The movie studios were convinced that FreeSync was a reincarnation of the Analog Hole, so AMD tried to invest in Popcorn Time but bought out some lousy movie ticket subscription company on accident.
The obvious question is why they don't limit first-run films to a single showing per day, and then blow it up on the second run? AMC has enough theaters they could get away with it, far fewer people would see it on the first run. The only thing I can think of is that losing concession stand sales to first-run masses going to another theater loses them more than they gain from increased second-run ticket sales. And there'd be potential antitrust issues if they attempted it.
An AMC near me has a theater with reclining plush seats and an HDR laser projector; a ticket is ~$18. If this could be used for that, it'd pay for itself after 2 movies. If it could be paid for month to month, one could wait for a few good films to be showing at one time, maybe subscribe every other month, or only when a 'must-see' film comes out. I'm sure the details will be finessed so that it ends up making AMC more money rather than less; if it turns into a Netflix-style "why pirate when you can see it for free anyway?" deal then it could be a winner.
This could be useful for self-driving cars that aren't quite Level 5. If the human operator notices a problem (car doesn't seem to be stopping for that box of kittens in the road), dumping control to the human with a "take control now" alert might be better than hoping they decide to take the wheel. Being able to detect ErrP AND using eye-tracking to determine what concerning thing the human is looking at, could allow for human-moderated automated avoidance of that thing, even if the software itself doesn't see it/thinks it's safe to ignore.
Progress doesn't always resemble a 100% complete, fully-baked solution. Intelligence has a large continuum with human intelligence at the top, which itself has a range.
Presumably you compared the BBC transcript against another independent source? Otherwise wouldn't an equally valid conclusion be that the BBC added a couple paragraphs in the middle? I'm surprised noone looked into this at the time and called out the guilty party for it. CNN did enough other sketchy stuff with the embedded journalists and the retired Generals that I'd consider it plausible.
I'm not sure it'll become a prop for propaganda, since that'd rely on people trusting its logic, and someone else could give the software the same position and see it come up with different lines of argument. If someone mods it to play strawman, it would be found out, and it wouldn't be trusted when deployed by a biased source. I find it more likely that this'll be used to create a 'script' for a human politician to follow, in speeches or debates, although they probably all do this already with human writers. Listen to any of our recent presidents when they went off-script (Obama was decent at not sounding like an idiot when off-script, I have to give him credit for that.)
It's a socializing tactic, done in order to harmlessly deflate potentially inflated expectations. AKA to 'let someone down gently.' I.e. a ploy to calm your emotions, rather than something directly rational (calming your emotions may be a rational thing to do, however.)
Like journal retractions? It'll have an easier time keeping up with the latest data/retractions from journals, as opposed to doctors who stopped keeping up decades ago. These could be useful on ethics panels, which sometimes have problems keeping their emotions in check.
it also underscores the lack of real value in the level of political discourse that we have today. We spend a lot of time arguing over things we don't really know about.
My fellow Americans, I disagree with the preceding statements because reasons!
Order 2 for Congress and each of its committees. Or just toss out the humans and go straight to Minds acting as our overlords. Sure it may be a while until general AI is as smart as the average voter, but how short until it can surpass the average Congress-critter?
Most IT help desks could be replaced with a normal electronic phone menu system, with one option for "X is down", another for "X broke, send a new one" and another for "I forgot my password".
Chess and some other games like Go have clear win/lose conditions, and are thus easier to train. There are leagues with hierarchies of skill levels, thus an easy way to measure progress of the AI's advancement. Once the techniques are developed that let it dominate at a relatively simple game, those same techniques can be applied to something more nebulous like diagnosing diseases based on an xray, or driving a car, or triaging patients at a hospital. All of those are things AI is being applied to, with some manner of success.
It's socially unacceptable for respected AI researchers to speculate that general AI might arrive in the next few years, and probably will remain so until the day it lands on their doorstep.
If you want to argue that meta-cognition and comprehension are required for 'intelligence', then you're going to need to use a more specific term, like 'wisdom'. Otherwise, recalling rote-memorized data and restating it in a relevant context is sufficient (hey, like in TFA.)
It may be logically impossible to prove that magic doesn't exist, if a scientific definition of 'magic' could even be devised. However, it is easy to point out that no scientific observations have been made of what might scientifically be called 'magic', and that using magic to explain general intelligence resembles a God of the Gaps, given that the gaps in our knowledge of general intelligence keep getting filled in yet 'magic' keeps failing to show up in that new knowledge.
One could argue that magic is inherently inscrutable and thus only the gaps surrounding it will be filled... but evidence for that (that's better than "we still don't know X") should be supplied, or at least a theoretical mechanism of action that is consistent with every other component of observed reality. That said, countless things once considered magic are now well-understood phenomena, and no longer considered magical.
It's a classical skill. However, it should be taught alongside dialectics so that politicians have more tools than endless bickering which is normally only resolved via corruption or quid pro quo. I'd rather we see televised Dialectic (how much a politician can learn, concede and grow) than Debate (how steadfast a politician can be.) Trying to out-hardhead one another leads to noone budging and no progress. One might argue that political deadlock is the ideal state of things... but consider how much shit gets logrolled into 'must-pass' legislation to figure out the fatal flaw with this strategy. And then there's 'let no tragedy go to waste.'
Why use Bitcoin and not something like Ripple or Ethereum that has 1% or less of Bitcoin's transaction fees? The obvious answer is name recognition, people know what Bitcoin is. However, they could advertise Bitcoin, and then try to steer the actual users towards superior cryptocurrencies once they actually try to use Bitcoin.
Slashdot Mirror
Some hoodlums the next table over did a 51% attack on my burger.
Pillarboxing FTW!
Hmm... the Hayabusa2 spacecraft is rendezvousing with the Ryugu asteroid? I wonder if someone at JAXA was a fan of Ninja Gaiden
The movie studios were convinced that FreeSync was a reincarnation of the Analog Hole, so AMD tried to invest in Popcorn Time but bought out some lousy movie ticket subscription company on accident.
The obvious question is why they don't limit first-run films to a single showing per day, and then blow it up on the second run? AMC has enough theaters they could get away with it, far fewer people would see it on the first run. The only thing I can think of is that losing concession stand sales to first-run masses going to another theater loses them more than they gain from increased second-run ticket sales. And there'd be potential antitrust issues if they attempted it.
An AMC near me has a theater with reclining plush seats and an HDR laser projector; a ticket is ~$18. If this could be used for that, it'd pay for itself after 2 movies. If it could be paid for month to month, one could wait for a few good films to be showing at one time, maybe subscribe every other month, or only when a 'must-see' film comes out. I'm sure the details will be finessed so that it ends up making AMC more money rather than less; if it turns into a Netflix-style "why pirate when you can see it for free anyway?" deal then it could be a winner.
This could be useful for self-driving cars that aren't quite Level 5. If the human operator notices a problem (car doesn't seem to be stopping for that box of kittens in the road), dumping control to the human with a "take control now" alert might be better than hoping they decide to take the wheel. Being able to detect ErrP AND using eye-tracking to determine what concerning thing the human is looking at, could allow for human-moderated automated avoidance of that thing, even if the software itself doesn't see it/thinks it's safe to ignore.
for $10.18, or wait until 5pm and pay $8.15, about 25% less
No, $8.15 is 19.94% less than $10.18 /pedant
However, $10.18 is ~25% MORE than $8.15
Progress doesn't always resemble a 100% complete, fully-baked solution. Intelligence has a large continuum with human intelligence at the top, which itself has a range.
Does that means it uses ad nominem attacks when he runs out of good points?
How's he going to win a debate if he attacks the nominators?
Presumably you compared the BBC transcript against another independent source? Otherwise wouldn't an equally valid conclusion be that the BBC added a couple paragraphs in the middle? I'm surprised noone looked into this at the time and called out the guilty party for it.
CNN did enough other sketchy stuff with the embedded journalists and the retired Generals that I'd consider it plausible.
I'm not sure it'll become a prop for propaganda, since that'd rely on people trusting its logic, and someone else could give the software the same position and see it come up with different lines of argument. If someone mods it to play strawman, it would be found out, and it wouldn't be trusted when deployed by a biased source. I find it more likely that this'll be used to create a 'script' for a human politician to follow, in speeches or debates, although they probably all do this already with human writers. Listen to any of our recent presidents when they went off-script (Obama was decent at not sounding like an idiot when off-script, I have to give him credit for that.)
It's a socializing tactic, done in order to harmlessly deflate potentially inflated expectations. AKA to 'let someone down gently.' I.e. a ploy to calm your emotions, rather than something directly rational (calming your emotions may be a rational thing to do, however.)
Like journal retractions? It'll have an easier time keeping up with the latest data/retractions from journals, as opposed to doctors who stopped keeping up decades ago. These could be useful on ethics panels, which sometimes have problems keeping their emotions in check.
The lobbyist-bot would resemble an ATM. And instead of going to $1000-a-plate dinners it'd go to $1000-a-CHAdeMO-port charging parties.
it also underscores the lack of real value in the level of political discourse that we have today. We spend a lot of time arguing over things we don't really know about.
My fellow Americans, I disagree with the preceding statements because reasons!
I'd suggest the Voter Apathy Party, but am afraid you'd get worked up over it.
Order 2 for Congress and each of its committees. Or just toss out the humans and go straight to Minds acting as our overlords. Sure it may be a while until general AI is as smart as the average voter, but how short until it can surpass the average Congress-critter?
Most IT help desks could be replaced with a normal electronic phone menu system, with one option for "X is down", another for "X broke, send a new one" and another for "I forgot my password".
Chess and some other games like Go have clear win/lose conditions, and are thus easier to train. There are leagues with hierarchies of skill levels, thus an easy way to measure progress of the AI's advancement. Once the techniques are developed that let it dominate at a relatively simple game, those same techniques can be applied to something more nebulous like diagnosing diseases based on an xray, or driving a car, or triaging patients at a hospital. All of those are things AI is being applied to, with some manner of success.
It's socially unacceptable for respected AI researchers to speculate that general AI might arrive in the next few years, and probably will remain so until the day it lands on their doorstep.
Proud Philosophical Zombie here!
If you want to argue that meta-cognition and comprehension are required for 'intelligence', then you're going to need to use a more specific term, like 'wisdom'. Otherwise, recalling rote-memorized data and restating it in a relevant context is sufficient (hey, like in TFA.)
It may be logically impossible to prove that magic doesn't exist, if a scientific definition of 'magic' could even be devised. However, it is easy to point out that no scientific observations have been made of what might scientifically be called 'magic', and that using magic to explain general intelligence resembles a God of the Gaps, given that the gaps in our knowledge of general intelligence keep getting filled in yet 'magic' keeps failing to show up in that new knowledge.
One could argue that magic is inherently inscrutable and thus only the gaps surrounding it will be filled... but evidence for that (that's better than "we still don't know X") should be supplied, or at least a theoretical mechanism of action that is consistent with every other component of observed reality. That said, countless things once considered magic are now well-understood phenomena, and no longer considered magical.
It's a classical skill. However, it should be taught alongside dialectics so that politicians have more tools than endless bickering which is normally only resolved via corruption or quid pro quo. I'd rather we see televised Dialectic (how much a politician can learn, concede and grow) than Debate (how steadfast a politician can be.) Trying to out-hardhead one another leads to noone budging and no progress. One might argue that political deadlock is the ideal state of things... but consider how much shit gets logrolled into 'must-pass' legislation to figure out the fatal flaw with this strategy. And then there's 'let no tragedy go to waste.'
Why use Bitcoin and not something like Ripple or Ethereum that has 1% or less of Bitcoin's transaction fees? The obvious answer is name recognition, people know what Bitcoin is. However, they could advertise Bitcoin, and then try to steer the actual users towards superior cryptocurrencies once they actually try to use Bitcoin.