Who the fuck is Rei, and how does Musk's latest dumb antic on dumb social media have any bearing on whether or not there's an organized disinformation campaign against SpaceX?
Look, I hate Trump probably as much as anyone. But this lawsuit is dumb.
It's just an emergency alert. The weather service can issue weather alerts, emergency services can issue alerts for wildfires and earthquakes and such. They're an obvious public good - informing the public of imminent dangers to life and limb.
Could it be abused? In theory, yeah. Not quite sure how you'd do it in practice - it's not like there's a special console in the Oval Office that controls it, any message has to pass through lines of people before it goes out, any one of which would be required to refuse it. I'd be more worried about some FEMA staffer accidentally running something in prod instead of test and spamming the country than the Tangerine Toddler using it as an unblockable twitter.
More to the point, if you're worried that the President is likely to abuse a top-level emergency warning system to shovel propaganda at an unwilling public... the solution is not "don't let the president do that", it's "don't let that person be president". Such an untrustworthy person should not have been elected in the first place, and such a breach of public trust is cause for immediate removal from office, whether by impeachment or 25A or any other means necessary.
After all, we trust the president with nukes. If we can't trust someone with an emergency broadcast system, how the hell can we trust them with thermonuclear weapons?
When I press the windows button, windows redraws my entire damned display with useless icons and grinds to a halt while it tries to predict what I'm typing. We used to push the windows button to pop up a quick an unobtrusive start menu, from where I could easily hit "r" for run.
Once you've switched it to a non-nauseating theme, it almost doesn't look like total garbage. The default theme is a crime against humanity.
Did your machine default to tablet mode for some reason? You normally only get the full-screen start menu if your machine thinks it's a tablet. Poke around the settings - there's System->Tablet Mode, which switches a whole bunch of settings at once, and Personalization->Start Menu->Use Start Full Screen, which just flips between the old-school start menu (or a modern reskin of it, at least) and the tablet-style full-screen one. Also take a look at System->Display->Scale, if you got defaulted into tablet mode it probably also set a rather high scaling factor, which would make the theme look pretty bad as well.
I've done a bunch of installs, I've never had W10 pick the wrong default (although I did flip my Surface into desktop mode and my MIDI station into tablet mode, but that's objectively not what the hardware is). But it really sounds like you got shunted into the wrong settings somehow.
The interface consistently gets in my way when I want to do things that were very simple in earlier versions of windows (for example starting a command prompt).
Opening a command line in W10 is simply right click the windows menu button (or press Win+X), and click "PowerShell" or "PowerShell (Admin)" (I think there's a setting to replace this with the old-school command line if you really want). Or tap the windows key, then type "cmd" as though it were a command line itself. Or win+R and type "cmd" into the Run menu (it stores history so if a command line was your last command, it's just Win+R, Enter). That last has been how I've opened command lines since XP, the second has been there since I think Vista, and IIRC the first is a new addition (maybe was in W8?).
I don't know how you expect it to get any simpler. I certainly don't recall any other simple way to do it existing in prior versions, and I've been running Windows since 95.
Second, their obsession with touchscreens is great for people who don't actually do any real work.
I agree, touchscreens are generally bad for traditional office or programming work. But a tablet makes a great media-consumption device, even with only a touchscreen for input. That's probably why tablets sell so well. After all, there are more consumers of media than producers.
There's also some work that actually benefits from a touchscreen. I use one for my audio workstation - between a pile of MIDI controllers and a basic 24" touchscreen, I can do everything I need, and it's a much more efficient use of space than a mouse+keyboard. I definitely appreciate Microsoft making their UI workable on a touch device. (And, I can even load that DAW software onto a tablet, making it a much more portable setup to take to jams).
While I've heard lot of complaints about W10 being too touch-oriented, I can't really see it, myself. I think it's more that it's designed for higher-resolution screens than older OSes. W10 looks fine as a desktop OS, provided you're on a 1080p screen or higher. Try to squeeze it onto an XGA-res screen, and you'll certainly suffer. But I'd rather my interface be designed for the kind of hardware I'm actually running, rather than the lowest common denominator.
Third, touchpads are garbage. The Apple touchpad is almost a valid pointing device but only just. Microsoft doesn't want to sell anything with a useful pointing device; users respond by buying mice to use with their Microsoft laptops and tablets.
Seems kind of contradictory to bitch about how you wish they'd focus on just old-school mouse+keyboard users, and then bitch about how you need a mouse to get the best experience.
I am also not a lawyer, but my understanding is that the SEC would only need to prove malicious intent to pursue a *criminal* action. For a civil action, simple negligence is sufficient - was the information false, and did the information alter the market? Both of those seem pretty clearly true - although Musk's rejection of the settlement makes a lot of sense if he has proof that funding was, in fact, secured.
I use it pretty heavily. I have a lot of computers - home desktop, work desktop, laptop, tablet, phone, that one with all the MIDI stuff. Sync lets me move between them fairly seamlessly.
Sync lets me see what tabs I have open on each machine. I saw this comment this morning on my tablet, didn't feel like finding my keyboard cover to type out a reply, so I left the tab open. I grabbed that tab from my work desktop, which is where I'm typing this now.
I can even send a tab to another computer. I use that a lot on my phone - "Oh, this looks interesting, let me send it to a bigger screen so I can actually read it". Or to send it closer to where I would use it - I can push an article on music theory over to my DAW computer, or a review for a game I might want to buy to my gaming machine, or a programming rant to my work computer.
It also just keeps my browser setup mostly the same between computers. My bookmarks bar has the exact same things on every computer, with no effort needed on my part. All my addons automatically install on new computers when I sign in.
There's a few defects. History syncs incompletely - redirect pages only sync the final destination, so if an RSS feed points to a redirect, only the original computer remembers that I've been there. Something goes wonky with the order of bookmarks sometimes. Settings for addons don't sync, so my custom Adblock list needs manual copying. But it's overall quite a useful feature.
Yeah, I refuse to buy Oculus for just that reason, and I'm hardly a hardliner for boycotting evil products and companies. Facebook is just that bad. But the only viable alternative is Vive, and they just don't seem to be trying anymore. They just came out with a wireless adapter that costs almost as much as the entire Quest system, and they haven't updated the core hardware in forever. The Rift, also, hasn't seen updates but they're at least shitting out new low-end stuff, the Vive ecosystem seems to be dead in comparison.
What would a broken-up Twitter look like? They only have the Twitter network itself, and Periscope, and 99% of the company is Twitter. Splitting them up would still leave Twitter being just as big and problematic. Trying to split the Twitter network won't work - everyone will just switch to one of them. Even if you try to do it on national or regional lines, half the accounts I follow are foreign so I'd end up using them (or more likely, an aggregation service), and then you're right back where you started.
Facebook has some more substantial products besides their core Facebook. There's WhatsApp, Instagram, Oculus... I'd love for Oculus to go independent, the main reason I refuse to buy their hardware is that they're owned by Facebook and are thus guaranteed to turn evil at some point. A breakup here would actually do something. I'm not sure it's a good idea, but it's not completely unproductive like a Twitter breakup.
Google is too big. Search, GMail, Android, Chrome, Chromebooks/ChromeOS, Youtube, Drive, Docs, Pay, Play, Plus, Blogger, AppEngine/Cloud, Waze, Project Fi... the network effect is huge and it's clearly anticompetitive - and I didn't even list Alphabet's separate holdings, which include Waymo and Google Fiber. They need to be broken up. They're already anticompetitive as hell.
I've not tried it myself, but this article was a quite interesting look at the OS and its design.
It's a very minimalist OS, by intent. Like DOS or AmigaOS, everything just runs at ring 0, no memory protection. This would be a security risk, but a) that's the user's fault and b) networking is the work of Satan anyways. Likewise, there's no "users" or "file permissions". There's minimal hardware support - VGA graphics, PC speaker sound, and that's it.
While it will clearly never amount to anything in itself, it would not be bad if Linux took some ideas from it, and a similar project with more focus on modern hardware might be able to find success.
Moore's Law in the phrasing "transistor count per unit area doubles every X months" is probably dying soon. I don't think we've reached the absolute end but we're well out of the exponential section of the curve. We've been lagging since 22nm or so, and transistor shrinks are only going to get slower and slower.
A tweaked phrasing may still be viable, though: "transistor count per unit money doubles every X months". Near the end of the long 32nm/28nm era, the GPU vendors were making some pretty massive chips, on the order of 600mm^2, and still got perfectly usable yields. Consider what might happen if we get stuck at ~14nm, but keep improving cost and yields so that a 1000mm^2 die or larger becomes viable for a standard consumer-grade part.
What will that do for performance? It won't help clock speeds, at least directly. We'll probably not spend it all on more cores. Maybe a huge on-die cache, maybe even your entire memory system in ~8GB of on-die SRAM? Or ASICs - smartphone SoCs already have a lot of die space going towards dedicated coprocessors, since they're more bound by power usage anyways, but what would that look like on the PC side? Would an on-die FPGA be useful? I don't know how things would go, which actually makes it kind of exciting compared to "oh hey, transistor density doubled, clock speeds went up a bit and power draw is down. again."
They might still be useful in SLI/Crossfire configurations. Only one GPU needs video outputs for that to work. Do mining-specific cards still have the SLI edge connector?
GPU production did not substantially increase. Both AMD and Nvidia looked at cryptocoins, realized they were almost certainly a bubble that could pop at any time, and accordingly made no long-term investments to increase production. They pretty much maxed out their contracts with chip foundries (neither make their own chips - Nvidia uses TSMC and Samsung, AMD uses GlobalFoundries and TSMC), but they could never keep up with such an absurd demand spike with existing infrastructure.
Further, the GPUs are hardly useless now. They're already starting to appear on the secondhand market, and be snatched up by gamers. I myself am contemplating getting a second card, if they get cheap enough. The really useless mining hardware are the ASICs - which are pretty tailor-made for the purpose of mining, although I suspect the Bitcoin ASICs might be able to be repurposed to brute-force SHA-256 passwords.
And finally, the headline is a bit inaccurate. Nvidia isn't "giving up". They're just letting shareholders know not to expect another quarter of massive sales volume, because they expect the remaining crypto miners to mostly buy ASICs, not general-purpose cards. (And the glut of used cards is probably going to hurt at least their low-end sales for a year or two - given a choice between this year's $200 card, and last year's $800 card, used, for $200, lots of people will pick the latter.)
... that's not what TFA says (I know, reading the articles? On Slashdot?). Planar flash and DRAM are similar enough to be switched easily, about as easily as CPU fabs switching from one design to another on the same node. 3D flash is another story, but there's still several planar flash fabs running, which (being the least cost-effective flash fabs) would almost certainly switch to DRAM if the price for flash crashes.
While Flash has been steadily dropping in price for the past decade, DRAM has been jumping in price, nearly quadrupling over the past two years. If a Flash price crash will cause fabs to switch over to DRAM and bring those prices down, I say it can't happen soon enough.
I suppose that would be better described as "non-cryogenic". You can get that cold with just a basic heat pump system, no need for liquid nitrogen (as with "high-temperature superconductors") or liquid helium (as with "conventional superconductors").
A subpoena is one step below a warrant, and one does not generally have to convince a judge of anything to obtain one. They are often issued by the clerk of the court, or even by the lawyers themselves. Or, as in this case, it was directly issued by a federal agency.
If one is served a subpoena that they believe is invalid, they must file a formal objection with the court. Simply ignoring a subpoena, even an improper one, is grounds for penalty.
A warrant, in contrast, is an authorization by a judge or other "competent official" for an officer to perform an act that would otherwise be illegal, most frequently to arrest someone or to perform a search, but is also used for executions and myriad other things.
It's not a boolean choice. It's not "either we only let rightthink be the onlythink and throw into the gulags anyone who deviates" or "literally anything must be given the same platform". There's a spectrum of how right an idea is and a spectrum of how much a voice we give to it.
Some ideas are so harmful that we criminalize them - incitement to violence being basically the only one in the US, but Europe generally outlaws holocaust denial and hate speech. Some ideas are not at that level of harmful but deserve a lessened voice - white supremacists ought not to be given an hour-long cable TV program, militant environmentalists shouldn't get nonprofit status, and jihadists don't deserve a Facebook page.
The diversity of ways to spread an idea inherently acts as a limit on how effective any such soft limits can be. If an overzealous social media site decides to ban a correct but contentious idea (say, "global warming is a looming catastrophe"), the idea will be spread through countless other channels and, if the idea has traction, it will cause backlash that decreases that site's viability. The system is ultimately self-correcting over the long term.
The fundamental problem is not that people don't think critically. The problem is that, absent any normalizing forces, certain sorts of harmful ideas spread faster than the counteridea. "[ETHNIC/RELIGIOUS MINORITY GROUP] is secretly infiltrating our country and government and plan to kill us all, we need to kill them first" is one such idea - if you accept it as true, that belief demands you begin spreading it as fast as possible, but if you encounter it and disbelieve it, you have no motivation to spread your disbelief. At most, you will respond only to who you heard it from, if you don't discredit it as a joke or the ramblings of a madman, and only once the idea reaches dangerously wide acceptance will a serious countereffort begin. All I seek is a moderating force - something that slows and impedes the spread of pathogenic ideas.
What do you mean by "worldviews contrary to their own"?
If you mean "opinions they disagree with", I think most Americans would disagree (isn't that ironic?). If you want to blather on about your horoscope or the superiority of Apple products or how Ocarina of Time is the best Zelda game, you should be perfectly free to do so, and I believe that is the majority opinion by a wide margin.
If, however, you mean "calls to take action that is wholly incompatible with free society", then yes, I think most Americans think we need to have less of that, and they are right to think so. People calling for a genocide ought to be excluded from civilized society, and in fact any society that does not attempt to exclude such people cannot rightly be considered civilized. Propaganda whose logical conclusion is an atrocity, even if it does not yet openly call for it, should not be given a megaphone. The large social media platforms - Facebook, Twitter, Instagram, Youtube, Tumblr - are not the final arbiters of what we see and hear. They are simply a powerful amplifier - and there are some ideas that do not deserve to be amplified.
And if we return to the original headline, and define "fake news" in the simple and obvious manner ("news which is demonstrably false"), we reach the almost axiomatic "we should give preferential treatment to truth over lies". Forget all the fiddly details for a moment - can we agree that information that is objectively true, ought to be spread faster and said louder than information which is objectively false?
The mixing of the two concepts is a deliberate ploy on the part of those telling manipulative lies. They're trying to reframe it from "truth vs lies" to "freedom of speech".
G-force and vibration are only an issue during launch, and Falcon is entirely autonomous. The only control active during that period is the abort sequence, which can be triggered by that big lever in the middle.
The crew has no control over the Falcon launch vehicle - the only control they need to be able to operate during launch, while pulling multiple Gs, is the abort sequence, which is that nice big lever in the center. The touchscreens (and all the other buttons) will only be used while in orbit.
There's searches. I use Google because they're actually the best I've found, and I don't bother using any of the "anonymizing" Google frontends because it's futile. All it would tell them is that I'm trying to hide.
There's Android. I do run Firefox on my phone, as gimped as the mobile version is, but I have to assume they get at least broad-level telemetry of what I'm browsing on mobile from their OS-level crap. And worst-case, they snoop my history on that, which I have set to sync with my desktops because that's really convenient, and now they've got 100%. I don't think they're that evil yet, but it's certainly possible. (They also get my location data, call history, and app usage, but that's not relevant to my browsing history)
There's their CDN. Tons of fonts, javascript libraries and web frameworks get hosted by Google for all kinds of unrelated sites. Maybe they aren't tracking that, but I wouldn't bet against it. Hit a random site, it's 50/50 whether something from Google loads. Probably more like 95% if you don't have adblock.
Finally, there's GMail. It isn't directly snooping my browser history but they get every link I follow out of an email.
So after that, I'd guess 70-80% of my browsing is in Google's hands. And I don't really know which 20-30% they don't have, so I have to assume they have all of it.
And I honestly wouldn't mind if they actually managed to do something useful with it. The whole concept of targeted ads is literally a joke at this point, their news suggestions are a crapshoot, and Youtube's recommendations are pathetic (quick study: of the 18 videos in the "recommended" right now, 12 are from channels I'm already subscribed to, 1 is from a channel I used to be subscribed to, and the last 5 are from channels who I have recently watched a video from). I can't help but feel like I could do better myself if I tried.
If they can do a better job of it than Google, I'm all for it.
Google *also* has access to all my browsing, as much as I try not to let them, and they make some hilariously bad recommendations despite all that info. "Oh, you're really big into astronomy and space exploration? Here's a horoscope (for a different astrological sign, not that it matters), a Nabiru conspiracy theory, some Apollo conspiracy theories, and some supermoon crap", "Oh, you watch Youtube documentaries on WW1? Did you know Obama is still coming to take your guns away?", "Oh, you've been searching for info on how Super Nintendo graphics worked and haven't found as much as you hoped? Let's fill that void with rants about SJWs ruining video games. I know you've got a Tumblr and voted for Hillary and probably actually *are* a SJW, but trust me, you want to read this.", "Oh, you're into Magic: The Gathering? Here's the results of the last big Texas Hold'Em tournament. Card games are all basically the same, right?".
Mozilla's also shown a higher willingness to consider privacy. Their existing "you might be interested in" system is strictly client-side - it downloads a small database, and then client-side determines which ones are relevant to you, so your history never has to leave your machine. I don't know if this new one will work the same way, but my baseline expectations are higher than for just about anyone else.
I have a strong personal aversion to phones that are too wide. 5.5" on a 16:9 phone is already borderline, 6"+ is just too big for me. That cuts out a surprisingly large proportion of the Android phone lineup. The Nokia 6.1 has everything I'd want otherwise (well, I wouldn't say no to USB-C, but it's not a must-have), so I feel like waiting on the 5.1 is a good call.
Slashdot Mirror
Who the fuck is Rei, and how does Musk's latest dumb antic on dumb social media have any bearing on whether or not there's an organized disinformation campaign against SpaceX?
Look, I hate Trump probably as much as anyone. But this lawsuit is dumb.
It's just an emergency alert. The weather service can issue weather alerts, emergency services can issue alerts for wildfires and earthquakes and such. They're an obvious public good - informing the public of imminent dangers to life and limb.
Could it be abused? In theory, yeah. Not quite sure how you'd do it in practice - it's not like there's a special console in the Oval Office that controls it, any message has to pass through lines of people before it goes out, any one of which would be required to refuse it. I'd be more worried about some FEMA staffer accidentally running something in prod instead of test and spamming the country than the Tangerine Toddler using it as an unblockable twitter.
More to the point, if you're worried that the President is likely to abuse a top-level emergency warning system to shovel propaganda at an unwilling public... the solution is not "don't let the president do that", it's "don't let that person be president". Such an untrustworthy person should not have been elected in the first place, and such a breach of public trust is cause for immediate removal from office, whether by impeachment or 25A or any other means necessary.
After all, we trust the president with nukes. If we can't trust someone with an emergency broadcast system, how the hell can we trust them with thermonuclear weapons?
When I press the windows button, windows redraws my entire damned display with useless icons and grinds to a halt while it tries to predict what I'm typing. We used to push the windows button to pop up a quick an unobtrusive start menu, from where I could easily hit "r" for run.
Once you've switched it to a non-nauseating theme, it almost doesn't look like total garbage. The default theme is a crime against humanity.
Did your machine default to tablet mode for some reason? You normally only get the full-screen start menu if your machine thinks it's a tablet. Poke around the settings - there's System->Tablet Mode, which switches a whole bunch of settings at once, and Personalization->Start Menu->Use Start Full Screen, which just flips between the old-school start menu (or a modern reskin of it, at least) and the tablet-style full-screen one. Also take a look at System->Display->Scale, if you got defaulted into tablet mode it probably also set a rather high scaling factor, which would make the theme look pretty bad as well.
I've done a bunch of installs, I've never had W10 pick the wrong default (although I did flip my Surface into desktop mode and my MIDI station into tablet mode, but that's objectively not what the hardware is). But it really sounds like you got shunted into the wrong settings somehow.
The interface consistently gets in my way when I want to do things that were very simple in earlier versions of windows (for example starting a command prompt).
Opening a command line in W10 is simply right click the windows menu button (or press Win+X), and click "PowerShell" or "PowerShell (Admin)" (I think there's a setting to replace this with the old-school command line if you really want). Or tap the windows key, then type "cmd" as though it were a command line itself. Or win+R and type "cmd" into the Run menu (it stores history so if a command line was your last command, it's just Win+R, Enter). That last has been how I've opened command lines since XP, the second has been there since I think Vista, and IIRC the first is a new addition (maybe was in W8?).
I don't know how you expect it to get any simpler. I certainly don't recall any other simple way to do it existing in prior versions, and I've been running Windows since 95.
Second, their obsession with touchscreens is great for people who don't actually do any real work.
I agree, touchscreens are generally bad for traditional office or programming work. But a tablet makes a great media-consumption device, even with only a touchscreen for input. That's probably why tablets sell so well. After all, there are more consumers of media than producers.
There's also some work that actually benefits from a touchscreen. I use one for my audio workstation - between a pile of MIDI controllers and a basic 24" touchscreen, I can do everything I need, and it's a much more efficient use of space than a mouse+keyboard. I definitely appreciate Microsoft making their UI workable on a touch device. (And, I can even load that DAW software onto a tablet, making it a much more portable setup to take to jams).
While I've heard lot of complaints about W10 being too touch-oriented, I can't really see it, myself. I think it's more that it's designed for higher-resolution screens than older OSes. W10 looks fine as a desktop OS, provided you're on a 1080p screen or higher. Try to squeeze it onto an XGA-res screen, and you'll certainly suffer. But I'd rather my interface be designed for the kind of hardware I'm actually running, rather than the lowest common denominator.
Third, touchpads are garbage. The Apple touchpad is almost a valid pointing device but only just. Microsoft doesn't want to sell anything with a useful pointing device; users respond by buying mice to use with their Microsoft laptops and tablets.
Seems kind of contradictory to bitch about how you wish they'd focus on just old-school mouse+keyboard users, and then bitch about how you need a mouse to get the best experience.
I am also not a lawyer, but my understanding is that the SEC would only need to prove malicious intent to pursue a *criminal* action. For a civil action, simple negligence is sufficient - was the information false, and did the information alter the market? Both of those seem pretty clearly true - although Musk's rejection of the settlement makes a lot of sense if he has proof that funding was, in fact, secured.
I use it pretty heavily. I have a lot of computers - home desktop, work desktop, laptop, tablet, phone, that one with all the MIDI stuff. Sync lets me move between them fairly seamlessly.
Sync lets me see what tabs I have open on each machine. I saw this comment this morning on my tablet, didn't feel like finding my keyboard cover to type out a reply, so I left the tab open. I grabbed that tab from my work desktop, which is where I'm typing this now.
I can even send a tab to another computer. I use that a lot on my phone - "Oh, this looks interesting, let me send it to a bigger screen so I can actually read it". Or to send it closer to where I would use it - I can push an article on music theory over to my DAW computer, or a review for a game I might want to buy to my gaming machine, or a programming rant to my work computer.
It also just keeps my browser setup mostly the same between computers. My bookmarks bar has the exact same things on every computer, with no effort needed on my part. All my addons automatically install on new computers when I sign in.
There's a few defects. History syncs incompletely - redirect pages only sync the final destination, so if an RSS feed points to a redirect, only the original computer remembers that I've been there. Something goes wonky with the order of bookmarks sometimes. Settings for addons don't sync, so my custom Adblock list needs manual copying. But it's overall quite a useful feature.
Yeah, I refuse to buy Oculus for just that reason, and I'm hardly a hardliner for boycotting evil products and companies. Facebook is just that bad. But the only viable alternative is Vive, and they just don't seem to be trying anymore. They just came out with a wireless adapter that costs almost as much as the entire Quest system, and they haven't updated the core hardware in forever. The Rift, also, hasn't seen updates but they're at least shitting out new low-end stuff, the Vive ecosystem seems to be dead in comparison.
What would a broken-up Twitter look like? They only have the Twitter network itself, and Periscope, and 99% of the company is Twitter. Splitting them up would still leave Twitter being just as big and problematic. Trying to split the Twitter network won't work - everyone will just switch to one of them. Even if you try to do it on national or regional lines, half the accounts I follow are foreign so I'd end up using them (or more likely, an aggregation service), and then you're right back where you started.
Facebook has some more substantial products besides their core Facebook. There's WhatsApp, Instagram, Oculus... I'd love for Oculus to go independent, the main reason I refuse to buy their hardware is that they're owned by Facebook and are thus guaranteed to turn evil at some point. A breakup here would actually do something. I'm not sure it's a good idea, but it's not completely unproductive like a Twitter breakup.
Google is too big. Search, GMail, Android, Chrome, Chromebooks/ChromeOS, Youtube, Drive, Docs, Pay, Play, Plus, Blogger, AppEngine/Cloud, Waze, Project Fi... the network effect is huge and it's clearly anticompetitive - and I didn't even list Alphabet's separate holdings, which include Waymo and Google Fiber. They need to be broken up. They're already anticompetitive as hell.
I've not tried it myself, but this article was a quite interesting look at the OS and its design.
It's a very minimalist OS, by intent. Like DOS or AmigaOS, everything just runs at ring 0, no memory protection. This would be a security risk, but a) that's the user's fault and b) networking is the work of Satan anyways. Likewise, there's no "users" or "file permissions". There's minimal hardware support - VGA graphics, PC speaker sound, and that's it.
While it will clearly never amount to anything in itself, it would not be bad if Linux took some ideas from it, and a similar project with more focus on modern hardware might be able to find success.
Moore's Law in the phrasing "transistor count per unit area doubles every X months" is probably dying soon. I don't think we've reached the absolute end but we're well out of the exponential section of the curve. We've been lagging since 22nm or so, and transistor shrinks are only going to get slower and slower.
A tweaked phrasing may still be viable, though: "transistor count per unit money doubles every X months". Near the end of the long 32nm/28nm era, the GPU vendors were making some pretty massive chips, on the order of 600mm^2, and still got perfectly usable yields. Consider what might happen if we get stuck at ~14nm, but keep improving cost and yields so that a 1000mm^2 die or larger becomes viable for a standard consumer-grade part.
What will that do for performance? It won't help clock speeds, at least directly. We'll probably not spend it all on more cores. Maybe a huge on-die cache, maybe even your entire memory system in ~8GB of on-die SRAM? Or ASICs - smartphone SoCs already have a lot of die space going towards dedicated coprocessors, since they're more bound by power usage anyways, but what would that look like on the PC side? Would an on-die FPGA be useful? I don't know how things would go, which actually makes it kind of exciting compared to "oh hey, transistor density doubled, clock speeds went up a bit and power draw is down. again."
Right, I forgot about those.
They might still be useful in SLI/Crossfire configurations. Only one GPU needs video outputs for that to work. Do mining-specific cards still have the SLI edge connector?
GPU production did not substantially increase. Both AMD and Nvidia looked at cryptocoins, realized they were almost certainly a bubble that could pop at any time, and accordingly made no long-term investments to increase production. They pretty much maxed out their contracts with chip foundries (neither make their own chips - Nvidia uses TSMC and Samsung, AMD uses GlobalFoundries and TSMC), but they could never keep up with such an absurd demand spike with existing infrastructure.
Further, the GPUs are hardly useless now. They're already starting to appear on the secondhand market, and be snatched up by gamers. I myself am contemplating getting a second card, if they get cheap enough. The really useless mining hardware are the ASICs - which are pretty tailor-made for the purpose of mining, although I suspect the Bitcoin ASICs might be able to be repurposed to brute-force SHA-256 passwords.
And finally, the headline is a bit inaccurate. Nvidia isn't "giving up". They're just letting shareholders know not to expect another quarter of massive sales volume, because they expect the remaining crypto miners to mostly buy ASICs, not general-purpose cards. (And the glut of used cards is probably going to hurt at least their low-end sales for a year or two - given a choice between this year's $200 card, and last year's $800 card, used, for $200, lots of people will pick the latter.)
... that's not what TFA says (I know, reading the articles? On Slashdot?). Planar flash and DRAM are similar enough to be switched easily, about as easily as CPU fabs switching from one design to another on the same node. 3D flash is another story, but there's still several planar flash fabs running, which (being the least cost-effective flash fabs) would almost certainly switch to DRAM if the price for flash crashes.
While Flash has been steadily dropping in price for the past decade, DRAM has been jumping in price, nearly quadrupling over the past two years. If a Flash price crash will cause fabs to switch over to DRAM and bring those prices down, I say it can't happen soon enough.
I suppose that would be better described as "non-cryogenic". You can get that cold with just a basic heat pump system, no need for liquid nitrogen (as with "high-temperature superconductors") or liquid helium (as with "conventional superconductors").
A subpoena is one step below a warrant, and one does not generally have to convince a judge of anything to obtain one. They are often issued by the clerk of the court, or even by the lawyers themselves. Or, as in this case, it was directly issued by a federal agency.
If one is served a subpoena that they believe is invalid, they must file a formal objection with the court. Simply ignoring a subpoena, even an improper one, is grounds for penalty.
A warrant, in contrast, is an authorization by a judge or other "competent official" for an officer to perform an act that would otherwise be illegal, most frequently to arrest someone or to perform a search, but is also used for executions and myriad other things.
It's not a boolean choice. It's not "either we only let rightthink be the onlythink and throw into the gulags anyone who deviates" or "literally anything must be given the same platform". There's a spectrum of how right an idea is and a spectrum of how much a voice we give to it.
Some ideas are so harmful that we criminalize them - incitement to violence being basically the only one in the US, but Europe generally outlaws holocaust denial and hate speech. Some ideas are not at that level of harmful but deserve a lessened voice - white supremacists ought not to be given an hour-long cable TV program, militant environmentalists shouldn't get nonprofit status, and jihadists don't deserve a Facebook page.
The diversity of ways to spread an idea inherently acts as a limit on how effective any such soft limits can be. If an overzealous social media site decides to ban a correct but contentious idea (say, "global warming is a looming catastrophe"), the idea will be spread through countless other channels and, if the idea has traction, it will cause backlash that decreases that site's viability. The system is ultimately self-correcting over the long term.
The fundamental problem is not that people don't think critically. The problem is that, absent any normalizing forces, certain sorts of harmful ideas spread faster than the counteridea. "[ETHNIC/RELIGIOUS MINORITY GROUP] is secretly infiltrating our country and government and plan to kill us all, we need to kill them first" is one such idea - if you accept it as true, that belief demands you begin spreading it as fast as possible, but if you encounter it and disbelieve it, you have no motivation to spread your disbelief. At most, you will respond only to who you heard it from, if you don't discredit it as a joke or the ramblings of a madman, and only once the idea reaches dangerously wide acceptance will a serious countereffort begin. All I seek is a moderating force - something that slows and impedes the spread of pathogenic ideas.
What do you mean by "worldviews contrary to their own"?
If you mean "opinions they disagree with", I think most Americans would disagree (isn't that ironic?). If you want to blather on about your horoscope or the superiority of Apple products or how Ocarina of Time is the best Zelda game, you should be perfectly free to do so, and I believe that is the majority opinion by a wide margin.
If, however, you mean "calls to take action that is wholly incompatible with free society", then yes, I think most Americans think we need to have less of that, and they are right to think so. People calling for a genocide ought to be excluded from civilized society, and in fact any society that does not attempt to exclude such people cannot rightly be considered civilized. Propaganda whose logical conclusion is an atrocity, even if it does not yet openly call for it, should not be given a megaphone. The large social media platforms - Facebook, Twitter, Instagram, Youtube, Tumblr - are not the final arbiters of what we see and hear. They are simply a powerful amplifier - and there are some ideas that do not deserve to be amplified.
And if we return to the original headline, and define "fake news" in the simple and obvious manner ("news which is demonstrably false"), we reach the almost axiomatic "we should give preferential treatment to truth over lies". Forget all the fiddly details for a moment - can we agree that information that is objectively true, ought to be spread faster and said louder than information which is objectively false?
The mixing of the two concepts is a deliberate ploy on the part of those telling manipulative lies. They're trying to reframe it from "truth vs lies" to "freedom of speech".
G-force and vibration are only an issue during launch, and Falcon is entirely autonomous. The only control active during that period is the abort sequence, which can be triggered by that big lever in the middle.
SpaceX has designed their flight suits with little pads on the fingers to allow the operation of capacitive touchscreens.
The crew has no control over the Falcon launch vehicle - the only control they need to be able to operate during launch, while pulling multiple Gs, is the abort sequence, which is that nice big lever in the center. The touchscreens (and all the other buttons) will only be used while in orbit.
My privacy died a death by a thousand cuts.
There's searches. I use Google because they're actually the best I've found, and I don't bother using any of the "anonymizing" Google frontends because it's futile. All it would tell them is that I'm trying to hide.
There's Android. I do run Firefox on my phone, as gimped as the mobile version is, but I have to assume they get at least broad-level telemetry of what I'm browsing on mobile from their OS-level crap. And worst-case, they snoop my history on that, which I have set to sync with my desktops because that's really convenient, and now they've got 100%. I don't think they're that evil yet, but it's certainly possible. (They also get my location data, call history, and app usage, but that's not relevant to my browsing history)
There's their CDN. Tons of fonts, javascript libraries and web frameworks get hosted by Google for all kinds of unrelated sites. Maybe they aren't tracking that, but I wouldn't bet against it. Hit a random site, it's 50/50 whether something from Google loads. Probably more like 95% if you don't have adblock.
Finally, there's GMail. It isn't directly snooping my browser history but they get every link I follow out of an email.
So after that, I'd guess 70-80% of my browsing is in Google's hands. And I don't really know which 20-30% they don't have, so I have to assume they have all of it.
And I honestly wouldn't mind if they actually managed to do something useful with it. The whole concept of targeted ads is literally a joke at this point, their news suggestions are a crapshoot, and Youtube's recommendations are pathetic (quick study: of the 18 videos in the "recommended" right now, 12 are from channels I'm already subscribed to, 1 is from a channel I used to be subscribed to, and the last 5 are from channels who I have recently watched a video from). I can't help but feel like I could do better myself if I tried.
If they can do a better job of it than Google, I'm all for it.
Google *also* has access to all my browsing, as much as I try not to let them, and they make some hilariously bad recommendations despite all that info. "Oh, you're really big into astronomy and space exploration? Here's a horoscope (for a different astrological sign, not that it matters), a Nabiru conspiracy theory, some Apollo conspiracy theories, and some supermoon crap", "Oh, you watch Youtube documentaries on WW1? Did you know Obama is still coming to take your guns away?", "Oh, you've been searching for info on how Super Nintendo graphics worked and haven't found as much as you hoped? Let's fill that void with rants about SJWs ruining video games. I know you've got a Tumblr and voted for Hillary and probably actually *are* a SJW, but trust me, you want to read this.", "Oh, you're into Magic: The Gathering? Here's the results of the last big Texas Hold'Em tournament. Card games are all basically the same, right?".
Mozilla's also shown a higher willingness to consider privacy. Their existing "you might be interested in" system is strictly client-side - it downloads a small database, and then client-side determines which ones are relevant to you, so your history never has to leave your machine. I don't know if this new one will work the same way, but my baseline expectations are higher than for just about anyone else.
I have a strong personal aversion to phones that are too wide. 5.5" on a 16:9 phone is already borderline, 6"+ is just too big for me. That cuts out a surprisingly large proportion of the Android phone lineup. The Nokia 6.1 has everything I'd want otherwise (well, I wouldn't say no to USB-C, but it's not a must-have), so I feel like waiting on the 5.1 is a good call.