You seem to want to punish manufacturers for wanting to stay in business. What you propose is all stick with no carrot and would greatly increase the cost of new products to the detriment of the buyers (possibly majority) who don't really want to have the product last for more than a relatively short time.
It is not new technology that drives the cycle. It is primarily more evolved application of old technology, and it happens fast - much faster than devices usually wear out even when non-repairable.
Most manufacturers want to stay in business with a product long term. To do this, they must continually improve the product. There will always be new versions that incorporate lessons learned or engineering development that was not ready at the time of an earlier product release. Even if the gains are fully in software, how do the companies get paid for the work of making them if they just give those gains to the old products forever? With the current system, they only gain via new sales.
Most consumers want those new features. Many, if not most, throw away or sell/trade their old devices for pennies on the dollar to get those new features - even when the old devices are in "perfect" working condition. This is because they aren't in perfect working condition. They are broke because they don't have the new features. There may be a scratch or two or some minor annoyances like the battery doesn't have quite the umph it used to. But if they look at the cost of fixing those versus getting a new device without those flaws AND with the new features, most will talk themselves into a pricier trade-in deal instead of a cheaper repair. Why? Because they feel they are wasting their money on the cheaper repair that leaves them missing out on new capabilities.
So, from both the manufacturers' and most consumers' point of view, the system is not broken. It is only broken when you throw society's legitimate desire to reduce waste or some individuals' desire to be more frugal with their expenditures and use products for a longer period into the equation.
The question is, how do we serve those interests without slowing development, reducing the manufacturers' profit, or increasing product lifecycle costs to those consumers that would like to have the new best thing every year?
The manufacturer will incur additional costs both in manufacturing for repairability and in supporting older devices. It is unfair for the initial consumer to have to pay those costs unless they do keep the device long term.
So any solution that fairly handles this situation must provide the manufacturer some means of deriving a revenue stream from the continued usage of the device beyond the point where many consumers want the new device. The only way I can think of to do this is to support some system where the manufacturer profits from every repair or software upgrade delivered to the old devices. That means even cheap third party repair centers must in some way be forced to pay back. Anything else simply isn't fair to all parties.
Also, though I write this primarily thinking of smartphones, it is really applicable to just about everything from the lowly Instant Pot up to the latest car models. Most who buy new cars do so because they want a new car, not because the old one is already broke.
I'm a man and the only way I could carry my smartphone in a jeans pocket is via the cargo pocket of a pair of cargo jeans. If you put it in the front pocket, it is likely going to snap in half when you sit. In the back, you're going to smash it on something when you sit.
Most of the time, I'm in shorts given where I live. I buy cargo or hiking shorts and put the phone in the lower pocket. Even there, it has a bad habit of turning sideways where the bend of my leg threatens it (because the pocket is wider than the XL is tall).
I have one pair of shorts with a "device" pocket. The phone sits in it nicely, but anyone walking by could slip it out with ease. In addition, every time I sit down in my car that pocket is slightly upside down and the phone ends up on the floor of the vehicle beside the seat or on the pavement outside.
The general form factor of smartphones sucks for convenience of carry and use while mobile. It would be better to separate the phone and processor from the peripherals. Put the peripherals in some glasses and turn the phone into a pocket computer with a more body conforming shape and a non-slip surface so that it tends to stay in pockets. Then make all input elements use speech or handwaving. Problem solved.
For AI to shine with anything this complicated, you really need detailed and consistent testing. This is where tech needs to go first in medicine. In very many areas, it would help to have a massive increase in testing and fidelity of the tests.
After the success of the genome project, we should have moved to a massive effort to develop cost efficient full system scanning and testing instead of starting the brain project.
We should launch a project to figure out how to measure every aspect of health imaginable in a fashion that allows it to be done as a checkup once a year. It should include things like somehow measuring gene activity at thousands of points within the body, very detailed measurement of cognitive performance to detect changes, detailed hearing and sight measurements (not just clarity but speed of focus, comprehension, etc.), checking the performance of nerves throughout the body, etc. - detail like never before.
Then, we would have the data needed to advance medicine and we'd definitely have a need for AI to interpret that volume of data.
In short, our medical sensor technologies suck. Take $50 billion per year away from defense and spend it on this until the problem is fixed. The payoff would change the world.
As I said, underground. Scoop the dirt out, build the home, put much of the dirt back on top.
My first boss did that in Missouri back in the 80s when building his retirement home. It was not very difficult, actually reduced expense versus other homes he could build at the time, and reduced his electric bills to little more than the lights. I've known others since that have done the same thing.
Even after a really hot forest fire, dig down a few inches and you'll find life. Dirt is a tremendous insulator, heat rises, and the fire usually moves fast enough that it doesn't have to insulate for long.
And, as to the car, that same boss also had his garage underground.
In a fire as severe as the ones California has, I'd fill the vents and entrances in with dirt and leave, but I'd be confident that the damage would be no more than the need to dig back out and clean entrances.
Why don't we just learn the lesson and ban any construction that would cause us to care about the forest fires?
I've known several people who built very cost effective homes where they just scooped a place out, built the home, and put the dirt over the top. As long as no trees are nearby to fall on it, those would likely survive most of these fires. They may have to leave to avoid suffocation, but their home would be fine when they return.
Rather than fight mother nature, respect her and build with the assumption that the worst will always happen over any significant time span.
I can see that. We forget how long things can take.
In relation to Apple, Tesla is right around 1990 or so now. There is a long ways to go. The same slogan could easily be in place in another decade. But, watch out 28 years from now. Perhaps the first $10 trillion company?
uuhhhh, I'm pretty sure he did warn them - in June.
Musk: They have about three weeks before their short position explodes
As usual, a little off in the timing, but his record for making things happen is actually very good if you simply disregard the fact that he always uses the stretch goal time he is pushing for instead of something realistic - a characteristic I believe is the key to success.
In addition, this could be considered a very nice warning that they wouldn't get from other companies. They should be very thankful for this early, unclear disclosure. If he warned of major news coming, they would likely have shorted some more by talking themselves into believing it was a major round of financing or factory build that needed another couple of billion real soon. If he came out with a mostly settled buyout offer for $420, the stock price would have shot to near that and stayed.
Instead, he has created an opportunity to get out of those short positions very near to where the stock was before he made the announcement. It is a huge gift. Will they use it? Of course not. They'll probably double down. Tesla's doom is a religion to them, not an investment strategy.
Welcome to the "explosion" he warned the shorts of in June. ROFLMAO.
The only way he could do that is by offering more stock to the new buyers and watering down the current shares. People would be screaming. That is actually one of the main things the shorts he so nicely warned have been using to try to talk the stock down - that he would water the stock down with new stock offerings in order to raise money.
But, with it being private, they will be better able to throw their own money in now and then. I've estimated Tesla will need at least another $70 billion or so of funding over around 10 years to get where I think he wants to go. These new investors and others down the road will have much more than this $72 billion in the company before it is over, they know that, and they are OK with that.
I think he wants to be mostly TaaS by 2030 or so. He wants the vehicles to be mostly owned by the company and routinely going a million miles on power produced by the company. Basically, he's aiming for a wipeout of both the auto and oil industry as it stands today. A few will be able to reorganize in time or be protected by their nations (like the Chinese ones), but it will mostly collapse.
That is much easier to do as an internationally held private company that doesn't have to publicly answer to stockholders all of the time.
And you're making the exact mistake that everybody today seems to make - short term thinking. If you look at the cost of always having a sofa over your lifetime for example, buying four or five cheap trash ones instead of one that lasts your lifetime and can be handed down is a very different comparison. The one that lasts a lifetime is much cheaper.
When I left home in the 80s, most of the furniture I took with me had been handed down to my parents by my grandparents and was still in perfect condition though few would want to lift it. It weighs a ton. From my point of view, that wasn't just cheap, it was free. My parents replaced it with furniture they still have because they were raised to recognize quality. None of that furniture has yet touched a landfill and my grandparent's furniture was first purchased in the 40s.
And as for your recognition of quality example, as a programmer I can generally recognize quality from one language to the next. Expertise is a different thing. Specifics don't disprove generalizations. In general, I have observed most younger people today judge quality at the surface. I've seen very few feel how heavy something is, turn it over and see how it is made, put your weight on furniture and see if it has any flimsiness, etc. Fifty years ago, that's the type of thing you usually did.
Cars have definitely improved, and I believe will hit the point where most are obsolete or wrecked before they wear out with the electric generation. But those are heavily recycled.
I'm thinking of things like appliances which have seen a huge drop but more things like clothes, tables, chairs, sofas, dressers, beds, children's toys that used to last generations, toy boxes, etc. Tons of the common everyday household items have severely degraded in durability.
I've seen sofas go for 50 years with one reupholstering but saw a recent one that was not cheap get thrown away after 6 because a cheap OSB board used in its construction had busted and it was not easily repairable.
In another case, a delivery man warned on a very expensive hide-a-bed that we should be careful to pull the hide-a-bed out evenly because the support bars were known to bend otherwise creating a non-level sleeping surface that had been the cause of many returns. The support "bars" were very thin tubular metal.
The dumpsters where I live are routinely filled with furniture being thrown away. Many throw away everything rather than moving it when they leave.
If you want to talk "used to"s, we used to buy things that were high quality, durable and repairable and keep them for generations. Recycling was as simple as handing it down.
When people don't produce things for a living, they don't know how to recognize the quality under the pretty paint. That secondary effect is compounds the loss of the industries.
This would not (or at least not for many years into the future) have occurred without "illegal" downloading.
Without the proof that it wouldn't break the internet, they'd have stuck by the mantra that the internet couldn't handle it.
Without the proof of the market, they'd have stuck by the mantra that the market wasn't there.
Without the competition, they would have priced it out of reach.
If the competition of "illegal" downloading were to disappear today, streaming would price itself out of reach and start dying. The presence of some "illegal" activity is critical for the health and survival of many industries. This is definitely one of them.
Maybe. I suspect they just customized the pipeline here to keep things from having to go to external memory and come back in. Human vision processing works in that way. There are several stages of specialized hardware producing results and sending them off to the next stage which operates in parallel. The early stages don't need NNs at all. The NN usage increases as you get to later stages. Using the same HW for every stage means that there is a lot of wasted HW. It also often means that multiple stages that can run simultaneously are trying to use the same cache.
It would be relatively easy to define stages that will likely never move around though the operations within them might change and split them to greatly increase the pipeline performance.
This type of thing has been a key to performance in real-time applications for many decades. Almost all the computers I worked with in avionics in the 80s and 90s used custom bit slice processors and gate arrays to pipeline the operations and perform tasks that would have been impossible with GP processors at the time. It is nothing new. And a large portion of existing automotive processors today are customized for the application.
In essence, I take this as a sign of the beginning of maturation of their tech.
They aren't unplugging a chip. This new chip goes into a new computer that replaces the entire previous computer. The computers are plug compatible, not the chip. So, yes, I've turned off one smart phone on the network and turned on a new one with all the old one's stuff, even the sim. Worked great.
And 10 NVidia boards instead of 1 would drop the range of the vehicle in half and multiply the cost of an already expensive computer by 10. If there is something better in 10 years and the car is still on the road, you just unplug the old and plug in the new - probably for less than half what this one costs.
Every NN is proprietary, and that is where the functionality to worry about is at. The performance on "edge cases" in driving is directly related to how much compute power you can throw at it. Tesla is multiplying its compute power. The edge cases will improve. Staying with the general purpose GPU instead of true NN hardware will guarantee continued unhandled edge cases.
This HW is undoubtedly also more energy efficient. That is the real key. They could stack on more boards, but these units are already consuming a significant amount of the vehicle's energy. The trick is to get more compute power with the same or less energy. NN specific HW is going to be a requirement to have that happen.
Everyone in the industry has known that GPUs will not be used past the first generation or so. They are development HW. Someone will eventually come up with a general purpose NPU that will win the market, but it hasn't happened yet - mostly because NN implementations haven't settled.
Have the wheels being driven by electric motors and if necessary charged by a diesel or gas engine for range extension is a total win.
The perfect example is the "diesel" locomotive which is actually almost universally diesel electric today. Low-end torque is more important in this app than almost any other, and that is why they are hybrids. Even the few that aren't diesel electric are not pure diesel. They are diesel hydraulic or diesel steam. None has a clutch/mechanical transmission type of mechanism.
Pickups and other work vehicles are just getting onto the winning team late because the low-end torque wasn't quite as important to them.
A user misled by a very convincing bot will also be misled by a very convincing human. The problem is, the user is taking the word from something/someone else just because it sounds convincing. That is a vulnerable user no matter what you do for them. Looking out for them just shifts the vulnerability to being vulnerable to the someone/something looking out for them.
The AC did say a $60K electric Honda Accord. I believe the $35K model is advertised to be 5.6 seconds. The long range single motor model is claimed by Tesla to be 5.1 seconds due to the greater amperage capability in its power pack and has actually tested at 4.6 seconds. From that accounting, we can surmise that Tesla is being nice and giving ratings based on a not-so-new battery. When new, it will likely be well under 5.6 seconds. In any case, every single version performs very well for its price point compared to other non-electric cars at the same price point.
In short, EVs in the end will win because dollar-for-dollar, they are going to be more fun.
I've thought some about the idea of decreasing their value, but I'm finding that some of the bots are getting unbelievably sophisticated. I like to periodically review the friends for one of their biggest accounts (which I won't name because it takes away from any ability to discuss) just for fun. I routinely find that over 50% of the friends I investigate are bots. It used to be easy. At first, the bots didn't bother having too many other friends or activity. Now, you have to look for patterns in the friends and activity. A bot is no longer "an" account. It manages a massive network of mixed bot and not bot friends and makes posts that seem somewhat reasonable. I'd bet that if you dig into them with the more in depth access that twitter has, you'll also find they use different IPs scattered in the appropriate country's for the languages used. They are likely using networks of compromised home devices such as the routers that were recently compromised. Harder and harder to beat.
What Honda Accord does 0-60 in 3.5? Every Tesla outperforms every BMW on the road at the same price point and many cars at higher price points. Performance is why electric is suddenly filling the future roadmaps for all car companies. We love performance.
When electric pickups do hit the market, they are going to have to ban them from tractor pulls. Otherwise, a lot of sensitive folks will be real embarrassed. There have already been a lot of incidents when stock Tesla sedans took the slips from people with hot rods and the folks with the hot rods sort of lost it.
You'll see the way non-hybrid, non-EV gas or diesel trucks are advertised change overnight. They will never talk about their towing power again. The low end torque of the electric blows everything else in the consumer market away, and the stability added by the battery packs placed down low will allow them to put more of that torque on the road.
There is no reliable way to "clearly and conspicuously label bots". Even if you require all registered users to prove a human identity, they could still let a bot use their account. A smart bot implementor could even pay users to utilize their identity and to install a VPN host so that the bot can make its posts from their IP.
But, you can be assured that proving identities would still be the first thing they jump to if this is required. It moves the blame. Kaspersky would get the internet where all users have to prove their identities that they have pushed all of these years.
They may also criminalize it. Criminalizing anything that has value usually increases its value. We should know well by now how well that works.
Are bots really the problem? Or is it the way we are making decisions?
I know people who will debate me with statements like "well, so and so had xyz happen, so it happens". First of all, they don't know that so and so told them the truth or even really knows the truth. People rarely remember even their own experiences in a accurate fashion. Second, anecdotal evidence doesn't rise to a level of any statistical relevance in trying to understand any truth about the massive society or complex world we live in. Basing your beliefs or fears on miniscule samples from your personal life or news stories you've seen without tempering it with large sample sets is the path to a very twisted and often unreasonably panicked view of the world.
The problem is that people aren't being taught why we have the scientific method. They aren't being taught just how horribly flawed their reasoning without it is and why our society started exploding in success when we started using it. You can't protect people from that lack of knowledge. If they don't have it, others will always find a way to use their ignorance.
Given that there is more money *per capita* present in this country than ever, the people who have your money are really glad you think that. You'll never find it if they can keep you looking in the wrong direction.
Slashdot Mirror
You seem to want to punish manufacturers for wanting to stay in business. What you propose is all stick with no carrot and would greatly increase the cost of new products to the detriment of the buyers (possibly majority) who don't really want to have the product last for more than a relatively short time.
It is not new technology that drives the cycle. It is primarily more evolved application of old technology, and it happens fast - much faster than devices usually wear out even when non-repairable.
Most manufacturers want to stay in business with a product long term. To do this, they must continually improve the product. There will always be new versions that incorporate lessons learned or engineering development that was not ready at the time of an earlier product release. Even if the gains are fully in software, how do the companies get paid for the work of making them if they just give those gains to the old products forever? With the current system, they only gain via new sales.
Most consumers want those new features. Many, if not most, throw away or sell/trade their old devices for pennies on the dollar to get those new features - even when the old devices are in "perfect" working condition. This is because they aren't in perfect working condition. They are broke because they don't have the new features. There may be a scratch or two or some minor annoyances like the battery doesn't have quite the umph it used to. But if they look at the cost of fixing those versus getting a new device without those flaws AND with the new features, most will talk themselves into a pricier trade-in deal instead of a cheaper repair. Why? Because they feel they are wasting their money on the cheaper repair that leaves them missing out on new capabilities.
So, from both the manufacturers' and most consumers' point of view, the system is not broken. It is only broken when you throw society's legitimate desire to reduce waste or some individuals' desire to be more frugal with their expenditures and use products for a longer period into the equation.
The question is, how do we serve those interests without slowing development, reducing the manufacturers' profit, or increasing product lifecycle costs to those consumers that would like to have the new best thing every year?
The manufacturer will incur additional costs both in manufacturing for repairability and in supporting older devices. It is unfair for the initial consumer to have to pay those costs unless they do keep the device long term.
So any solution that fairly handles this situation must provide the manufacturer some means of deriving a revenue stream from the continued usage of the device beyond the point where many consumers want the new device. The only way I can think of to do this is to support some system where the manufacturer profits from every repair or software upgrade delivered to the old devices. That means even cheap third party repair centers must in some way be forced to pay back. Anything else simply isn't fair to all parties.
Also, though I write this primarily thinking of smartphones, it is really applicable to just about everything from the lowly Instant Pot up to the latest car models. Most who buy new cars do so because they want a new car, not because the old one is already broke.
I'm a man and the only way I could carry my smartphone in a jeans pocket is via the cargo pocket of a pair of cargo jeans. If you put it in the front pocket, it is likely going to snap in half when you sit. In the back, you're going to smash it on something when you sit.
Most of the time, I'm in shorts given where I live. I buy cargo or hiking shorts and put the phone in the lower pocket. Even there, it has a bad habit of turning sideways where the bend of my leg threatens it (because the pocket is wider than the XL is tall).
I have one pair of shorts with a "device" pocket. The phone sits in it nicely, but anyone walking by could slip it out with ease. In addition, every time I sit down in my car that pocket is slightly upside down and the phone ends up on the floor of the vehicle beside the seat or on the pavement outside.
The general form factor of smartphones sucks for convenience of carry and use while mobile. It would be better to separate the phone and processor from the peripherals. Put the peripherals in some glasses and turn the phone into a pocket computer with a more body conforming shape and a non-slip surface so that it tends to stay in pockets. Then make all input elements use speech or handwaving. Problem solved.
For AI to shine with anything this complicated, you really need detailed and consistent testing. This is where tech needs to go first in medicine. In very many areas, it would help to have a massive increase in testing and fidelity of the tests.
After the success of the genome project, we should have moved to a massive effort to develop cost efficient full system scanning and testing instead of starting the brain project.
We should launch a project to figure out how to measure every aspect of health imaginable in a fashion that allows it to be done as a checkup once a year. It should include things like somehow measuring gene activity at thousands of points within the body, very detailed measurement of cognitive performance to detect changes, detailed hearing and sight measurements (not just clarity but speed of focus, comprehension, etc.), checking the performance of nerves throughout the body, etc. - detail like never before.
Then, we would have the data needed to advance medicine and we'd definitely have a need for AI to interpret that volume of data.
In short, our medical sensor technologies suck. Take $50 billion per year away from defense and spend it on this until the problem is fixed. The payoff would change the world.
As I said, underground. Scoop the dirt out, build the home, put much of the dirt back on top.
My first boss did that in Missouri back in the 80s when building his retirement home. It was not very difficult, actually reduced expense versus other homes he could build at the time, and reduced his electric bills to little more than the lights. I've known others since that have done the same thing.
Even after a really hot forest fire, dig down a few inches and you'll find life. Dirt is a tremendous insulator, heat rises, and the fire usually moves fast enough that it doesn't have to insulate for long.
And, as to the car, that same boss also had his garage underground.
In a fire as severe as the ones California has, I'd fill the vents and entrances in with dirt and leave, but I'd be confident that the damage would be no more than the need to dig back out and clean entrances.
It's not like I said don't build there. Just don't build there in a way that fire can destroy.
Why don't we just learn the lesson and ban any construction that would cause us to care about the forest fires?
I've known several people who built very cost effective homes where they just scooped a place out, built the home, and put the dirt over the top. As long as no trees are nearby to fall on it, those would likely survive most of these fires. They may have to leave to avoid suffocation, but their home would be fine when they return.
Rather than fight mother nature, respect her and build with the assumption that the worst will always happen over any significant time span.
Exactly. Nuke it from orbit.
I can see that. We forget how long things can take.
In relation to Apple, Tesla is right around 1990 or so now. There is a long ways to go. The same slogan could easily be in place in another decade. But, watch out 28 years from now. Perhaps the first $10 trillion company?
uuhhhh, I'm pretty sure he did warn them - in June.
Musk: They have about three weeks before their short position explodes
As usual, a little off in the timing, but his record for making things happen is actually very good if you simply disregard the fact that he always uses the stretch goal time he is pushing for instead of something realistic - a characteristic I believe is the key to success.
In addition, this could be considered a very nice warning that they wouldn't get from other companies. They should be very thankful for this early, unclear disclosure. If he warned of major news coming, they would likely have shorted some more by talking themselves into believing it was a major round of financing or factory build that needed another couple of billion real soon. If he came out with a mostly settled buyout offer for $420, the stock price would have shot to near that and stayed.
Instead, he has created an opportunity to get out of those short positions very near to where the stock was before he made the announcement. It is a huge gift. Will they use it? Of course not. They'll probably double down. Tesla's doom is a religion to them, not an investment strategy.
Welcome to the "explosion" he warned the shorts of in June. ROFLMAO.
The only way he could do that is by offering more stock to the new buyers and watering down the current shares. People would be screaming. That is actually one of the main things the shorts he so nicely warned have been using to try to talk the stock down - that he would water the stock down with new stock offerings in order to raise money.
But, with it being private, they will be better able to throw their own money in now and then. I've estimated Tesla will need at least another $70 billion or so of funding over around 10 years to get where I think he wants to go. These new investors and others down the road will have much more than this $72 billion in the company before it is over, they know that, and they are OK with that.
I think he wants to be mostly TaaS by 2030 or so. He wants the vehicles to be mostly owned by the company and routinely going a million miles on power produced by the company. Basically, he's aiming for a wipeout of both the auto and oil industry as it stands today. A few will be able to reorganize in time or be protected by their nations (like the Chinese ones), but it will mostly collapse.
That is much easier to do as an internationally held private company that doesn't have to publicly answer to stockholders all of the time.
This isn't the last takeover trick up his sleeve.
And you're making the exact mistake that everybody today seems to make - short term thinking. If you look at the cost of always having a sofa over your lifetime for example, buying four or five cheap trash ones instead of one that lasts your lifetime and can be handed down is a very different comparison. The one that lasts a lifetime is much cheaper.
When I left home in the 80s, most of the furniture I took with me had been handed down to my parents by my grandparents and was still in perfect condition though few would want to lift it. It weighs a ton. From my point of view, that wasn't just cheap, it was free. My parents replaced it with furniture they still have because they were raised to recognize quality. None of that furniture has yet touched a landfill and my grandparent's furniture was first purchased in the 40s.
And as for your recognition of quality example, as a programmer I can generally recognize quality from one language to the next. Expertise is a different thing. Specifics don't disprove generalizations. In general, I have observed most younger people today judge quality at the surface. I've seen very few feel how heavy something is, turn it over and see how it is made, put your weight on furniture and see if it has any flimsiness, etc. Fifty years ago, that's the type of thing you usually did.
Cars have definitely improved, and I believe will hit the point where most are obsolete or wrecked before they wear out with the electric generation. But those are heavily recycled.
I'm thinking of things like appliances which have seen a huge drop but more things like clothes, tables, chairs, sofas, dressers, beds, children's toys that used to last generations, toy boxes, etc. Tons of the common everyday household items have severely degraded in durability.
I've seen sofas go for 50 years with one reupholstering but saw a recent one that was not cheap get thrown away after 6 because a cheap OSB board used in its construction had busted and it was not easily repairable.
In another case, a delivery man warned on a very expensive hide-a-bed that we should be careful to pull the hide-a-bed out evenly because the support bars were known to bend otherwise creating a non-level sleeping surface that had been the cause of many returns. The support "bars" were very thin tubular metal.
The dumpsters where I live are routinely filled with furniture being thrown away. Many throw away everything rather than moving it when they leave.
If you want to talk "used to"s, we used to buy things that were high quality, durable and repairable and keep them for generations. Recycling was as simple as handing it down.
When people don't produce things for a living, they don't know how to recognize the quality under the pretty paint. That secondary effect is compounds the loss of the industries.
This would not (or at least not for many years into the future) have occurred without "illegal" downloading.
Without the proof that it wouldn't break the internet, they'd have stuck by the mantra that the internet couldn't handle it.
Without the proof of the market, they'd have stuck by the mantra that the market wasn't there.
Without the competition, they would have priced it out of reach.
If the competition of "illegal" downloading were to disappear today, streaming would price itself out of reach and start dying. The presence of some "illegal" activity is critical for the health and survival of many industries. This is definitely one of them.
Maybe. I suspect they just customized the pipeline here to keep things from having to go to external memory and come back in. Human vision processing works in that way. There are several stages of specialized hardware producing results and sending them off to the next stage which operates in parallel. The early stages don't need NNs at all. The NN usage increases as you get to later stages. Using the same HW for every stage means that there is a lot of wasted HW. It also often means that multiple stages that can run simultaneously are trying to use the same cache.
It would be relatively easy to define stages that will likely never move around though the operations within them might change and split them to greatly increase the pipeline performance.
This type of thing has been a key to performance in real-time applications for many decades. Almost all the computers I worked with in avionics in the 80s and 90s used custom bit slice processors and gate arrays to pipeline the operations and perform tasks that would have been impossible with GP processors at the time. It is nothing new. And a large portion of existing automotive processors today are customized for the application.
In essence, I take this as a sign of the beginning of maturation of their tech.
They aren't unplugging a chip. This new chip goes into a new computer that replaces the entire previous computer. The computers are plug compatible, not the chip. So, yes, I've turned off one smart phone on the network and turned on a new one with all the old one's stuff, even the sim. Worked great.
And 10 NVidia boards instead of 1 would drop the range of the vehicle in half and multiply the cost of an already expensive computer by 10. If there is something better in 10 years and the car is still on the road, you just unplug the old and plug in the new - probably for less than half what this one costs.
Every NN is proprietary, and that is where the functionality to worry about is at. The performance on "edge cases" in driving is directly related to how much compute power you can throw at it. Tesla is multiplying its compute power. The edge cases will improve. Staying with the general purpose GPU instead of true NN hardware will guarantee continued unhandled edge cases.
This HW is undoubtedly also more energy efficient. That is the real key. They could stack on more boards, but these units are already consuming a significant amount of the vehicle's energy. The trick is to get more compute power with the same or less energy. NN specific HW is going to be a requirement to have that happen.
Everyone in the industry has known that GPUs will not be used past the first generation or so. They are development HW. Someone will eventually come up with a general purpose NPU that will win the market, but it hasn't happened yet - mostly because NN implementations haven't settled.
Have the wheels being driven by electric motors and if necessary charged by a diesel or gas engine for range extension is a total win.
The perfect example is the "diesel" locomotive which is actually almost universally diesel electric today. Low-end torque is more important in this app than almost any other, and that is why they are hybrids. Even the few that aren't diesel electric are not pure diesel. They are diesel hydraulic or diesel steam. None has a clutch/mechanical transmission type of mechanism.
Pickups and other work vehicles are just getting onto the winning team late because the low-end torque wasn't quite as important to them.
A user misled by a very convincing bot will also be misled by a very convincing human. The problem is, the user is taking the word from something/someone else just because it sounds convincing. That is a vulnerable user no matter what you do for them. Looking out for them just shifts the vulnerability to being vulnerable to the someone/something looking out for them.
The AC did say a $60K electric Honda Accord. I believe the $35K model is advertised to be 5.6 seconds. The long range single motor model is claimed by Tesla to be 5.1 seconds due to the greater amperage capability in its power pack and has actually tested at 4.6 seconds. From that accounting, we can surmise that Tesla is being nice and giving ratings based on a not-so-new battery. When new, it will likely be well under 5.6 seconds. In any case, every single version performs very well for its price point compared to other non-electric cars at the same price point.
In short, EVs in the end will win because dollar-for-dollar, they are going to be more fun.
I've thought some about the idea of decreasing their value, but I'm finding that some of the bots are getting unbelievably sophisticated. I like to periodically review the friends for one of their biggest accounts (which I won't name because it takes away from any ability to discuss) just for fun. I routinely find that over 50% of the friends I investigate are bots. It used to be easy. At first, the bots didn't bother having too many other friends or activity. Now, you have to look for patterns in the friends and activity. A bot is no longer "an" account. It manages a massive network of mixed bot and not bot friends and makes posts that seem somewhat reasonable. I'd bet that if you dig into them with the more in depth access that twitter has, you'll also find they use different IPs scattered in the appropriate country's for the languages used. They are likely using networks of compromised home devices such as the routers that were recently compromised. Harder and harder to beat.
What Honda Accord does 0-60 in 3.5? Every Tesla outperforms every BMW on the road at the same price point and many cars at higher price points. Performance is why electric is suddenly filling the future roadmaps for all car companies. We love performance.
When electric pickups do hit the market, they are going to have to ban them from tractor pulls. Otherwise, a lot of sensitive folks will be real embarrassed. There have already been a lot of incidents when stock Tesla sedans took the slips from people with hot rods and the folks with the hot rods sort of lost it.
You'll see the way non-hybrid, non-EV gas or diesel trucks are advertised change overnight. They will never talk about their towing power again. The low end torque of the electric blows everything else in the consumer market away, and the stability added by the battery packs placed down low will allow them to put more of that torque on the road.
There is no reliable way to "clearly and conspicuously label bots". Even if you require all registered users to prove a human identity, they could still let a bot use their account. A smart bot implementor could even pay users to utilize their identity and to install a VPN host so that the bot can make its posts from their IP.
But, you can be assured that proving identities would still be the first thing they jump to if this is required. It moves the blame. Kaspersky would get the internet where all users have to prove their identities that they have pushed all of these years.
They may also criminalize it. Criminalizing anything that has value usually increases its value. We should know well by now how well that works.
Are bots really the problem? Or is it the way we are making decisions?
I know people who will debate me with statements like "well, so and so had xyz happen, so it happens". First of all, they don't know that so and so told them the truth or even really knows the truth. People rarely remember even their own experiences in a accurate fashion. Second, anecdotal evidence doesn't rise to a level of any statistical relevance in trying to understand any truth about the massive society or complex world we live in. Basing your beliefs or fears on miniscule samples from your personal life or news stories you've seen without tempering it with large sample sets is the path to a very twisted and often unreasonably panicked view of the world.
The problem is that people aren't being taught why we have the scientific method. They aren't being taught just how horribly flawed their reasoning without it is and why our society started exploding in success when we started using it. You can't protect people from that lack of knowledge. If they don't have it, others will always find a way to use their ignorance.
Given that there is more money *per capita* present in this country than ever, the people who have your money are really glad you think that. You'll never find it if they can keep you looking in the wrong direction.