). Even better, while wifi hotspots are managed by zillions of different people and companies, making it somewhat harder to aggregate tracking data for a given user across all the APs they use; you voluntarily connect to your VPN provider; so they get all your traffic no matter where you are.
Honestly, given the numerous alarming things you can do when you are a man-in-the-middle; I'm a bit surprised that adding local malware(and thus substantially increasing the risk of detection) was seen as a good strategy. If I were running an evil VPN I'd want my client(which pesky AV companies or security researchers might well download and inspect) to be squeaky clean and standard; basically just an idiot-proof wrapper around the system provided VPN protocols; and instead load up the malice, ad/malware injection, etc. on the server side.
Luckily, we have a fair number of historical examples drawn from the lower-tech world of competition for the water of a given river between multiple parties that live along it.
Atmospheric moisture flows require more sophistication to track and exploit; but conflict patters should be pretty similar. Good thing land wars in Asia never go poorly!
The trouble is that, while that's a good question; it applies all to readily to using custom ROMs or using stock ones.
Should you trust some random dude on the internet who totally got AOSP+CM tweaks working on a newer kernel by aggregating blobs from 4 different devices? No, probably not. He may well be acting in good faith; but you have zero assurance of that; or much reason to trust that he hasn't made some potentially serious error in the process of making it work.
Should you trust your handset vendor/(and telco, if it's a phone that they've had a hand in)/Google? No, very probably not. The vendors do seem to care slightly more about bugs that might cause customer support calls or returns; and a lot less about security patches or providing vaguely recent versions of anything; but aside from those somewhat different technical priorities they aren't markedly more trustworthy than some random person on the XDA forums.
They don't appear to have abandoned the product line; but it's been ages since I've seen a VIA x86 in the wild. HP used to build thin clients around them, after Transmeta died horribly; and prior to Atoms they showed up reasonably frequently on embedded boards(slow; but markedly cheaper than a Pentium M and markedly smaller and cooler than P4); but they don't seem to have done well recently. They were always pretty slow, and ran pretty warm unless clocked quite low, plus their GPU offering is a descendant of the old S3 'Chrome' designs which is...not good...when it comes to software support.
Between Atoms and the AMD G-series SoCs, it was a bit of a slaughter.
It probably helps that the techniques for neutralizing locks and cameras, while typically not legal if used during a burglary, aren't all that interesting to a potential jury; while the techniques for neutralizing dogs are either rather unreliable or deeply unsympathetic. Some dogs will roll right over for a charm offensive and a treat; but you can't rely on that; and if you kill a dog you've probably made yourself less popular than at least half of the actual murders on the docket, which isn't a good plan for a relatively petty property crime.
If this bill's author has had the temerity to claim to be in favor of 'freedom' or 'free markets'; and then pushes this nonsense, somebody needs to feed him to a wood chipper.
Intel has parts that would work(albeit a bit light on GPIO); I've got a dreadful little tablet here based on the Z3735G, and they packed that CPU, a gig of RAM, 16GB of flash, a 1024x600 touchscreen, some sort of BT and wifi, and a battery together for under $50.
If they hadn't also burdened the device with some of the more agonizing firmware I've had the pleasure of dealing with(AMI's dysfunctional take on 32-bit UEFI, no compatibility support module, on a 64-bit platform? Sign me up!); it'd actually be a decent little Linux toy, since that Atom is supported by intel GPU drivers, not the freaky PowerVR stuff.
As best I can tell, though, the Z-series Atoms didn't attract all that much interest(they are comparable to ARM devices aimed as similar price performance niches; but not particularly superior); and vendors weren't exactly clamoring to buy the chips; and Intel doesn't really like selling parts that cheap. They'd much rather try to sell you on a Core M or the like.
There isn't a whole lot of reason to do it; or apparent interest; but it could be done.
I can understand why the Secret Service might be the ones who actually break the bad news and hand over the handset; but does anyone know if they actually have any role in 'approving' it; or do they just provide whatever the Information Assurance Directorate tells them is the correct answer?
Even aside from their recent history of embarrassing scandal, you don't really think of them as one of the techie outfits.
Rich guy has PR flacks, news at 11?
I'm not quite sure why the iRiver IHP-120/140s didn't do FLAC out of the box. They supported some other specialty goodies(line level and optical in and out) that required more hardware and are probably even more esoteric; and they had ogg vorbis support, so it's not like they were MP3 only or wedded to whatever Microsoft was pushing at the time(the 300 series, though, leaned dangerously in that direction).
Luckily rockbox support is quite good on those models, which takes most of the pain away. LCD isn't good enough to do Doom justice, however.
If memory serves, the original logic behind the existence of this thing was dissatisfaction with Twitter jerking around 3rd party client developers in order to ensure that their freeloading peasants were exposed to enough advertising and had suitably limited control over layout, presentation, etc.
This service was going to be the one where developers came first and you were the customer, not the product. As far as I know that part of the vision was delivered; it just turns out that demand for "Like twitter, except basically empty" isn't all that robust, no matter how nice the service is.
Zardoz is going to be pissed when he finds out that someone is building guns that shoot seeds and makes new life to poison the earth...
I don't disagree that Apple makes good hardware; my point was that (presumably because they care more about iDevices on the low end; and just don't care on the high end; and because, if only because MS and Intel have been cluebatting them as hard as they can for several years, PC OEMs have stepped up their game a little bit) Apple's offerings have gotten comparatively less exciting. They are still very good, unless you are one of the customers they decided they don't care about anymore; but the difference is not as dramatic as it once was.
Back in the bad old days, getting a genuinely thin and light PC laptop was downright hard. Sony and Fujitsu had some slightly eccentric offerings for moderately alarming amounts of money, some of the X-series Thinkpads were pretty good; but ibooks and powerbooks were often actually cheaper once you ignored the janky plastic crap and barely portable stuff in the bargain bin. That situation eased a bit once Intel dropped the pitiful farce that "Pentium 4M" was actually a mobile CPU and accepted that Pentium M parts were going to have to be available across the board, not just as a high end price-gouge product; but even once suitably low power CPUs were available, atrocious screens, shit build quality, and assorted other sins remained the rule.
On the desktop side, the minis were actually pretty aggressive(you could usually 'beat' them with some mini-tower eMachine that managed to be noisier despite having 10-20 times the volume to put a cooling system in; but that wasn't very impressive); The iMacs compared less well in a straight spec-fight; but good all-in-ones were practically nonexistent elsewhere; and the workstation hardware tended to get gimped GPUs; but was otherwise a pretty solid competitor among its peers.
All of this just isn't as true anymore. You can't get a screen that isn't something of an embarrassment for less than ~$1400(there is the macbook air; but 1440x900, in 2017, for $1000?); and once you move north of a thousand bucks; PC laptops suck far less than they used to. The macbook pros are nice; but more 'nice' than 'pro'. iMacs are still pretty good as AIO options; but the less said about the 'Mac Pro' the better.
I have no interest in arguing that what Apple is doing is bad business, they certainly make enough money on it; but it is pretty hard to be surprised that it isn't doing OSX's market share any favors.
PC laptop screens went through some dark, dark times. The cheap crap still has lousy screens; but there was some time where it was hard to find anything decent, at any price(especially after the harrowing of the 4:3 panels and the massacre of what few 19:10s existed). At least now you can get decent panels again, if you stay out of the bargain basement.
While they continue to pull defeat from the jaws of victory with baffling regularity(eg. needlessly atrocious touchpads for no obvious reason); it's amazing how much less-bad your average PC laptop is today, when compared to the race-to-the-bottom and "Yeah, it's a 15in low-res screen and 2 inches thick" era. Models that can go directly head-to-head with Apple's finest are rarer; but you can often save enough money, vs. the really classy Apple gear, that a few minor sins can be overlooked. Combine that with Apple's more or less total neglect of anything desktop/workstation, which is a boring segment but moves a lot of hardware; and the fair success of Chromebooks as practically-disposable cheap 'n portable options; and you have a few reasons why OSX marketshare might not be doing as well outside of the truly devoted.
Back in the day, an ibook/macbook was both good and actually one of the cheaper options if you needed something small and light; mac minis stacked up reasonably favorably against all but the most atrocious cheapy towers; and Mac Pros were pretty respectably priced workstation offerings. I remember, back when they were still doing the intel-based 'cheese grater' case Pros; we were a Dell shop but when we priced out the Pros vs. equivalent Precisions our Dell rep turned a slightly unhealthy color and had to cut us a deal to make it worth going with those rather than just bootcamping the macs. That...isn't exactly...how the world works anymore.
I certainly wouldn't bet against that; I just don't think that they need to kill cmd.exe to do it.
It's just shell, and not even a terribly good one; and the shell is only as powerful as the programs and commands you can use it to invoke. Going pure GUI tends to involve some loss of control/dumbing down, just because you can't realistically cram everything a CLI can do into a GUI that any sane person would want to look at; but if the OS vendor doesn't want you to do something, making it impossible via CLI isn't a particularly different problem than making it impossible via GUI.
No, Nintendo. Their(crude by modern standards; but quite clear in intention) CIC/10NES lockout chips were in full production well before Sony even had a console in the race; and back when 'Microsoft' meant 'MS-DOS 2.0'; and they have been enthusiastically litigating against vendors and distributors of flash carts and assorted unauthorized accessories for ages.
Sony and Microsoft are also control freaks; and quite possibly better at it than Nintendo(they've made mistakes of their own, like the hilarious PS3 LV0 key leak; or the original Xbox's naive assumption that fast busses were enough to keep low end adversaries at bay even though FPGAs exist); but Nintendo has been at least attempting to keep things locked up nice and tight since before Sony and MS had even entered the market.
It looks like Nintendo did their own, slightly quirky, thing in terms of how the ROMs are stored; but the procedure otherwise uses the same tools you use to manipulate Allwinner SoCs over USB. Since this console is just a cut-down Allwinner board, that isn't a surprise; but (as we know from dealing with cellphones and some tablets from the more obnoxious vendors) the ability to lock the bootloader so that it flatly refuses to do anything with an unsigned payload is a pretty standard feature. Some vendors don't turn it on; or allow it to be turned off; but the hardware is generally capable of it.
Given Nintendo's historical opposition to basically anything they don't explicitly allow happening on their consoles, it seems like a real surprise that this one cheerfully accepts being reflashed with a modified system image. Does Nintendo just not care in this case? Are they doing console lockdown almost as retro as the games being emulated?
Microsoft seems to have been getting less reliable in terms of not killing things for novelty's sake(Ballmer may have been a jerkass; but he understood what the job of a OS company is better than the ipad-envy faction); but it seems hard to imagine killing cmd.exe About a zillion legacy customers depend on it; and, because it's a legacy dependency, they actively don't want it to change.
It's pretty obvious that all of Microsoft's future love and attention are going toward Powershell; but what would they gain by not shipping cmd.exe until hell freezes over?
There's also the minor fact (how do you get called an 'analyst' while screwing up something this simple?) that Samsung is pretty much the vendor who does stylus support in phone-size devices, and the 'Galaxy Note 7', as its name suggests, was one of their stylus equipped models.
Even if you aren't especially wedded to Android, Apple simply doesn't make a comparable device(apparently voluntarily; their 'Apple pencil' thing suggests that they could do a stylus supporting phone if they felt like it). If your best explanation for why the people who spent nontrivial amounts of money to get this device would want a similar replacement is 'committed loyalists'; your analysis is fairly pitiful.
Yeah, it seems to be down to individual preference. I was fine with text based stuff, and most video isn't more than 30fps anyway; but I kept running into a situation where I would move the mouse a bit too fast and the cursor disappearing from where it was in one frame and reappearing where it was for the next frame would show up as a 'blink'/'jump' effect that tweaked my visual response to sudden movement and just drove me nuts.
Saves you some trouble if you aren't bothered, though, things are way better than they used to be; but 4k at 60Hz is still off the table for a lot of real world hardware combinations.
This case highlights a more general problem with most(not quite all, Nexus devices and a few others aren't affected) Android hardware:
Vendors just don't supply system images. If they are in a good mood, you might get some OTA updates; and there will be some key combo that allows you to initiate a 'system restore', which may do the trick if nothing has tampered with or corrupted the 'system' side of things and just wiping the user-writeable data is good enough; but if you want to reflash the entire device? Haha, good luck with that.
Doing this with iDevices generally requires installing the weeping pustule that is iTunes; but if you are willing to do that it's pretty trivial: click, click, new system image. ADB is a trifle clunkier and definitely not intended for general consumer use; but for many models the vendor simply does not provide a system image, period.
It's like the bad old days when Wintel OEMs treated Windows install CDs(rather than 'restore partitions') as practically a controlled substance; except that there is no such thing as a 'generic' Android install, so you either get vendor cooperation; hope for a usable 3rd party build; or get nothing.
I really don't understand why this is the case. If you are willing to distribute the ROM written to handsets; you know full well that anyone who really cares will be able to inspect it without too much trouble; and there are a variety of situations where being able to just reflash everything above the bootloader is really convenient(and, even if you are a control freak, you don't sacrifice any control by providing a signed image with a locked bootloader); but a substantial portion of Android devices just don't have system images available, expect perhaps as unofficial extracted versions pulled from devices. Why not?