Nothing is ever enough for these corporations while there are still dollars in circulation that don't belong to them. Facebook is about making amazing solutions to fill Mark Zuckerberg's pocket, nothing more, nothing less. "Real people in the real world" my bollox
And yet... how many of us have actually made a product 1+ billion "real people in the real world" use and how many are just being Internet warriors in the comments field? I've seen a lot of engineers get lost in technical or academic challenges, philosophical issues or just perceived wants and needs the customers/users would have that they really don't. Cue the infamous "No wireless. Less space than a nomad. Lame." when Apple released the iPod. Or why "assume a spherical cow" has become a running joke of theoretical models.
There is a lot of hubris that it's what the STEM people do that's important and everything else is fluff. If nobody else will stick their neck out, I will: I don't really understand people, like what the mainstream wants or why. And that's okay, because the people who do generally haven't don't know much about actually building it. And sometimes the economists will tell you that yes, people want it and yes, we could build it but you'd be spending $100 for a solution to a $10 problem. And sometimes you build it and it's the best solution nobody's heard about.
real world <-- investigate, analyze --> problems problems <-- design, build, test --> solutions solutions <-- communicate, distribute --> real world
I've written a lot of good code that went to naught not because there was anything wrong with it, but because it wasn't actually solving real world problems. And you can of course blame the spec or that the user is holding it wrong, but at the end of the day it just isn't providing value. So I try to go beyond what they teach in STEM classes and work on what's the user really trying to achieve and can I deliver on that. Or maybe it's just in-house what the business analyst wants or the way the architect or development team wants to build things but there's hardly any position where you don't need those skills.
I have this person at work in mind, no doubt he's very bright but he's also quite terrible at talking to people in a way they can understand. Even for an IT person it becomes an incoherent rant of technical details, niche concepts and proscribing solutions instead of explaining them. Meanwhile I pushed through a good technical change recently asking "What if the person changes his mind?" because the proposed solution failed to take that into account. That I pointed out how we'd already solved this other places helped, of course. But if nobody pays attention, we're going to reinvent the wheel and poorly, over and over again.
I guess it's meta-news, the state of the Linux desktop in 2016... I found this piece from 2008 named The writers who cried YOTLD thats starts like this:
If you have followed tech news closely at all within the last ten years, you've probably heard the phrase year of the Linux desktop before. This is the year that Linux makes a breakthrough with home users, and suddenly Microsoft's dominant market share comes toppling down. I believe people have been proclaiming various years as the year of the Linux desktop since as early as 1998 (possibly even earlier).
The Steam client for Linux is now over three years old and market share is at 0.91% and falling, expand OS version line for the details. And AMD has been working on open source drivers for eight and a half years now. At this pace I'm mostly just wondering where it'll be when Win7 sunsets in 2020 and whether I really want to struggle with a 1% market share desktop again. When I ditched it for Win7 back in 2010 I expected it'd take a few more years in the oven, now I'm not sure even a decade will do. But at least I can put the menu on the bottom, unless that's removed again in the next "reinvention" of the desktop.
And yet, the alternative would be to use the same thing forever. Was it a mistake for Netscape at that time? Arguably. Does that example apply to open source? Probably not, open source often doesn't (can't) "go out of business" the way a company like Netscape can. Any time there are 2 software products that solve the same type of problem, one of them could have just not been written because there was already something else. Is writing something new any different than re-writing something from scratch? Not if it is open source and you don't really care about user numbers.
The difference is natural and forced adoption. If you create something new, you don't have existing users and people would have to start using it because they find the pros outweigh the cons. If you see a massive voluntary migration it's pretty obvious you're doing something right. When you rewrite something you have existing users and features that used to work and when things stop working we call those regressions. Most are unintentional side effects that developers agree are bugs and should be fixed.
Rewrites are typically when those rules go out the window. Not supported anymore. That's not a bug it's a feature. That's gone and not coming back. Might come back someday, but not a development priority. Often with a solid dose of hubris like "you shouldn't be doing that" or "once you get use to it you'll love it" or "users want simple, minimalist interfaces" to say "we heard you, but you're wrong". Or to belittle you by saying this is a "tempest in a teapot" by a "vocal minority" of luddites that oppose all change.
I guess if you don't care about users and functionality but just more code and new code it can't fail, but then you could just hire monkeys to bang on typewriters. There are good rewrites where the code was just a buggy spaghetti mess and the rewrite was a huge success, you just don't hear much about those. But often it's just a version of NIH syndrome, I don't understand the code so the code must be stupid and if I could just write it myself I'm sure it would be much cleaner and better. Usually those end up as a total disaster, with the culprit abandoning it midway.
But stop replacing customer service positions with computers. People know how to interact with people. When we get a "friendly AI" on the phone, it's usually an exercise in frustration. It won't fair any better in meatspace. If I wanted my bank teller to be a robot, I'd just use the #**$ing ATM.
I'm guessing you're somewhat like me, when you contact customer support it's because the answer isn't on the screen. There's no such option in the menus and the FAQ doesn't list the answer to your question. Roughly 99,9% of the time when I call customer service, I need an actual live person. But if you've ever worked first line help desk, the unfiltered common sense of the general public you'll know we're a rare breed. I've kinda accepted that the companies I deal with need to have an idiot filter to keep support costs down. If they're reliable and deliver as promised that is, the best customer service I get is the one where I don't have to call customer service much at all. That's really why I put up with it, if I had to call them regularly I wouldn't want do to business with them no matter how much they kissed ass.
Basically we're either going to redistribute the wealth of the machines or enter a new Dark Ages. Everybody sorta forgets the human race spent 1200 years with everyone but the 1% and their servants living like shit.
But the standard for "shit" has pretty much always improved, not regressed. Unless we end up in some kind of resource crunch where we run out of the raw materials to make things, it's probably easier to get some of those automated tools to make happy meals and iPhones (aka bread and circus) than deal with the riots. People need to be pretty damn desperate to start a revolution.
Ok, google, park next to the elevator/blue sedan/in spot 14A/etc...
A relatively simple touch screen should do the trick, if it's not the passenger's responsibility to check the mirrors and such. You just say where you want to go, the car works out if it's safe to do so.
They're enough of a threat to TV networks they get harassed into fixing problems that aren't really theirs, if they deliver to a US-registered IP that should be the end of their responsibility. This is just a policy of appeasement while hopefully kicking them to the curb, it's a global market and you sell to the whole world. It's called globalization and you're only like 50 years behind the times.
This shows what a horrible idea it was for Google to remove the standard driver controls from their car design. First, it gives absolutely no backup when the inevitable failure occurs and the car doesn't know WTF to do. For example how exactly are you supposed to direct the car to a specific parking spot inside a garage?
Just like human drivers Google's car is free to take directions from the passengers but still be the one legally responsible. I'm sure they have a plan B, but it's obvious the car would have a ton more value if it didn't require a licensed and capable driver. And principally humans don't have a backup, if the driver is incapacitated well call an ambulance and a tow truck. Yes, that might mean the passengers are shit out of luck. But if they don't have a license or are drunk or whatever they would be anyway, I don't think it should be a legal condition to be road-worthy. That many people would want manual controls so they can carry on anyway is another story.
The first is that writing a graphics driver is REALLY HARD. I think a lot of the people who were complaining and asking didn't really understand the magnitude of what they were talking about. They were people who'd maybe messed around with a network driver or something and said "Huh, drivers aren't that bad." Graphics drivers are ENORMOUS things, exceedingly complex. Lots and lots and lots of code that interacts with a lot of stuff in different ways. I mean the GPU is literally a little computer in many respects. Also GPUs change fast. New generations come out every 2 years or so and are often radically different architectures with tons of new features. So you have continual new work to do. It isn't like a NIC or RAID controller where 95%+ of the features might be copy-paste from the previous gen. I don't think a lot of people understood just how big an undertaking a GPU driver is.
The real issue is that there's no "assembler-level" standard like x86 and ARM. Before Vulkan, there hasn't even been an "intermediate representation-level" standard, which would be something like Java bytecode. Implementing DirectX/OpenGL has been like re-implementing say.NET or Java, huge and evolving high level libraries. Open source has kind of made the divide internally with Gallium3D, drivers write towards that and mesa runs on top of any Gallium3D driver. AMDGPU is another such divide for the latest AMD generation, where the same kernel driver works with both open and closed source UMDs (user mode drivers) which is cutting the stack at a lower level but for AMD only.
You can kinda see the stack fracturing now: Hardware <--> AMDGPU <--> User mode driver (UMD) for Vulkan <--> OpenGL to Vulkan translation layer <--> "Traditional" applications Hardware <--> open source <--> closed source, but going open according to AMD <--> closed UMD, Vulkan native or mesa-based shim <--> "Traditional" applications
We're finally approaching a situation where there are parts of the stack to replace, it doesn't have to be an all-or-nothing choice. That will make a huge difference.
-Does it have deployed fiber or wiring for internet and phone service?
Well in his defense:
The website informed him, "Comcast Business is available at your address."
Depending on your actual experience with ISPs, you might not be aware that this is what we call a boldfaced lie. It might be. It might not be. It might take forever to install. It just means it's within an area they think they might deliver service. I've experienced that here in Norway, twice. The parents of my best buddy as well. Due to some particular extender, they couldn't deliver. No room in the central. And they probably won't do a real check until sometime after you actually order, which is rather premature if you're just considering renting. The only thing you can probably believe is if it's installed and working right now.
The other lesson here is that contracts where the other side has an inifinite amount of time to deliver are bad. What the sales rep says doesn't matter, in 99% of the cases it's not going to stick and be legally binding. Or at least you don't want the legal costs to make it stick, get it in writing. Sadly this is a large part of running a business, dealing with various other parts of your supply chain or support infrastructure not delivering or not with the scope and quality you were expecting. Like when you make SLAs, the penalties for violating the uptime requirements and conditions for termination are just as important as the agreed level, perhaps even more so.
No company sells the big money makers, if you want to sell something buyers want reasons. This is not like eBay where "I don't need it anymore" counts as a plausible reason, well sometimes you do get the fluff about strategic realignment and such but it's a weak excuse. What you don't want to say is that the business runs poor or the technology is bad. That it's taking too long is actually not that bad, you get potential buyers who:
a) Think they can get it cheap from a company with ADD and take it to maturity. IBM for example would be a company that likes to run the long game. b) Think that they can bring it to market faster, because they got ideas for quick wins that Google missed. Like back to the military. c) Think their R&D can speed up the buyer's projects which may be much closer to maturity. It's not like Google is the only game in town with robotics projects.
At any rate, I'm guessing Google scooped them up mostly for the sensor/vision/navigation technology not the actual robotics. I'm sure they were primarily looking for things to speed up their autonomous car, which could become a massive industry and where the first company get a road-certified AI might get a huge lead on the competition. I'm guessing they either got what they wanted or didn't find what they wanted, so out they go again.
The simple fact is that this is perfect for just about everything EXCEPT heavy-duty gaming and massive 3D work, and allows replacing massive ATX cases with a box the size of a paperback book in a tiny corner of the desk.
No, it's massive overkill. I got my parents a small NUC, slapped another 8GB RAM in there for 10GB total and with a small 64GB SSD it's also perfectly fine for everything EXCEPT heavy-duty gaming and massive 3D work. And it was something like $300 total, not $1000. If I'd have to throw a bigger disk in there, add $50. I never understood the (small, powerful, non-portable) market. Sure, if you change one of those:
big, powerful, non-portable = gaming/workstation PC small, low-power, non-portable = HTPC/casual use small, powerful, portable = power laptop
Those all make sense to me. But cramming it all into a tiny box with heat and noise issues and paying a huge premium for saving a cubic feet of space? I just don't see it. I guess there's a market for it, but I'm not in it...
Apple has been arguing that the All Writs Act, which the FBI is using in order to conscript Apple's assistance, is inapplicable in situations where Congress has passed laws that provide more specificity. 200 years of legal precedent agree with that understanding. And, contrary to your assertions, CALEA clearly provides a higher degree of specificity that's directly applicable in this case, since it explicitly states that law enforcement cannot make these sorts of demands of manufacturers.
Well in this case I think Apple's logic is a little weak. The law says CALEA doesn't require it. It doesn't explicitly prohibit some other law from requiring it. Say you run a restaurant and you're up to all health codes. That doesn't protect you from a negligent manslaughter charge under the penal code if one of your customers die from food poisoning. Or it turns out that one of your exotic dishes was an endangered species under the Lacey act. The general rule is that you must obey all the laws, simultaneously. Like the fun with search warrants and national security letters, you can't claim that because you respond to one you don't have to respond to the other, that's not a legal theory that would fly.
Well the former case at least involves intentionally using the music, which you can resolve using the off button. The latter is more like being charged with aiding copyright infringement selling blank CDs because the customer might make pirated copies. I hope he wins, because it's really a bizarre law.
Uh huh. Totally. Look, have you seen the way they're pushing this? They're going to entice/scare people into upgrading now while it's still free, then turn around and make another extension. They're not really planning to sell this version, ever.
Remind me why competition among public utilities is bad again?
Don't confuse giants competing over monopoly rent with healthy competition. Pre-fiber it was a near-monopoly. Post-fiber it'll be a near-monopoly. Short term they're willing to do almost anything including service upgrades and price dumping to keep you, because they know long term they got you over a barrel. Nobody's going to run a second fiber network after the first one is hooked up, they're going to make back what they lost and more and it's coming out of your hide. That's why we arrange natural monopolies as public utilities, nobody's going to lay down new water or sewage pipes if 99% in that area already get service from somebody else. Fiber will be the same, enjoy the honeymoon but it won't last very long.
However, and this is where I can speak from personal experience, multi-GPU cards are not always a great use of money
Are they ever? It seems like two cards in CF/SLI have pretty much the exact same performance and drawbacks, with much less of a premium. Now I'd really like a single card that could drive my 4K monitor, but even the 980 Ti/Fury X aren't quite there. I'm guessing I'll stick with what I have until 14/16nm and 8GB of HBM2 is an option.
Nobody is trying to make a general intelligence because nobody wants it. What is wanted is domain specific algorithms that are very good at what they do.
Well it's bits and pieces of it. Imagine you could start to combine Watson and AlphaGo, you tell it to "I'd like to play a round of Go with you" and Watson does the natural language parsing of the request and the rules, finds some games that would make good training material, spawns up an instance of AlphaGo that does the initial training and self-training to improve its play. Yes, the end result is to be able to solve domain specific tasks, but the goal is not to create domain specific solutions but more of a "solution factory". It's still not a generic intelligence that'll learn across the domain-specific ones but a collection of them might mimic it fairly well for a wide variety of tasks.
If the goal was to settle, we'd be priorizing Venus, which offers (...) temperatures that can be - and have been, on 1960s Soviet tech - withstood by simple thermal inertia
...for minutes. No probe has lasted even an hour on the surface, while Opportunity is running for the 13th year and counting on Mars. It's a lot more robot friendly which might be even more important than if it's human-friendly.
You can't effectively teach 1-10th grade math in the same class. So we age-tier because we think that's a reasonable approximation of skill-tier or just to split it up so you've had all the parts of the curriculum. Or think why you have divisions in leagues, you learn by playing roughly equal levels with a few better, a few worse than yourself but if you're just trivially beating them or being crushed you learn nothing. That doesn't mean I think individual learning is the one true answer, you learn a lot of valuable lessons explaining and being explained to, cooperating, correcting each other and so on. But ideally you'd do that with your peers in skill.
If you have some time dedicated to working individually it'd be a lot easier to create a dynamic, personalized schedule where you are in peer groups with others of comparable skill. Today it's mostly impossible to say follow the class of the grade above you because then you have something else, either you must jump a year in every subject or you're pretty much stuck where you are. Of course there's also other concern like like a stable social group you can develop inter-personal relations and skills with but it's not like you exclusively played with those in your class anyway. In the pauses between classes you'd play with other kids anyway.
As I recall, the older versions of the kernel added some features and offered others as third party patches. The 2.0.x, 2.2.x, 2.4.x, and 2.6.x branches were supported for a very long time, (...) This was a very successful development model for over a decade and I don't understand why that's changed. Arguably, it would require fewer work since there wouldn't be as many branches to maintain.
Actually it wasn't. Distros were massively cherry-picking from the odd-numbered branches creating huge variations from the stock kernel, creating strange bugs and making the big jumps was a huge pain because the more relaxed requirements to put it in a development branch led to poor quality. Linus tightened ship and basically said do development in your own branch, when it's ready merge it to the released kernel and ~2 months after the merge window closes it'll be released instead of years like the old kernel. No more hodge-podge development kernels full of half-assed changes. It got distros to work more on the upstream kernel than their own variations, leading to more manpower and higher quality in the core project. It was a great success.
And "gay" is the first word that needs to be reclaimed; as "carefree happiness" it has no synonym.
I think "merry" is pretty close, though apart from expressions like Robin Hood and his merry men and merry Chistmas it's fallen somewhat out of fashion. Blissful, cheerful, jolly are also very much in the vicinity so I don't think you have a shortage of alternatives they're just not used much. And I think we feel a linguistic need to have a formal/informal divide when it comes to sexuality like penis/cock, vagina/pussy, sex/intercourse, gay/homosexual for some reason, so I doubt it goes away unless a different informal term takes over which seems rather unlikely at this point.
General issues are everybody's problem. Issues with one particular piece of hardware is:
a) Not really the problem of anybody else b) Not something most people can reproduce c) Not anybody's job
If Dell delivers laptops with Linux preinstalled, then it's their problem, they got the hardware to reproduce it and they got paid people working on it. If $random_user installs $random_distro on $random_laptop, well the manufacturer doesn't care. And while there's always a few people working to make Linux run on everything, they're few and they can't go around buying laptops just because and there's new models all the time. Red Hat will work on supporting the servers that RHEL runs on, they won't generally work on random hardware. And the kernel is mostly driven by paid development, other hardware is very much in the "you want support for that? great, submit a patch and we'll review it" mode.
Some might say that it's because OEMs do not "support ACPI spec properly", but in practice most PCs don't... It could be more practical to just find the patterns that Windows uses, and imitate them.
Says no person who has tried imitating an undocumented binary blob ever. Basically manufacturers just bang the code until it stops crashing, unless you can replicate it exactly which is hopeless in practice you're going to run into random issues. And random issues here aren't just glitches, they're usually crash/hang bugs. I'm sure a lot could be done if you brought the right people together with the right hardware and gave them some money to work on that. But I don't really see who'd do that, because it's not just free time and there's no profit in it.
That's the last kernel that I as an individual was vaguely able to keep up with. Now it's such a bloated mess that unless you've been part of the development team for decades, you're really not going to be able to maintain an understanding. It seems nearly every project in computing goes this way. Oh well, sic transit and stuff.
According to this chart 2.6.32 had ~9.8 MLoC so I'd say you weren't doing too shabby. In fact if you can keep up with over half the current kernel at ~16.8 MLoC, Linus might be hiring...
It would be interesting to set up a Go Turing Test. Either have another top Go player or AlphaGo behind a wall calling the moves.
Can the human champ Lee Se-dol determine if he is playing against a computer or a human... ?
Well at least in the end game the pros were pretty clear that this was not the kind of plays you'd make to try to confuse a 9 dan pro into losing a slightly favorable position. It was forcing Lee Se-dol to counter but all it really did was give him more time to consider the remaining contested areas while playing moves he could blitz if he'd wanted to. Also previously they felt AlphaGo took some really convoluted ways to win where a human would just simplify to claim the win. So when you step out of the game and into the meta-game it seems obvious - to them at least - that you're playing a computer.
Slashdot Mirror
Nothing is ever enough for these corporations while there are still dollars in circulation that don't belong to them. Facebook is about making amazing solutions to fill Mark Zuckerberg's pocket, nothing more, nothing less. "Real people in the real world" my bollox
And yet... how many of us have actually made a product 1+ billion "real people in the real world" use and how many are just being Internet warriors in the comments field? I've seen a lot of engineers get lost in technical or academic challenges, philosophical issues or just perceived wants and needs the customers/users would have that they really don't. Cue the infamous "No wireless. Less space than a nomad. Lame." when Apple released the iPod. Or why "assume a spherical cow" has become a running joke of theoretical models.
There is a lot of hubris that it's what the STEM people do that's important and everything else is fluff. If nobody else will stick their neck out, I will: I don't really understand people, like what the mainstream wants or why. And that's okay, because the people who do generally haven't don't know much about actually building it. And sometimes the economists will tell you that yes, people want it and yes, we could build it but you'd be spending $100 for a solution to a $10 problem. And sometimes you build it and it's the best solution nobody's heard about.
real world <-- investigate, analyze --> problems
problems <-- design, build, test --> solutions
solutions <-- communicate, distribute --> real world
I've written a lot of good code that went to naught not because there was anything wrong with it, but because it wasn't actually solving real world problems. And you can of course blame the spec or that the user is holding it wrong, but at the end of the day it just isn't providing value. So I try to go beyond what they teach in STEM classes and work on what's the user really trying to achieve and can I deliver on that. Or maybe it's just in-house what the business analyst wants or the way the architect or development team wants to build things but there's hardly any position where you don't need those skills.
I have this person at work in mind, no doubt he's very bright but he's also quite terrible at talking to people in a way they can understand. Even for an IT person it becomes an incoherent rant of technical details, niche concepts and proscribing solutions instead of explaining them. Meanwhile I pushed through a good technical change recently asking "What if the person changes his mind?" because the proposed solution failed to take that into account. That I pointed out how we'd already solved this other places helped, of course. But if nobody pays attention, we're going to reinvent the wheel and poorly, over and over again.
I guess it's meta-news, the state of the Linux desktop in 2016... I found this piece from 2008 named The writers who cried YOTLD thats starts like this:
If you have followed tech news closely at all within the last ten years, you've probably heard the phrase year of the Linux desktop before. This is the year that Linux makes a breakthrough with home users, and suddenly Microsoft's dominant market share comes toppling down. I believe people have been proclaiming various years as the year of the Linux desktop since as early as 1998 (possibly even earlier).
The Steam client for Linux is now over three years old and market share is at 0.91% and falling, expand OS version line for the details. And AMD has been working on open source drivers for eight and a half years now. At this pace I'm mostly just wondering where it'll be when Win7 sunsets in 2020 and whether I really want to struggle with a 1% market share desktop again. When I ditched it for Win7 back in 2010 I expected it'd take a few more years in the oven, now I'm not sure even a decade will do. But at least I can put the menu on the bottom, unless that's removed again in the next "reinvention" of the desktop.
And yet, the alternative would be to use the same thing forever. Was it a mistake for Netscape at that time? Arguably. Does that example apply to open source? Probably not, open source often doesn't (can't) "go out of business" the way a company like Netscape can. Any time there are 2 software products that solve the same type of problem, one of them could have just not been written because there was already something else. Is writing something new any different than re-writing something from scratch? Not if it is open source and you don't really care about user numbers.
The difference is natural and forced adoption. If you create something new, you don't have existing users and people would have to start using it because they find the pros outweigh the cons. If you see a massive voluntary migration it's pretty obvious you're doing something right. When you rewrite something you have existing users and features that used to work and when things stop working we call those regressions. Most are unintentional side effects that developers agree are bugs and should be fixed.
Rewrites are typically when those rules go out the window. Not supported anymore. That's not a bug it's a feature. That's gone and not coming back. Might come back someday, but not a development priority. Often with a solid dose of hubris like "you shouldn't be doing that" or "once you get use to it you'll love it" or "users want simple, minimalist interfaces" to say "we heard you, but you're wrong". Or to belittle you by saying this is a "tempest in a teapot" by a "vocal minority" of luddites that oppose all change.
I guess if you don't care about users and functionality but just more code and new code it can't fail, but then you could just hire monkeys to bang on typewriters. There are good rewrites where the code was just a buggy spaghetti mess and the rewrite was a huge success, you just don't hear much about those. But often it's just a version of NIH syndrome, I don't understand the code so the code must be stupid and if I could just write it myself I'm sure it would be much cleaner and better. Usually those end up as a total disaster, with the culprit abandoning it midway.
But stop replacing customer service positions with computers. People know how to interact with people. When we get a "friendly AI" on the phone, it's usually an exercise in frustration. It won't fair any better in meatspace. If I wanted my bank teller to be a robot, I'd just use the #**$ing ATM.
I'm guessing you're somewhat like me, when you contact customer support it's because the answer isn't on the screen. There's no such option in the menus and the FAQ doesn't list the answer to your question. Roughly 99,9% of the time when I call customer service, I need an actual live person. But if you've ever worked first line help desk, the unfiltered common sense of the general public you'll know we're a rare breed. I've kinda accepted that the companies I deal with need to have an idiot filter to keep support costs down. If they're reliable and deliver as promised that is, the best customer service I get is the one where I don't have to call customer service much at all. That's really why I put up with it, if I had to call them regularly I wouldn't want do to business with them no matter how much they kissed ass.
Basically we're either going to redistribute the wealth of the machines or enter a new Dark Ages. Everybody sorta forgets the human race spent 1200 years with everyone but the 1% and their servants living like shit.
But the standard for "shit" has pretty much always improved, not regressed. Unless we end up in some kind of resource crunch where we run out of the raw materials to make things, it's probably easier to get some of those automated tools to make happy meals and iPhones (aka bread and circus) than deal with the riots. People need to be pretty damn desperate to start a revolution.
Ok, google, park next to the elevator/blue sedan/in spot 14A/etc...
A relatively simple touch screen should do the trick, if it's not the passenger's responsibility to check the mirrors and such. You just say where you want to go, the car works out if it's safe to do so.
They're enough of a threat to TV networks they get harassed into fixing problems that aren't really theirs, if they deliver to a US-registered IP that should be the end of their responsibility. This is just a policy of appeasement while hopefully kicking them to the curb, it's a global market and you sell to the whole world. It's called globalization and you're only like 50 years behind the times.
This shows what a horrible idea it was for Google to remove the standard driver controls from their car design. First, it gives absolutely no backup when the inevitable failure occurs and the car doesn't know WTF to do. For example how exactly are you supposed to direct the car to a specific parking spot inside a garage?
Just like human drivers Google's car is free to take directions from the passengers but still be the one legally responsible. I'm sure they have a plan B, but it's obvious the car would have a ton more value if it didn't require a licensed and capable driver. And principally humans don't have a backup, if the driver is incapacitated well call an ambulance and a tow truck. Yes, that might mean the passengers are shit out of luck. But if they don't have a license or are drunk or whatever they would be anyway, I don't think it should be a legal condition to be road-worthy. That many people would want manual controls so they can carry on anyway is another story.
The first is that writing a graphics driver is REALLY HARD. I think a lot of the people who were complaining and asking didn't really understand the magnitude of what they were talking about. They were people who'd maybe messed around with a network driver or something and said "Huh, drivers aren't that bad." Graphics drivers are ENORMOUS things, exceedingly complex. Lots and lots and lots of code that interacts with a lot of stuff in different ways. I mean the GPU is literally a little computer in many respects. Also GPUs change fast. New generations come out every 2 years or so and are often radically different architectures with tons of new features. So you have continual new work to do. It isn't like a NIC or RAID controller where 95%+ of the features might be copy-paste from the previous gen. I don't think a lot of people understood just how big an undertaking a GPU driver is.
The real issue is that there's no "assembler-level" standard like x86 and ARM. Before Vulkan, there hasn't even been an "intermediate representation-level" standard, which would be something like Java bytecode. Implementing DirectX/OpenGL has been like re-implementing say .NET or Java, huge and evolving high level libraries. Open source has kind of made the divide internally with Gallium3D, drivers write towards that and mesa runs on top of any Gallium3D driver. AMDGPU is another such divide for the latest AMD generation, where the same kernel driver works with both open and closed source UMDs (user mode drivers) which is cutting the stack at a lower level but for AMD only.
You can kinda see the stack fracturing now:
Hardware <--> AMDGPU <--> User mode driver (UMD) for Vulkan <--> OpenGL to Vulkan translation layer <--> "Traditional" applications
Hardware <--> open source <--> closed source, but going open according to AMD <--> closed UMD, Vulkan native or mesa-based shim <--> "Traditional" applications
We're finally approaching a situation where there are parts of the stack to replace, it doesn't have to be an all-or-nothing choice. That will make a huge difference.
-Does it have deployed fiber or wiring for internet and phone service?
Well in his defense:
The website informed him, "Comcast Business is available at your address."
Depending on your actual experience with ISPs, you might not be aware that this is what we call a boldfaced lie. It might be. It might not be. It might take forever to install. It just means it's within an area they think they might deliver service. I've experienced that here in Norway, twice. The parents of my best buddy as well. Due to some particular extender, they couldn't deliver. No room in the central. And they probably won't do a real check until sometime after you actually order, which is rather premature if you're just considering renting. The only thing you can probably believe is if it's installed and working right now.
The other lesson here is that contracts where the other side has an inifinite amount of time to deliver are bad. What the sales rep says doesn't matter, in 99% of the cases it's not going to stick and be legally binding. Or at least you don't want the legal costs to make it stick, get it in writing. Sadly this is a large part of running a business, dealing with various other parts of your supply chain or support infrastructure not delivering or not with the scope and quality you were expecting. Like when you make SLAs, the penalties for violating the uptime requirements and conditions for termination are just as important as the agreed level, perhaps even more so.
No company sells the big money makers, if you want to sell something buyers want reasons. This is not like eBay where "I don't need it anymore" counts as a plausible reason, well sometimes you do get the fluff about strategic realignment and such but it's a weak excuse. What you don't want to say is that the business runs poor or the technology is bad. That it's taking too long is actually not that bad, you get potential buyers who:
a) Think they can get it cheap from a company with ADD and take it to maturity. IBM for example would be a company that likes to run the long game.
b) Think that they can bring it to market faster, because they got ideas for quick wins that Google missed. Like back to the military.
c) Think their R&D can speed up the buyer's projects which may be much closer to maturity. It's not like Google is the only game in town with robotics projects.
At any rate, I'm guessing Google scooped them up mostly for the sensor/vision/navigation technology not the actual robotics. I'm sure they were primarily looking for things to speed up their autonomous car, which could become a massive industry and where the first company get a road-certified AI might get a huge lead on the competition. I'm guessing they either got what they wanted or didn't find what they wanted, so out they go again.
The simple fact is that this is perfect for just about everything EXCEPT heavy-duty gaming and massive 3D work, and allows replacing massive ATX cases with a box the size of a paperback book in a tiny corner of the desk.
No, it's massive overkill. I got my parents a small NUC, slapped another 8GB RAM in there for 10GB total and with a small 64GB SSD it's also perfectly fine for everything EXCEPT heavy-duty gaming and massive 3D work. And it was something like $300 total, not $1000. If I'd have to throw a bigger disk in there, add $50. I never understood the (small, powerful, non-portable) market. Sure, if you change one of those:
big, powerful, non-portable = gaming/workstation PC
small, low-power, non-portable = HTPC/casual use
small, powerful, portable = power laptop
Those all make sense to me. But cramming it all into a tiny box with heat and noise issues and paying a huge premium for saving a cubic feet of space? I just don't see it. I guess there's a market for it, but I'm not in it...
Apple has been arguing that the All Writs Act, which the FBI is using in order to conscript Apple's assistance, is inapplicable in situations where Congress has passed laws that provide more specificity. 200 years of legal precedent agree with that understanding. And, contrary to your assertions, CALEA clearly provides a higher degree of specificity that's directly applicable in this case, since it explicitly states that law enforcement cannot make these sorts of demands of manufacturers.
Well in this case I think Apple's logic is a little weak. The law says CALEA doesn't require it. It doesn't explicitly prohibit some other law from requiring it. Say you run a restaurant and you're up to all health codes. That doesn't protect you from a negligent manslaughter charge under the penal code if one of your customers die from food poisoning. Or it turns out that one of your exotic dishes was an endangered species under the Lacey act. The general rule is that you must obey all the laws, simultaneously. Like the fun with search warrants and national security letters, you can't claim that because you respond to one you don't have to respond to the other, that's not a legal theory that would fly.
Well the former case at least involves intentionally using the music, which you can resolve using the off button. The latter is more like being charged with aiding copyright infringement selling blank CDs because the customer might make pirated copies. I hope he wins, because it's really a bizarre law.
Uh huh. Totally. Look, have you seen the way they're pushing this? They're going to entice/scare people into upgrading now while it's still free, then turn around and make another extension. They're not really planning to sell this version, ever.
Remind me why competition among public utilities is bad again?
Don't confuse giants competing over monopoly rent with healthy competition. Pre-fiber it was a near-monopoly. Post-fiber it'll be a near-monopoly. Short term they're willing to do almost anything including service upgrades and price dumping to keep you, because they know long term they got you over a barrel. Nobody's going to run a second fiber network after the first one is hooked up, they're going to make back what they lost and more and it's coming out of your hide. That's why we arrange natural monopolies as public utilities, nobody's going to lay down new water or sewage pipes if 99% in that area already get service from somebody else. Fiber will be the same, enjoy the honeymoon but it won't last very long.
However, and this is where I can speak from personal experience, multi-GPU cards are not always a great use of money
Are they ever? It seems like two cards in CF/SLI have pretty much the exact same performance and drawbacks, with much less of a premium. Now I'd really like a single card that could drive my 4K monitor, but even the 980 Ti/Fury X aren't quite there. I'm guessing I'll stick with what I have until 14/16nm and 8GB of HBM2 is an option.
Nobody is trying to make a general intelligence because nobody wants it. What is wanted is domain specific algorithms that are very good at what they do.
Well it's bits and pieces of it. Imagine you could start to combine Watson and AlphaGo, you tell it to "I'd like to play a round of Go with you" and Watson does the natural language parsing of the request and the rules, finds some games that would make good training material, spawns up an instance of AlphaGo that does the initial training and self-training to improve its play. Yes, the end result is to be able to solve domain specific tasks, but the goal is not to create domain specific solutions but more of a "solution factory". It's still not a generic intelligence that'll learn across the domain-specific ones but a collection of them might mimic it fairly well for a wide variety of tasks.
If the goal was to settle, we'd be priorizing Venus, which offers (...) temperatures that can be - and have been, on 1960s Soviet tech - withstood by simple thermal inertia
...for minutes. No probe has lasted even an hour on the surface, while Opportunity is running for the 13th year and counting on Mars. It's a lot more robot friendly which might be even more important than if it's human-friendly.
You can't effectively teach 1-10th grade math in the same class. So we age-tier because we think that's a reasonable approximation of skill-tier or just to split it up so you've had all the parts of the curriculum. Or think why you have divisions in leagues, you learn by playing roughly equal levels with a few better, a few worse than yourself but if you're just trivially beating them or being crushed you learn nothing. That doesn't mean I think individual learning is the one true answer, you learn a lot of valuable lessons explaining and being explained to, cooperating, correcting each other and so on. But ideally you'd do that with your peers in skill.
If you have some time dedicated to working individually it'd be a lot easier to create a dynamic, personalized schedule where you are in peer groups with others of comparable skill. Today it's mostly impossible to say follow the class of the grade above you because then you have something else, either you must jump a year in every subject or you're pretty much stuck where you are. Of course there's also other concern like like a stable social group you can develop inter-personal relations and skills with but it's not like you exclusively played with those in your class anyway. In the pauses between classes you'd play with other kids anyway.
As I recall, the older versions of the kernel added some features and offered others as third party patches. The 2.0.x, 2.2.x, 2.4.x, and 2.6.x branches were supported for a very long time, (...) This was a very successful development model for over a decade and I don't understand why that's changed. Arguably, it would require fewer work since there wouldn't be as many branches to maintain.
Actually it wasn't. Distros were massively cherry-picking from the odd-numbered branches creating huge variations from the stock kernel, creating strange bugs and making the big jumps was a huge pain because the more relaxed requirements to put it in a development branch led to poor quality. Linus tightened ship and basically said do development in your own branch, when it's ready merge it to the released kernel and ~2 months after the merge window closes it'll be released instead of years like the old kernel. No more hodge-podge development kernels full of half-assed changes. It got distros to work more on the upstream kernel than their own variations, leading to more manpower and higher quality in the core project. It was a great success.
And "gay" is the first word that needs to be reclaimed; as "carefree happiness" it has no synonym.
I think "merry" is pretty close, though apart from expressions like Robin Hood and his merry men and merry Chistmas it's fallen somewhat out of fashion. Blissful, cheerful, jolly are also very much in the vicinity so I don't think you have a shortage of alternatives they're just not used much. And I think we feel a linguistic need to have a formal/informal divide when it comes to sexuality like penis/cock, vagina/pussy, sex/intercourse, gay/homosexual for some reason, so I doubt it goes away unless a different informal term takes over which seems rather unlikely at this point.
General issues are everybody's problem. Issues with one particular piece of hardware is:
a) Not really the problem of anybody else
b) Not something most people can reproduce
c) Not anybody's job
If Dell delivers laptops with Linux preinstalled, then it's their problem, they got the hardware to reproduce it and they got paid people working on it. If $random_user installs $random_distro on $random_laptop, well the manufacturer doesn't care. And while there's always a few people working to make Linux run on everything, they're few and they can't go around buying laptops just because and there's new models all the time. Red Hat will work on supporting the servers that RHEL runs on, they won't generally work on random hardware. And the kernel is mostly driven by paid development, other hardware is very much in the "you want support for that? great, submit a patch and we'll review it" mode.
Some might say that it's because OEMs do not "support ACPI spec properly", but in practice most PCs don't... It could be more practical to just find the patterns that Windows uses, and imitate them.
Says no person who has tried imitating an undocumented binary blob ever. Basically manufacturers just bang the code until it stops crashing, unless you can replicate it exactly which is hopeless in practice you're going to run into random issues. And random issues here aren't just glitches, they're usually crash/hang bugs. I'm sure a lot could be done if you brought the right people together with the right hardware and gave them some money to work on that. But I don't really see who'd do that, because it's not just free time and there's no profit in it.
That's the last kernel that I as an individual was vaguely able to keep up with. Now it's such a bloated mess that unless you've been part of the development team for decades, you're really not going to be able to maintain an understanding. It seems nearly every project in computing goes this way. Oh well, sic transit and stuff.
According to this chart 2.6.32 had ~9.8 MLoC so I'd say you weren't doing too shabby. In fact if you can keep up with over half the current kernel at ~16.8 MLoC, Linus might be hiring...
It would be interesting to set up a Go Turing Test. Either have another top Go player or AlphaGo behind a wall calling the moves.
Can the human champ Lee Se-dol determine if he is playing against a computer or a human... ?
Well at least in the end game the pros were pretty clear that this was not the kind of plays you'd make to try to confuse a 9 dan pro into losing a slightly favorable position. It was forcing Lee Se-dol to counter but all it really did was give him more time to consider the remaining contested areas while playing moves he could blitz if he'd wanted to. Also previously they felt AlphaGo took some really convoluted ways to win where a human would just simplify to claim the win. So when you step out of the game and into the meta-game it seems obvious - to them at least - that you're playing a computer.