If you were trying to change the behavior of the government, it wouldn't make sense to release all the most up to date exploits - as soon as you've done it, you've lost all your leverage. At that point it would be "and if you don't do what I want, I'll continue to release the old exploits that you probably don't care about"... not much of a threat.
By releasing older exploits the message is, "see - I can release your tools to the wild... do what I want or I'll start releasing the exploits you really do care about"
11. Chance to use that body armor you got for christmas!
Seriously, I have about as much desire to go to the movies as to travel by airplane (and get raped by the TSA). The experience has so many downsides these days that it's easy to decide to just do something else rather than put up with all the expense and hassle.
Not all of the problems are new: concession food has always been outrageously expensive (but we always knew that was how the theatre made it's money). But cell phones, obnoxious patrons, night vision equipped theater employees, small screens, and automatic weapons fire... those are mostly new inventions. Seriously, with the big flat screen TV and nice sound system I have, the home experience as a whole is better than the theater experience as a whole. And the difference in price? $5 on Amazon streaming vs $50 or more to go to the theater. Really, they've made it a very easy decision.
But harder to make an excuse to the wife for needing to wear that new body armor!
Yes, I had thought about that, but I like smaller cars (MR2, STi) and the Model S is a huge car. The Model 3 is just about the size I'm looking for... BMW 300 series size, not Lincoln Continental Ocean Liner;-) And besides, I'm assuming Tesla will be charging that much for the Ludicrous version with maximum range...
The Model 3 won't come to the UK before next year anyway, so I'm looking at the end of 2018 to pick one up after they work all the bugs out.
I'm doing the same thing. I'm actually thinking 2019-2020 to get a model 3, specifically because of what you said: Tesla seems to have a lot of problems building a car with good reliability. I tend to like sporty cars that are very reliable (Toyota MR2, Subaru STi). I anticipate a lot of problems with early Model 3s, but am hopeful that Tesla will fairly quickly figure out the problem areas and address them. I already have an electric car, but look forward to one with greater range. A fast 0-60 would be nice. I'm figuring I'm going to have to spend between $50K-70K to get what I want... but I want to be sure it'll be reliable (and that Tesla will be around to fix it when it breaks).
Israel has been doing it for years (including access to your emails), for example. I do not like it, but I don't like my luggage and person being searched either — this is not especially more outrageous.
I actually feel this is a lot worse than searching my luggage and person, although I don't like that either. As our lives increasingly leave digital fingerprints, access to that data conveys a lot about you (which is of course why they want the data). The search surface of my luggage and my body is pretty small. The search surface of social media is pretty darn huge, especially for younger people (and some of my friends who seem to have to post about every little boring detail of their life).
Personally I'd be much happier if they just make sure I'm not carrying any bombs or weapons and not try to determine at the border whether I'm a nice person or not.
I've watched some of those CBP reality shows on Netflix, and I've been amazed at some of the stuff people get turned away (from Canada and Australia I think it was) for. Like... omg, you committed a crime 45 years ago... Sorry, we don't want you in the country... Kinda surprises me that they look/care about stuff so far back. Is that going to happen with social media? "You made a bad posting on Twitter 20 years ago, we don't want you in our country" ???
Yeah, the TV was state of the art when I got it, replaced a 34" CRT. Since then I'm boycotting Sony because of the Sony_BMG_copy_protection_rootkit_scandal. Every now and then they come out with something really nice that I would otherwise be tempted to buy, but I don't because of that dick move they did. They deserved to go out of business at the time, and I'm still hoping they will.
I must have bought that TV back around 2003 - at the time a good plasma would have cost at least $10,000, would have been 780p instead of 1080p, and would have been thicker and used a lot more power. So, at the time it seemed like a good tradeoff even if the black level left a lot to be desired.
Agreed. Few cars make it to 300K without being so completely worn out that a majority of parts would need to be replaced. 900K is completely ridiculous. GP has an agenda and is making up silly requirements to try to support it...
What happens then? Does cold weather affect battery performance? Without an internal combustion engine, the only way to get heat in the cabin is via electricity, which is going to impose a considerable burden on the battery.
Not really...Consider this, if the battery is powerful enough to drive the car around for miles then powering a little heater and fan isn't going to tax it that much. I'm driving my Nissan LEAF in Pennsylvania in 20 degree F weather this week and I drove it all through last winter too. There is an ~10-20% hit to the range in colder weather, but the heater [and my heated steering wheel] isn't that big of deal, certainly no more than running the AC in the summer. It gets along just fine. In fact, last winter I jumped my wife's Honda Accord car with my LEAF when her battery died on a very cold morning... in that regard, EVs are more reliable to at least start in the cold.
I drive a Honda Fit EV and have a slightly different experience. In cold weather (it's been around 0 here in Boston the last few days) I lose about 50% of my range if I don't use the electric cabin heater. I attribute this to:
1) Denser air increasing aerodynamic drag (happens to ICE cars as well) 2) Can't charge the battery to the full 19kwh
Later versions of the Leaf use a heat pump rather than just a resistive cabin heater, so that may explain why your Leaf does better than my Fit. I believe the Tesla uses heat from either the battery pack or electric motor to help warm the cabin - I'm sure a Tesla person will jump in and correct me.
My point is that a compliance car like the Fit EV (so, I'm curious about the Bolt) can take a huge hit if you try to keep the cabin warm. I find that I lose about 1 mile per khw if I keep the cabin at about 65F, my daughter tends to keep it 70-75F when she's driving around town, and that can eat up another mile/kwh...
My only purpose for posting is that this is something that EV manufacturers need to keep in mind - seems like Tesla may have done the best job of this, Nissan addressed it a couple years into production, and some compliance cars don't do anything special and it can make them much less useful during winter months in northern states.
and then I bought a 4K LED backlit LCD with full dimming
I'm waiting for the OLED price to drop a bit (and hope that any kinks get worked out), but the reason I don't want to replace my old shitty-contrast LED TV is that the local dimming ones bother me - in lots of images I see the local dimming effect as an artifact, and it drives me crazy. I'll see a scene where it's apparent and suddenly I'm no longer watching the movie, I'm looking at the screen for more artifacts.
To be fair the "article" does mention Sony's OLED TVs
Not really replying to what you said just responding to the "Sony" part; I currently have a ~12 year old Sony, and they make some good stuff, but when they did the root kit thing in 2005 I swore I would never buy another piece of Sony gear, and I'm sticking with that... They deserved to go out of business over that snafu, so I'll just keep buying stuff from their competition and hope they eventually go out of business.
Not surprising that you might not notice. Question is, would a photographer/graphic designer/architect/interior designer/ etc. notice? Certainly as a part time photographer I notice it a lot!
The other thing is that people are amazingly good at getting used to crap. Many years ago when TVs were all CRTs I was visiting a friend. There was clearly a problem with his TV because the image was very very green. I asked him about it and he didn't know what I was talking about. He had gotten used to it and saw it as "normal". I couldn't believe that he didn't see how screwed up the picture was.
There are clearly changes which are very subtle that the average person won't notice. 4K is probably an example of that, except in very large screens. If you walk an average person into a room with an LED set, and then walk them into a room with an OLED set, yeah, lots of them won't notice, whereas if you show them side by side they probably will notice. It really is a huge difference in image quality, so just because you're used to LED crap, doesn't mean that OLED isn't a big improvement...
Would I pay $4K right now? No, especially because the light output is lower than LED and my screen is in a bright room. But when it hits $2K for a 65 or 70 inch screen, I'll probably upgrade from my LED.
It is a larger change from the Plasma -> LCD switch that happened years ago.
People changed from plasma to LCD? Not me - plasma looks great. LCD looks washed out and can't display black.
So, isn't that exactly the point? OLED gets you most of the visual improvements of Plasma without some of the drawbacks (power, depth).
But it's an incremental upgrade in quality. Clearly most people don't care much about color fidelity, or LCD never would have taken off. Compared to plasma, OLED has some incremental advantages, mostly lower power consumption.
Well, in addition to power consumption, how about price, weight and depth? (and, knowing people, it was probably mostly price).
One remaining problem with OLED is light output. I think it's about 1/2 of a LED display. It's an issue for some people: my tv is in a room that's a wall of glass... so, even the LED is pretty unwatchable during certain hours of the day, which is fine: I only watch it at night so it works out okay, but lots of people probably value the light output over the darker blacks.
Personally, I paid $5K for an early Sony Bravia LED and the black level is pretty horrible, so I've been waiting for OLED, mostly for the prices to drop but also for the technology to become more mature. This might be the year that I pull the trigger.
Yeah.... you are a genius for making this comparison. Nothing has changed in 100 years to make ramping up a car company easier.
Not sure if you're implying that it's easier now to ramp up a car company? I would have guessed the opposite: lots more governmental regulations to meet now. You obviously couldn't bring a Model T car to market today - it would fail on so many fronts it would be crazy to try. Also, people's expectations of even a basic car are pretty high now. No electric start? No air conditioning? No AM/FM/BT/SAT/USB entertainment system? People have pretty high expectations even for a basic car.
Additionally, when the Model T was introduced, there was a huge market of first time buyers available (most people didn't have cars). Now, not so much...
Carnival get's the leftovers, who will complain when poked. There are very few talented people working there for long.
Well, that may be true, but I wonder whether it's because of the qualifications of the staff, or whether the environment sucks so bad that the good people leave after a short stay. If so, that's the company selecting for poor employees.
I'm not IT, but any modern company who treats it as mere overhead does so at their own peril. Remember back in June when SouthWest had computer issues? "Flight delays across our network have resulted in 600-700 canceled and delayed flights". That cost the airline a lot of money and hassle.
Then you think about companies being hacked... you can literally bankrupt the company if you are hacked big time.
Does every IT department need to be all A-Team people? No, and in fact people like that might end up being bored in some shops. But a good solid team that can be depended on for reliable uptime and quick reaction when stuff does go wrong? Seems pretty important to me.
And offshoring like Carnival is doing? Why would you want to put the fate of the company in some external company that you have no control over day-to-day? Yeah, great, you can fire them when a major IT problem hurts you, and you might even be able to recover some of the losses in a lawsuit, but me, I'd much rather put together a solid team of people who have loyalty to the company and can be depended upon to do a good solid job.
Until the robot factories are mass producing killdroids, the people still have the edge in terms of power.
So, say, 6 months from now:-(
Seriously, we're already pretty far down the road of marginalized citizens, and no major riots yet. As much as I'd like to see people stand up and not take it anymore, I'm not holding my breath.
I totally agree. We have the worst of all worlds: we're paying top dollar for mediocre care. We should just look around, see how other countries are doing it, and pick one to copy. I don't care if it's Canada, UK, Germany, Japan, Sweden; just pick one that seems to be working and emulate it. 'Cause the system we have sure as heck isn't working for most people (and as you say - even for businesses except for the healthcare companies).
heheh... reminds me of the startup I was at... they brought in a jerk who didn't do his research and started making hasty decisions. As soon as he showed up I started looking, so luckily was gone by the time he sacked the guy in charge of all the servers. (not really IT, but close enough). Too bad they forgot to get the passwords from him before they sacked him.
Then they tried to go back to the guy and get the passwords - don't remember if they offered him money or not, but to his credit he gave them a nice big "fuck you".
I was at a company that bought a startup/product, let the original guys go, and sent it to India. They tried the "one smart guy who runs the India team". That didn't work, and when they were in jeopardy of being in breach of contract to the tune of several million dollars, they then had to hire a local American team to fix what the Indian team had done.
The thing is, the Indian team didn't cost *that* much less. They could have had a smaller American team from the start for about the same amount of money and they would have avoided the crap that the India engineers wrote. It was like watching a high school kid write code (and the hardware they designed wasn't any better).
But, you know, it probably gets the CEO a bonus at the end of the year because he made the bottom line look briefly better.
Back when Japan was the big competitor that everybody worried about, I was at a startup that didn't make it, and they arranged for the technology group to be sold off to a Japanese company (basically so that engineers would keep their jobs - really good E staff at our company). The point though, is that the Japanese company could articulate their plans not for the next quarter or next year, but for years ahead. They were playing for long term, not short term. Saving a few bucks a quarter so the CEO can give himself a bigger bonus is not the long term way to succeed, but CEOs aren't playing for the company, they're (mostly) in it for themselves. They have no more loyalty for the company than they have for the employees they screw over.
What message does it send when a company like Carnival pulls a dick move like this? They'll have all this great HR stuff about how much they value their employees, but after watching what happens to the IT staff, everybody else realizes what the score really is. And guess what? The remaining people end up with absolutely no loyalty for the company, doing the minimum required to get by, and screw the company. When the employee has to make a decision that might be good for him, but bad for the company... guess which way the decision will go? And the company will absolutely deserve it! If it's everybody for themselves then don't be surprised when the employees goals aren't the same as yours!
There was a local company in the Boston area (Digital Equipment Corporation) who had an amazing CEO. Talk to people who worked for this company and they'll all tell you stories about Ken Olsen. How much he cared about the employees and basically "doing the right thing". Now, hey, they're out of business... but same thing happened to a bunch of their competitors - basically the PC killed them. But the point is, people knew that Ken cared about the little guy. Example: he noticed in one building that all the managers had grabbed the window offices and all the rest of the staff were stuck in the middle with no outside view from their cubicles. At most companies people would have said "yeah, managers are more senior, of course they get the better offices". Instead Ken issued an edict that there would be NO window offices, that only corridors could be next to windows, so that *everybody* could leave their cubicle and look outside for a couple minutes during their workday. Cost the company nothing, but got a lot of goodwill from the employees. There were a ton of stories like that about the guy.
Many modern CEOs don't get how important that is. I worked my butt off when I was at that company and one of the reasons was that I knew Ken Olsen cared about me, a lowly engineer, and so I damn well cared about him and his company and I did my best to make them succeed (which they did while I was there!).
You're incentiveizing people to not bother to top up if they're going to get charged by the minute and they might be hours past due.
Yes, this is exactly the sort of behavior that this policy is trying to prevent.
When people start getting battery stranded what'll that do to the perception of electric cars then
Exactly why this policy is being put into effect. So that asshats won't tie up a charging spot that someone else might need in order to reach home.
I actually agree with the people who are posting that a discharge should start after a certain amount of time goes by. People who can afford a Tesla will be happy to spend $4/hr to get a parking spot right next to the store. But if they think the car will start discharging to the point that they can't make it home... well... that's a pretty good incentive to return to your car in a reasonable amount of time.
Long term, we can solve this with technology. Short term, it's an interesting problem in how to get selfish, privileged, wealthy people to act responsibly in sharing infrastructure with others (while not alienating exactly the sort of people who can afford to buy your cars).
My personal experience was with Level C, but our parent company did Level A autopilot work. The software was tested on special debugging platforms that allowed branches to be exhaustively tested, and this allowed testing that the software was resilient even when a glitch might make a compare/branch take a path that in theory was impossible. So, while the certification certainly was done on a complete system, the software was tested exhaustively by itself, and this was part of the basis for certification. The fact that it would take a year to do a complete test of the software both goes to show the level of inspection being done, and probably would be used as an example by many of how this process can slow development and thus should be avoided. Still, if you want to achieve the kinds of quality that Level A implies, that's what it takes.
I am curious about how you test a neural net, given that the training is presumably as important as the underlying computing platform. Has anyone ever attempted to quantify how you test such a system (besides taking it out on the open road and seeing what happens?).
dual CPU processors running in lock step with validated bit identical inputs and outputs.
That was how the system worked at one fault tolerant computer system company I worked at (each CPU board was married to an identical copy with comparator logic to detect any non-identical results. A pain with modern microprocessors because you had to be sure the two microprocessors were of exactly the same mask level, otherwise things like exceptions could cause a comparison failure (as micro-state was pushed onto the stack). This of course only lets you know the CPU has failed - you don't know which microprocessor is the culprit, so this particular company used some very intelligent software to achieve required uptime while the competitors all used triple redundancy in their processor design.
The certification consultant we had working with us on the Level C product worked on the 777 cert. That was three processing units, each one built out of three CPUs, using three different CPU architectures: Intel 80486, AMD 29050 and Motorola MC68040 (to avoid/detect issues with the implementation of the microprocessor). Each of the three CPUs had one chip of each architecture, the output of which was run through voting hardware, and then the output of THOSE three run through another voter.
Of course, there is a question whether we should relax the standards because when cars crash they only kill a half dozen or so people at a time (usually) rather than the larger numbers of deaths when airplanes fall out of the sky. Also, I think Aviation took a strong stance on avoiding accidents because there was a lot of (well deserved) fear about flying, enough to hurt revenue. For some weird reason, there's never been that fear about automobile fatalities even though the number of deaths each year is staggering.
I keep wondering why an automotive software system for autonomous driving shouldn't have to pass software design/coding standards at least as strict as DO178B Level A, i.e. "Catastrophic – Failure may cause a crash. Error or loss of critical function required to safely fly and land aircraft" with a similar failure rate (10^^9 hr between failures).
The current situation with cars is as if the FAA let Boeing test experimental software on revenue flights of their airplanes. NOBODY would accept that. Why do we accept it with cars?
I'm also a bit curious why Elon is willing to risk the company (due to a major lawsuit which is inevitable) for a feature that's not really necessary to sell his cars. I'm happy to buy a Tesla that doesn't have autonomous features. Just having a great battery electric car is enough for me. Seems like he could ship every Tesla he can build just based on that - the autonomous part is cool and all, but what happens when Tesla gets sued out of existence when some school bus gets clobbered?
I can see it a little more for Uber since it's a way to make a LOT more money (not having to pay their drivers) but in the Tesla case... I just don't get why it's worth the risk to the company?
Agreed... otherwise, if we follow grandparent's argument, my 5 year old can drive on the roads if I call myself the "operator".
In aviation the operator can be the pilot, or it can be the airline he works for. In road vehicles, I think we can all read "operator" as "entity that moves the controls".
My daughter recently got her driver's license here in MA. Not sure I think any of the current autonomous vehicles would pass that test. It was pretty tough.
Slashdot Mirror
If you were trying to change the behavior of the government, it wouldn't make sense to release all the most up to date exploits - as soon as you've done it, you've lost all your leverage. At that point it would be "and if you don't do what I want, I'll continue to release the old exploits that you probably don't care about"... not much of a threat.
By releasing older exploits the message is, "see - I can release your tools to the wild... do what I want or I'll start releasing the exploits you really do care about"
10. Bragging rights.
OP forgot:
11. Chance to use that body armor you got for christmas!
Seriously, I have about as much desire to go to the movies as to travel by airplane (and get raped by the TSA). The experience has so many downsides these days that it's easy to decide to just do something else rather than put up with all the expense and hassle.
Not all of the problems are new: concession food has always been outrageously expensive (but we always knew that was how the theatre made it's money). But cell phones, obnoxious patrons, night vision equipped theater employees, small screens, and automatic weapons fire... those are mostly new inventions. Seriously, with the big flat screen TV and nice sound system I have, the home experience as a whole is better than the theater experience as a whole. And the difference in price? $5 on Amazon streaming vs $50 or more to go to the theater. Really, they've made it a very easy decision.
But harder to make an excuse to the wife for needing to wear that new body armor!
Yes, I had thought about that, but I like smaller cars (MR2, STi) and the Model S is a huge car. The Model 3 is just about the size I'm looking for... BMW 300 series size, not Lincoln Continental Ocean Liner ;-) And besides, I'm assuming Tesla will be charging that much for the Ludicrous version with maximum range...
The Model 3 won't come to the UK before next year anyway, so I'm looking at the end of 2018 to pick one up after they work all the bugs out.
I'm doing the same thing. I'm actually thinking 2019-2020 to get a model 3, specifically because of what you said: Tesla seems to have a lot of problems building a car with good reliability. I tend to like sporty cars that are very reliable (Toyota MR2, Subaru STi). I anticipate a lot of problems with early Model 3s, but am hopeful that Tesla will fairly quickly figure out the problem areas and address them. I already have an electric car, but look forward to one with greater range. A fast 0-60 would be nice. I'm figuring I'm going to have to spend between $50K-70K to get what I want... but I want to be sure it'll be reliable (and that Tesla will be around to fix it when it breaks).
Israel has been doing it for years (including access to your emails), for example. I do not like it, but I don't like my luggage and person being searched either — this is not especially more outrageous.
I actually feel this is a lot worse than searching my luggage and person, although I don't like that either. As our lives increasingly leave digital fingerprints, access to that data conveys a lot about you (which is of course why they want the data). The search surface of my luggage and my body is pretty small. The search surface of social media is pretty darn huge, especially for younger people (and some of my friends who seem to have to post about every little boring detail of their life).
Personally I'd be much happier if they just make sure I'm not carrying any bombs or weapons and not try to determine at the border whether I'm a nice person or not.
I've watched some of those CBP reality shows on Netflix, and I've been amazed at some of the stuff people get turned away (from Canada and Australia I think it was) for. Like... omg, you committed a crime 45 years ago... Sorry, we don't want you in the country... Kinda surprises me that they look/care about stuff so far back. Is that going to happen with social media? "You made a bad posting on Twitter 20 years ago, we don't want you in our country" ???
Can you say FBI?
Yeah, the TV was state of the art when I got it, replaced a 34" CRT. Since then I'm boycotting Sony because of the Sony_BMG_copy_protection_rootkit_scandal. Every now and then they come out with something really nice that I would otherwise be tempted to buy, but I don't because of that dick move they did. They deserved to go out of business at the time, and I'm still hoping they will.
I must have bought that TV back around 2003 - at the time a good plasma would have cost at least $10,000, would have been 780p instead of 1080p, and would have been thicker and used a lot more power. So, at the time it seemed like a good tradeoff even if the black level left a lot to be desired.
Agreed. Few cars make it to 300K without being so completely worn out that a majority of parts would need to be replaced. 900K is completely ridiculous. GP has an agenda and is making up silly requirements to try to support it...
What happens then? Does cold weather affect battery performance? Without an internal combustion engine, the only way to get heat in the cabin is via electricity, which is going to impose a considerable burden on the battery.
Not really...Consider this, if the battery is powerful enough to drive the car around for miles then powering a little heater and fan isn't going to tax it that much. I'm driving my Nissan LEAF in Pennsylvania in 20 degree F weather this week and I drove it all through last winter too. There is an ~10-20% hit to the range in colder weather, but the heater [and my heated steering wheel] isn't that big of deal, certainly no more than running the AC in the summer. It gets along just fine. In fact, last winter I jumped my wife's Honda Accord car with my LEAF when her battery died on a very cold morning... in that regard, EVs are more reliable to at least start in the cold.
I drive a Honda Fit EV and have a slightly different experience. In cold weather (it's been around 0 here in Boston the last few days) I lose about 50% of my range if I don't use the electric cabin heater. I attribute this to:
1) Denser air increasing aerodynamic drag (happens to ICE cars as well)
2) Can't charge the battery to the full 19kwh
Later versions of the Leaf use a heat pump rather than just a resistive cabin heater, so that may explain why your Leaf does better than my Fit. I believe the Tesla uses heat from either the battery pack or electric motor to help warm the cabin - I'm sure a Tesla person will jump in and correct me.
My point is that a compliance car like the Fit EV (so, I'm curious about the Bolt) can take a huge hit if you try to keep the cabin warm. I find that I lose about 1 mile per khw if I keep the cabin at about 65F, my daughter tends to keep it 70-75F when she's driving around town, and that can eat up another mile/kwh...
My only purpose for posting is that this is something that EV manufacturers need to keep in mind - seems like Tesla may have done the best job of this, Nissan addressed it a couple years into production, and some compliance cars don't do anything special and it can make them much less useful during winter months in northern states.
so, do like the Alaskan bush pilots and throw a lighted pan of gasoline under the battery pack 1/2 hour before you want to go for a drive? ;-)
and then I bought a 4K LED backlit LCD with full dimming
I'm waiting for the OLED price to drop a bit (and hope that any kinks get worked out), but the reason I don't want to replace my old shitty-contrast LED TV is that the local dimming ones bother me - in lots of images I see the local dimming effect as an artifact, and it drives me crazy. I'll see a scene where it's apparent and suddenly I'm no longer watching the movie, I'm looking at the screen for more artifacts.
Am I the only one this bothers?
To be fair the "article" does mention Sony's OLED TVs
Not really replying to what you said just responding to the "Sony" part; I currently have a ~12 year old Sony, and they make some good stuff, but when they did the root kit thing in 2005 I swore I would never buy another piece of Sony gear, and I'm sticking with that... They deserved to go out of business over that snafu, so I'll just keep buying stuff from their competition and hope they eventually go out of business.
Sony_BMG_copy_protection_rootkit_scandal
Not surprising that you might not notice. Question is, would a photographer/graphic designer/architect/interior designer/ etc. notice? Certainly as a part time photographer I notice it a lot!
The other thing is that people are amazingly good at getting used to crap. Many years ago when TVs were all CRTs I was visiting a friend. There was clearly a problem with his TV because the image was very very green. I asked him about it and he didn't know what I was talking about. He had gotten used to it and saw it as "normal". I couldn't believe that he didn't see how screwed up the picture was.
There are clearly changes which are very subtle that the average person won't notice. 4K is probably an example of that, except in very large screens. If you walk an average person into a room with an LED set, and then walk them into a room with an OLED set, yeah, lots of them won't notice, whereas if you show them side by side they probably will notice. It really is a huge difference in image quality, so just because you're used to LED crap, doesn't mean that OLED isn't a big improvement...
Would I pay $4K right now? No, especially because the light output is lower than LED and my screen is in a bright room. But when it hits $2K for a 65 or 70 inch screen, I'll probably upgrade from my LED.
It is a larger change from the Plasma -> LCD switch that happened years ago.
People changed from plasma to LCD? Not me - plasma looks great. LCD looks washed out and can't display black.
So, isn't that exactly the point? OLED gets you most of the visual improvements of Plasma without some of the drawbacks (power, depth).
But it's an incremental upgrade in quality. Clearly most people don't care much about color fidelity, or LCD never would have taken off. Compared to plasma, OLED has some incremental advantages, mostly lower power consumption.
Well, in addition to power consumption, how about price, weight and depth? (and, knowing people, it was probably mostly price).
One remaining problem with OLED is light output. I think it's about 1/2 of a LED display. It's an issue for some people: my tv is in a room that's a wall of glass... so, even the LED is pretty unwatchable during certain hours of the day, which is fine: I only watch it at night so it works out okay, but lots of people probably value the light output over the darker blacks.
Personally, I paid $5K for an early Sony Bravia LED and the black level is pretty horrible, so I've been waiting for OLED, mostly for the prices to drop but also for the technology to become more mature. This might be the year that I pull the trigger.
Yeah.... you are a genius for making this comparison. Nothing has changed in 100 years to make ramping up a car company easier.
Not sure if you're implying that it's easier now to ramp up a car company? I would have guessed the opposite: lots more governmental regulations to meet now. You obviously couldn't bring a Model T car to market today - it would fail on so many fronts it would be crazy to try. Also, people's expectations of even a basic car are pretty high now. No electric start? No air conditioning? No AM/FM/BT/SAT/USB entertainment system? People have pretty high expectations even for a basic car.
Additionally, when the Model T was introduced, there was a huge market of first time buyers available (most people didn't have cars). Now, not so much...
Carnival get's the leftovers, who will complain when poked. There are very few talented people working there for long.
Well, that may be true, but I wonder whether it's because of the qualifications of the staff, or whether the environment sucks so bad that the good people leave after a short stay. If so, that's the company selecting for poor employees.
I'm not IT, but any modern company who treats it as mere overhead does so at their own peril. Remember back in June when SouthWest had computer issues? "Flight delays across our network have resulted in 600-700 canceled and delayed flights". That cost the airline a lot of money and hassle.
Then you think about companies being hacked... you can literally bankrupt the company if you are hacked big time.
Does every IT department need to be all A-Team people? No, and in fact people like that might end up being bored in some shops. But a good solid team that can be depended on for reliable uptime and quick reaction when stuff does go wrong? Seems pretty important to me.
And offshoring like Carnival is doing? Why would you want to put the fate of the company in some external company that you have no control over day-to-day? Yeah, great, you can fire them when a major IT problem hurts you, and you might even be able to recover some of the losses in a lawsuit, but me, I'd much rather put together a solid team of people who have loyalty to the company and can be depended upon to do a good solid job.
Until the robot factories are mass producing killdroids, the people still have the edge in terms of power.
So, say, 6 months from now :-(
Seriously, we're already pretty far down the road of marginalized citizens, and no major riots yet. As much as I'd like to see people stand up and not take it anymore, I'm not holding my breath.
I totally agree. We have the worst of all worlds: we're paying top dollar for mediocre care. We should just look around, see how other countries are doing it, and pick one to copy. I don't care if it's Canada, UK, Germany, Japan, Sweden; just pick one that seems to be working and emulate it. 'Cause the system we have sure as heck isn't working for most people (and as you say - even for businesses except for the healthcare companies).
heheh... reminds me of the startup I was at... they brought in a jerk who didn't do his research and started making hasty decisions. As soon as he showed up I started looking, so luckily was gone by the time he sacked the guy in charge of all the servers. (not really IT, but close enough). Too bad they forgot to get the passwords from him before they sacked him.
Then they tried to go back to the guy and get the passwords - don't remember if they offered him money or not, but to his credit he gave them a nice big "fuck you".
Ah, fond memories!
A few points...
I was at a company that bought a startup/product, let the original guys go, and sent it to India. They tried the "one smart guy who runs the India team". That didn't work, and when they were in jeopardy of being in breach of contract to the tune of several million dollars, they then had to hire a local American team to fix what the Indian team had done.
The thing is, the Indian team didn't cost *that* much less. They could have had a smaller American team from the start for about the same amount of money and they would have avoided the crap that the India engineers wrote. It was like watching a high school kid write code (and the hardware they designed wasn't any better).
But, you know, it probably gets the CEO a bonus at the end of the year because he made the bottom line look briefly better.
Back when Japan was the big competitor that everybody worried about, I was at a startup that didn't make it, and they arranged for the technology group to be sold off to a Japanese company (basically so that engineers would keep their jobs - really good E staff at our company). The point though, is that the Japanese company could articulate their plans not for the next quarter or next year, but for years ahead. They were playing for long term, not short term. Saving a few bucks a quarter so the CEO can give himself a bigger bonus is not the long term way to succeed, but CEOs aren't playing for the company, they're (mostly) in it for themselves. They have no more loyalty for the company than they have for the employees they screw over.
What message does it send when a company like Carnival pulls a dick move like this? They'll have all this great HR stuff about how much they value their employees, but after watching what happens to the IT staff, everybody else realizes what the score really is. And guess what? The remaining people end up with absolutely no loyalty for the company, doing the minimum required to get by, and screw the company. When the employee has to make a decision that might be good for him, but bad for the company... guess which way the decision will go? And the company will absolutely deserve it! If it's everybody for themselves then don't be surprised when the employees goals aren't the same as yours!
There was a local company in the Boston area (Digital Equipment Corporation) who had an amazing CEO. Talk to people who worked for this company and they'll all tell you stories about Ken Olsen. How much he cared about the employees and basically "doing the right thing". Now, hey, they're out of business... but same thing happened to a bunch of their competitors - basically the PC killed them. But the point is, people knew that Ken cared about the little guy. Example: he noticed in one building that all the managers had grabbed the window offices and all the rest of the staff were stuck in the middle with no outside view from their cubicles. At most companies people would have said "yeah, managers are more senior, of course they get the better offices". Instead Ken issued an edict that there would be NO window offices, that only corridors could be next to windows, so that *everybody* could leave their cubicle and look outside for a couple minutes during their workday. Cost the company nothing, but got a lot of goodwill from the employees. There were a ton of stories like that about the guy.
Many modern CEOs don't get how important that is. I worked my butt off when I was at that company and one of the reasons was that I knew Ken Olsen cared about me, a lowly engineer, and so I damn well cared about him and his company and I did my best to make them succeed (which they did while I was there!).
Yeah, hey, I agree this is what should happen.
What will happen:
Defence: I ask you again officer. Is it legal for your police department to mislead a court?
Officer: Absolutely not. (he says lying through his teeth)
You're incentiveizing people to not bother to top up if they're going to get charged by the minute and they might be hours past due.
Yes, this is exactly the sort of behavior that this policy is trying to prevent.
When people start getting battery stranded what'll that do to the perception of electric cars then
Exactly why this policy is being put into effect. So that asshats won't tie up a charging spot that someone else might need in order to reach home.
I actually agree with the people who are posting that a discharge should start after a certain amount of time goes by. People who can afford a Tesla will be happy to spend $4/hr to get a parking spot right next to the store. But if they think the car will start discharging to the point that they can't make it home... well... that's a pretty good incentive to return to your car in a reasonable amount of time.
Long term, we can solve this with technology. Short term, it's an interesting problem in how to get selfish, privileged, wealthy people to act responsibly in sharing infrastructure with others (while not alienating exactly the sort of people who can afford to buy your cars).
My personal experience was with Level C, but our parent company did Level A autopilot work. The software was tested on special debugging platforms that allowed branches to be exhaustively tested, and this allowed testing that the software was resilient even when a glitch might make a compare/branch take a path that in theory was impossible. So, while the certification certainly was done on a complete system, the software was tested exhaustively by itself, and this was part of the basis for certification. The fact that it would take a year to do a complete test of the software both goes to show the level of inspection being done, and probably would be used as an example by many of how this process can slow development and thus should be avoided. Still, if you want to achieve the kinds of quality that Level A implies, that's what it takes.
I am curious about how you test a neural net, given that the training is presumably as important as the underlying computing platform. Has anyone ever attempted to quantify how you test such a system (besides taking it out on the open road and seeing what happens?).
dual CPU processors running in lock step with validated bit identical inputs and outputs.
That was how the system worked at one fault tolerant computer system company I worked at (each CPU board was married to an identical copy with comparator logic to detect any non-identical results. A pain with modern microprocessors because you had to be sure the two microprocessors were of exactly the same mask level, otherwise things like exceptions could cause a comparison failure (as micro-state was pushed onto the stack). This of course only lets you know the CPU has failed - you don't know which microprocessor is the culprit, so this particular company used some very intelligent software to achieve required uptime while the competitors all used triple redundancy in their processor design.
The certification consultant we had working with us on the Level C product worked on the 777 cert. That was three processing units, each one built out of three CPUs, using three different CPU architectures: Intel 80486, AMD 29050 and Motorola MC68040 (to avoid/detect issues with the implementation of the microprocessor). Each of the three CPUs had one chip of each architecture, the output of which was run through voting hardware, and then the output of THOSE three run through another voter.
Of course, there is a question whether we should relax the standards because when cars crash they only kill a half dozen or so people at a time (usually) rather than the larger numbers of deaths when airplanes fall out of the sky. Also, I think Aviation took a strong stance on avoiding accidents because there was a lot of (well deserved) fear about flying, enough to hurt revenue. For some weird reason, there's never been that fear about automobile fatalities even though the number of deaths each year is staggering.
I keep wondering why an automotive software system for autonomous driving shouldn't have to pass software design/coding standards at least as strict as DO178B Level A, i.e. "Catastrophic – Failure may cause a crash. Error or loss of critical function required to safely fly and land aircraft" with a similar failure rate (10^^9 hr between failures).
The current situation with cars is as if the FAA let Boeing test experimental software on revenue flights of their airplanes. NOBODY would accept that. Why do we accept it with cars?
I'm also a bit curious why Elon is willing to risk the company (due to a major lawsuit which is inevitable) for a feature that's not really necessary to sell his cars. I'm happy to buy a Tesla that doesn't have autonomous features. Just having a great battery electric car is enough for me. Seems like he could ship every Tesla he can build just based on that - the autonomous part is cool and all, but what happens when Tesla gets sued out of existence when some school bus gets clobbered?
I can see it a little more for Uber since it's a way to make a LOT more money (not having to pay their drivers) but in the Tesla case... I just don't get why it's worth the risk to the company?
Agreed... otherwise, if we follow grandparent's argument, my 5 year old can drive on the roads if I call myself the "operator".
In aviation the operator can be the pilot, or it can be the airline he works for. In road vehicles, I think we can all read "operator" as "entity that moves the controls".
My daughter recently got her driver's license here in MA. Not sure I think any of the current autonomous vehicles would pass that test. It was pretty tough.