In years past, I've taken several vacation road trips topping 3000 miles that required carrying large volumes of equipment. Renting a luxury type vehicle for a couple of weeks that is perfect for the road trip and that I could not afford to actually own actually cost me less than putting that same mileage on the vehicle that I could afford to drive every day. Sure, I had to shop around to find rental deals that gave me a massive SUV with trailer tow capacity without killing me on mileage fees, but the needed deal has always been there and saved me enough money versus making do with my own vehicle to make the search very worth it.
I'd much rather take a road trip in luxury that I couldn't normally afford due to its features and low gas mileage than making do with the everyday vehicle I can afford. It is a vacation after all.
I will happily pay a bit more than average for awesome acceleration, quiet ride, state of the art functions, to never again have to visit a gas station, and to not have to be dealing with the thousands of dollars and multiple weeks in the shop that I've been encountering on our 100K mile ICE vehicle every year despite the fact that it isn't driven more than 10K miles in a year. I have noted that the majority of the post-warranty maintenance cost that people have been encountering on EVs are costs that I rarely bother to incur on my ICE. The story we are getting on EV repairs right now is from an unusually high class owner since prices have only recently dipped low enough to get the story from an average level owner. 100% of my ICE repairs that eat me alive are drive train related but only because that's all that has been important enough to repair. What normal driver actually fixes the dents, dings, scratches, etc that vehicles accumulate over time? The vehicle only has to go about 14 years because that is when virtually every rubber and plastic part in the vehicle starts failing in my climate due to age regardless of mileage, and it has to be dumped. If it reliably drives and I'm not getting wet inside, I'm good. I'm not rich.
For the last couple of decades, the math has not once worked out in favor of using my own vehicle for a long trip. I have always been able to find a rental, usually tailor picked for the exact needs of my trip, that charges either no mileage or low-enough mileage that the cost per mile works out much better for renting a vehicle perfect for the trip than for eating up my own mileage. Choosing the vehicles you drive every day on the basis of occasional needs better met by vehicles chosen for those needs alone is silly. I apply a bit more reason that that to an investment that for most sucks up around 1/4 of their income.
And that is a ridiculous discussion that does not speak to the average driver. I don't claim to be an average driver. I do drive almost every day, but the last time I drove more than 200 miles in a day was Thanksgiving of 2016.
Even Intel doesn't believe this. If they did, they wouldn't have so desperate to combine the Spectre and Meltdown issues into one announcement and jump the gun on the announcement in order to obfuscate the fact that their exposure to these problems was and is much greater.
Is it speculative execution that is the problem or backtracking? What if we were to pull the speculative execution out of the time domain and put in parallel in space? If we were to include multiple pipelines in the same core, we could speculatively execute multiple branches in parallel and just dump all but the good one when we find out which is right. Then do a single cycle copy of all stages of the good pipeline over all pipelines to resync. With completely separate HW, there would be no potential for one speculation to gain knowledge of another.
This approach would allow speculative execution with separation of HW and the timing would appear as though the correct branch was always taken. The downside is that much more power and HW is being used per core. This may be acceptable to produce what would likely be the fastest IPC rating ever. At least in desktop applications, we're not so desperate for more cores anymore. Cores that do more could be very good.
That would make a case for letting people use the autopilot sooner in cars. There is a far better chance of an aircraft pilot being safer than the autopilot than of a driver being safer than the autopilot.
We've already absolutely proven through decades of example that we can do little to control whether people drive drunk, talk on phones, text, etc. while driving. Pilots have a much better (though still not perfect because, hey, human) record at this sort of thing.
Anyway, the original point stands. Aircraft autopilots have always required an attentive pilot to be present at all times and have always been called autopilots.
What you need is a camper made for a Tesla. It could have its own battery pack. Of course, Tesla would have to provide some sort of kit to allow the connection of a trailer battery somewhere near the hitch. But, it would be cool.
Speaking of campers, every time I pass a camper struggling on a grade or pulled-over to cool the transmission and/or brakes, I scratch my head as to why the great American motor home was not the first type of vehicle to go full-on EV. Neither weight nor space is near the concern that it is for smaller vehicles. Many of them cost enough that a 250+ kWh battery pack wouldn't be much of a blip in their cost. They don't go very far on most days. They tend to park every night at a site that already has a 50A plug. They desperately need a torque upgrade for the mountains their owners always love to take them over. They could regain a lot of that energy from going up when they go down the other side, etc. etc.
Torque is just one aspect of a tow rating. Though your example is one that doesn't actually require a lot of torque and is only good for show to the naive, EVs do have much better low-end torque than other technologies. This is one of the reasons virtually all trains are diesel-electric hybrids.
But, the towing capacity is also dependent on how much weight the vehicle can hold and control at the hitch when doing crazy things like slamming on the brakes while going around sharp curves or weaving left and right at speed to go around an obstacle that suddenly appears. Towing more than your capacity can allow for things like side-to-side oscillations to start up while driving down the highway that can eventually go out of control and jackknife your vehicle. Or there may be braking situations that cause the weight to bear down on the hitch and pop your front wheels off of the ground causing immediate loss of all steering control.
I suspect Teslas with their low center of gravity and the high mass of the batteries will also be king in these areas, but not so much that you could ever have a true tow rating of something ridiculous like 100K pounds from a Model Y, X, or even the upcoming pickup truck.
First, you proved my point. Autopilot does not mean you can dispense of the pilot.
No commercially approved system can be used for takeoff due to regulation. There absolutely are systems that are capable of it, even from carriers, and many would argue that some of the very advanced ones are safer.
As for the obstacles in the air issue, yes, I have worked on semi-autonomous flight systems and understand. Though it isn't like the obstacles usually appear from behind buildings, trees, and other vehicles or from around curves.
As for how much the pilots are relied on for spotting obstacles, I worked with high speed, ground-hugging flight systems over 30 years ago that could not have performed their mission if the pilot had been relied on for spotting obstacles. Even then, yes, they failed. Having a large bird penetrate your cockpit when you're 100 ft off the ground tends to result in an immediate need for ejection - if you're lucky.
The word "autopilot" has a history. It has been applied to aircraft controls since the days when it was barely able to maintain straight and level flight. Even today, aircraft autopilots do not enable what we call "level 5" though only because regulations don't allow it. But it is fully right and proper to call them autopilots because the word does not mean and has never meant that you don't have to have a pilot.
I agree with Elon on this one. Today's aircraft autopilots are fully capable of piloting an aircraft from takeoff to landing and could likely handle gate-to-gate with just a little work. They can even handle takeoff and landing on carriers.
Yet, even in the air with almost nothing to hit, we still require an attentive pilot in the cockpit at all times. Aircraft autopilots are not allowed by regulation to operate as level 5. But we still call them autopilots and have done so since the days in which they were far less capable including being completely unable to handle their equivalent to city traffic - takeoff and landing.
The automobile application is not so different from the aircraft application to warrant a radical redefinition of an already common word.
Not genetically, epigenetically. New stuff here. It is not the DNA itself being changed. Instead, regulatory molecules are being attached to the DNA.
We have discovered that our systems are capable of adapting to environmental stresses by sending a signal throughout the body that causes gene expression to be changed via mechanisms like DNA methylation and that those signals target sperm and eggs as well as the other body cells. We are finding that this is why things like PTSD are near impossible to reverse. The body has epigenetically forced defensive mechanisms into overdrive. People who then parent children after experience of this kind of trauma can then pass on the over-activation of those defenses. Several studies have shown this to be true and they have even identified some of the signals now and succeeded in blocking them in mice.
Once these changes in expression occur, it can take a few generations for them to be cleaned out of the offspring.
That level of ingraining would not just be in the culture. Survival would favor those who are more passive and genetic selection would occur. Epigenetics would play a part too when traumas endured by parents are strong enough to turn on survival mechanisms whose activation is passed on to their children.
Just being in the lobby when you start to post the review doesn't seem like much of a barrier to me unless you didn't actually go to see the movie at all. I am very fond of the saying "trust, but verify".
They could take this a step further by requiring some indication of actual attendance. This might be as simple as only accepting reviews posted by the app from the theater during or within a half an hour of a showing until the movie has been released on DVD or for streaming. Or perhaps every review could require a photo of a unique ticket stub assuming there is some way to validate them reliably.
The audience score is the only aspect of Rotten Tomatoes I find to be of value. The scores from movie critics just don't seem to have a strong relationship with whether or not I'm going to enjoy a movie. Actually, maybe that's not quite correct. It does sometimes seem like a movie with a very low critics score and a high audience score is usually great. So there may be a strong inverse relationship.
I think the chicken and egg issue will dominate though. Until a large portion of the datacenter systems are ARM, there would be no compelling reason for a developer to switch their development platform and many compelling reasons not too. And until a large number of developers are on ARM, the datacenters would be fighting the developer's platform if they switch.
Why would I buy a development system as premiumly-priced as the Macs to target a platform that might be successful in a few years? These things don't happen overnight.
I CAN play devil's advocate with myself here. I do realize that this will help front end developers, and that is almost certainly in Apple's thoughts. But, in my experience, the back end is where the real tech is. If front end developers jump to ARM-based Macs because it makes their jobs easier, we'll see the already damaging gulf between front and back end development widen. That would be a bad result for the industry. Of course, I say this as more of a full-stack guy that believes it is much harder to develop a quality product when nobody on the team fully understands both worlds.
Linus Torvalds has stated that ARM won't win the server space because developers want to run their apps on the architecture it has been developed on and almost all are developing on x86. Many application bugs are still architecture specific. Application performance optimization is also highly architecture specific, especially for database applications.
Given the Mac's popularity among developers, this argument should apply to the Macs too when looked at from the opposite angle. The vast majority of servers are x86, and developers want to run their apps on the architecture they are developing for. Running in an emulator is nowhere near the same experience. I would think a switch from x86 to ARM would decimate the number of developers calling the Mac home.
Separately, I don't see the appeal of running phone apps on my laptop or desktop. Smartphone apps do not have the feature density that I'm looking for with a desktop app and desktop apps are not generally appropriate for smartphones. On my desktop, I don't want simplicity. I want to see everything I can at once and to be able to do almost everything with my keyboard.
True. That would cause the battery pack to have a bit less range than a fresh one. However, this works out a bit like buying prewashed versus fresh jeans. The prewashed will fit just as nice after a few washes but the ones that were not prewashed will be too tight.
The loss of capacity in a fresh pack is much faster than in one that has already made it down below 90% or so. The curve flattens out a lot. A 100 kWh battery that is down to 80 kWh due to having moved down to the 80% point in its curve will stay near that 80 kWh level for far longer than one that is starting out at 80 kWh. After some time, the true 80 kWh will be a 70 kWh battery and the 100 kWh battery may have just fallen from 80 kWh to 78 kWh.
A 100 kWh battery that is has reached 80% capacity due to having a few hundred thousand miles of use would be an awesome used purchase for the 80 kWh version of the same vehicle. When it drops to 80% of what that vehicle expects, send it on to a utility storage operation that uses acres of space and doesn't care as much about power density.
Using this approach, vehicle batteries might not have 9-lives, but they should have three or four.
There is already a market for used battery packs. As the volume of used packs available reaches thousands, entrepreneurs will come along and realize they could build racks and control systems ready made to accept Tesla (for example) battery packs, purchase 1,000 used 100 kWh packs, and deploy an 80 MWh utility backup system for a fraction of the cost of one using brand new batteries. This will push the recycling problem way back.
I had the same thought. It seems that many businesses are no longer doing their best to put their competition out of business in order to increase profits. Instead, they concentrate on ways to increase profits through internal changes without changes in sales. Many also no longer push to innovate. R&D cost money. I guess it is much easier to effect the stock on 3-month timeframes when you just silently agree to a truce with your competition and work on reducing internal costs and using accounting tricks to increase profit.
The thought that net neutrality will be an issue for the 2020 presidential election instead of being put to bed this year is very discouraging. The speed increase I get when using a VPN connection with streaming services is steadily increasing.
On a related note, I'm having frequent connection issues, up to a few times a week now for around an hour at a time, using multiple major public DNS systems including Cloudflare, Google, and OpenDNS. Interestingly, I can get to them through the VPN when I can't get to them directly. I believe my provider is interfering with competitive DNS systems that are taking away information that my provider sells.
Pretty much what I came to say. It would fit with the Adobe culture to have their product rely on a custom instruction set on a device in a USB dongle to execute using the excuse that it is a critical "accelerator".
I've probably been using gmail as long as you. Might have even got the invitation from the same place.
I'm fairly positive that all spam goes to the spam folder. I can't remember the last time I had to mark a spam message. I do look at the spam folder periodically for its entertainment value and to keep up with the flavor of the current scams. I haven't had a false positive in many years.
Slashdot Mirror
In years past, I've taken several vacation road trips topping 3000 miles that required carrying large volumes of equipment. Renting a luxury type vehicle for a couple of weeks that is perfect for the road trip and that I could not afford to actually own actually cost me less than putting that same mileage on the vehicle that I could afford to drive every day. Sure, I had to shop around to find rental deals that gave me a massive SUV with trailer tow capacity without killing me on mileage fees, but the needed deal has always been there and saved me enough money versus making do with my own vehicle to make the search very worth it.
I'd much rather take a road trip in luxury that I couldn't normally afford due to its features and low gas mileage than making do with the everyday vehicle I can afford. It is a vacation after all.
$30,000? The AVERAGE new US vehicle is over $36K this year. The fact that you're having to dip so far down the distribution curve to make your argument actually makes the EVs are here argument.
I will happily pay a bit more than average for awesome acceleration, quiet ride, state of the art functions, to never again have to visit a gas station, and to not have to be dealing with the thousands of dollars and multiple weeks in the shop that I've been encountering on our 100K mile ICE vehicle every year despite the fact that it isn't driven more than 10K miles in a year. I have noted that the majority of the post-warranty maintenance cost that people have been encountering on EVs are costs that I rarely bother to incur on my ICE. The story we are getting on EV repairs right now is from an unusually high class owner since prices have only recently dipped low enough to get the story from an average level owner. 100% of my ICE repairs that eat me alive are drive train related but only because that's all that has been important enough to repair. What normal driver actually fixes the dents, dings, scratches, etc that vehicles accumulate over time? The vehicle only has to go about 14 years because that is when virtually every rubber and plastic part in the vehicle starts failing in my climate due to age regardless of mileage, and it has to be dumped. If it reliably drives and I'm not getting wet inside, I'm good. I'm not rich.
For the last couple of decades, the math has not once worked out in favor of using my own vehicle for a long trip. I have always been able to find a rental, usually tailor picked for the exact needs of my trip, that charges either no mileage or low-enough mileage that the cost per mile works out much better for renting a vehicle perfect for the trip than for eating up my own mileage. Choosing the vehicles you drive every day on the basis of occasional needs better met by vehicles chosen for those needs alone is silly. I apply a bit more reason that that to an investment that for most sucks up around 1/4 of their income.
And that is a ridiculous discussion that does not speak to the average driver. I don't claim to be an average driver. I do drive almost every day, but the last time I drove more than 200 miles in a day was Thanksgiving of 2016.
Even Intel doesn't believe this. If they did, they wouldn't have so desperate to combine the Spectre and Meltdown issues into one announcement and jump the gun on the announcement in order to obfuscate the fact that their exposure to these problems was and is much greater.
Is it speculative execution that is the problem or backtracking? What if we were to pull the speculative execution out of the time domain and put in parallel in space? If we were to include multiple pipelines in the same core, we could speculatively execute multiple branches in parallel and just dump all but the good one when we find out which is right. Then do a single cycle copy of all stages of the good pipeline over all pipelines to resync. With completely separate HW, there would be no potential for one speculation to gain knowledge of another.
This approach would allow speculative execution with separation of HW and the timing would appear as though the correct branch was always taken. The downside is that much more power and HW is being used per core. This may be acceptable to produce what would likely be the fastest IPC rating ever. At least in desktop applications, we're not so desperate for more cores anymore. Cores that do more could be very good.
That would make a case for letting people use the autopilot sooner in cars. There is a far better chance of an aircraft pilot being safer than the autopilot than of a driver being safer than the autopilot.
We've already absolutely proven through decades of example that we can do little to control whether people drive drunk, talk on phones, text, etc. while driving. Pilots have a much better (though still not perfect because, hey, human) record at this sort of thing.
Anyway, the original point stands. Aircraft autopilots have always required an attentive pilot to be present at all times and have always been called autopilots.
What you need is a camper made for a Tesla. It could have its own battery pack. Of course, Tesla would have to provide some sort of kit to allow the connection of a trailer battery somewhere near the hitch. But, it would be cool.
Speaking of campers, every time I pass a camper struggling on a grade or pulled-over to cool the transmission and/or brakes, I scratch my head as to why the great American motor home was not the first type of vehicle to go full-on EV. Neither weight nor space is near the concern that it is for smaller vehicles. Many of them cost enough that a 250+ kWh battery pack wouldn't be much of a blip in their cost. They don't go very far on most days. They tend to park every night at a site that already has a 50A plug. They desperately need a torque upgrade for the mountains their owners always love to take them over. They could regain a lot of that energy from going up when they go down the other side, etc. etc.
Torque is just one aspect of a tow rating. Though your example is one that doesn't actually require a lot of torque and is only good for show to the naive, EVs do have much better low-end torque than other technologies. This is one of the reasons virtually all trains are diesel-electric hybrids.
But, the towing capacity is also dependent on how much weight the vehicle can hold and control at the hitch when doing crazy things like slamming on the brakes while going around sharp curves or weaving left and right at speed to go around an obstacle that suddenly appears. Towing more than your capacity can allow for things like side-to-side oscillations to start up while driving down the highway that can eventually go out of control and jackknife your vehicle. Or there may be braking situations that cause the weight to bear down on the hitch and pop your front wheels off of the ground causing immediate loss of all steering control.
I suspect Teslas with their low center of gravity and the high mass of the batteries will also be king in these areas, but not so much that you could ever have a true tow rating of something ridiculous like 100K pounds from a Model Y, X, or even the upcoming pickup truck.
First, you proved my point. Autopilot does not mean you can dispense of the pilot.
No commercially approved system can be used for takeoff due to regulation. There absolutely are systems that are capable of it, even from carriers, and many would argue that some of the very advanced ones are safer.
As for the obstacles in the air issue, yes, I have worked on semi-autonomous flight systems and understand. Though it isn't like the obstacles usually appear from behind buildings, trees, and other vehicles or from around curves.
As for how much the pilots are relied on for spotting obstacles, I worked with high speed, ground-hugging flight systems over 30 years ago that could not have performed their mission if the pilot had been relied on for spotting obstacles. Even then, yes, they failed. Having a large bird penetrate your cockpit when you're 100 ft off the ground tends to result in an immediate need for ejection - if you're lucky.
The word "autopilot" has a history. It has been applied to aircraft controls since the days when it was barely able to maintain straight and level flight. Even today, aircraft autopilots do not enable what we call "level 5" though only because regulations don't allow it. But it is fully right and proper to call them autopilots because the word does not mean and has never meant that you don't have to have a pilot.
I agree with Elon on this one. Today's aircraft autopilots are fully capable of piloting an aircraft from takeoff to landing and could likely handle gate-to-gate with just a little work. They can even handle takeoff and landing on carriers.
Yet, even in the air with almost nothing to hit, we still require an attentive pilot in the cockpit at all times. Aircraft autopilots are not allowed by regulation to operate as level 5. But we still call them autopilots and have done so since the days in which they were far less capable including being completely unable to handle their equivalent to city traffic - takeoff and landing.
The automobile application is not so different from the aircraft application to warrant a radical redefinition of an already common word.
Not genetically, epigenetically. New stuff here. It is not the DNA itself being changed. Instead, regulatory molecules are being attached to the DNA.
We have discovered that our systems are capable of adapting to environmental stresses by sending a signal throughout the body that causes gene expression to be changed via mechanisms like DNA methylation and that those signals target sperm and eggs as well as the other body cells. We are finding that this is why things like PTSD are near impossible to reverse. The body has epigenetically forced defensive mechanisms into overdrive. People who then parent children after experience of this kind of trauma can then pass on the over-activation of those defenses. Several studies have shown this to be true and they have even identified some of the signals now and succeeded in blocking them in mice.
Once these changes in expression occur, it can take a few generations for them to be cleaned out of the offspring.
That level of ingraining would not just be in the culture. Survival would favor those who are more passive and genetic selection would occur. Epigenetics would play a part too when traumas endured by parents are strong enough to turn on survival mechanisms whose activation is passed on to their children.
Just being in the lobby when you start to post the review doesn't seem like much of a barrier to me unless you didn't actually go to see the movie at all. I am very fond of the saying "trust, but verify".
They could take this a step further by requiring some indication of actual attendance. This might be as simple as only accepting reviews posted by the app from the theater during or within a half an hour of a showing until the movie has been released on DVD or for streaming. Or perhaps every review could require a photo of a unique ticket stub assuming there is some way to validate them reliably.
The audience score is the only aspect of Rotten Tomatoes I find to be of value. The scores from movie critics just don't seem to have a strong relationship with whether or not I'm going to enjoy a movie. Actually, maybe that's not quite correct. It does sometimes seem like a movie with a very low critics score and a high audience score is usually great. So there may be a strong inverse relationship.
I think the chicken and egg issue will dominate though. Until a large portion of the datacenter systems are ARM, there would be no compelling reason for a developer to switch their development platform and many compelling reasons not too. And until a large number of developers are on ARM, the datacenters would be fighting the developer's platform if they switch.
Why would I buy a development system as premiumly-priced as the Macs to target a platform that might be successful in a few years? These things don't happen overnight.
I CAN play devil's advocate with myself here. I do realize that this will help front end developers, and that is almost certainly in Apple's thoughts. But, in my experience, the back end is where the real tech is. If front end developers jump to ARM-based Macs because it makes their jobs easier, we'll see the already damaging gulf between front and back end development widen. That would be a bad result for the industry. Of course, I say this as more of a full-stack guy that believes it is much harder to develop a quality product when nobody on the team fully understands both worlds.
Linus Torvalds has stated that ARM won't win the server space because developers want to run their apps on the architecture it has been developed on and almost all are developing on x86. Many application bugs are still architecture specific. Application performance optimization is also highly architecture specific, especially for database applications.
Given the Mac's popularity among developers, this argument should apply to the Macs too when looked at from the opposite angle. The vast majority of servers are x86, and developers want to run their apps on the architecture they are developing for. Running in an emulator is nowhere near the same experience. I would think a switch from x86 to ARM would decimate the number of developers calling the Mac home.
Separately, I don't see the appeal of running phone apps on my laptop or desktop. Smartphone apps do not have the feature density that I'm looking for with a desktop app and desktop apps are not generally appropriate for smartphones. On my desktop, I don't want simplicity. I want to see everything I can at once and to be able to do almost everything with my keyboard.
True. That would cause the battery pack to have a bit less range than a fresh one. However, this works out a bit like buying prewashed versus fresh jeans. The prewashed will fit just as nice after a few washes but the ones that were not prewashed will be too tight.
The loss of capacity in a fresh pack is much faster than in one that has already made it down below 90% or so. The curve flattens out a lot. A 100 kWh battery that is down to 80 kWh due to having moved down to the 80% point in its curve will stay near that 80 kWh level for far longer than one that is starting out at 80 kWh. After some time, the true 80 kWh will be a 70 kWh battery and the 100 kWh battery may have just fallen from 80 kWh to 78 kWh.
A 100 kWh battery that is has reached 80% capacity due to having a few hundred thousand miles of use would be an awesome used purchase for the 80 kWh version of the same vehicle. When it drops to 80% of what that vehicle expects, send it on to a utility storage operation that uses acres of space and doesn't care as much about power density.
Using this approach, vehicle batteries might not have 9-lives, but they should have three or four.
There is already a market for used battery packs. As the volume of used packs available reaches thousands, entrepreneurs will come along and realize they could build racks and control systems ready made to accept Tesla (for example) battery packs, purchase 1,000 used 100 kWh packs, and deploy an 80 MWh utility backup system for a fraction of the cost of one using brand new batteries. This will push the recycling problem way back.
I had the same thought. It seems that many businesses are no longer doing their best to put their competition out of business in order to increase profits. Instead, they concentrate on ways to increase profits through internal changes without changes in sales. Many also no longer push to innovate. R&D cost money. I guess it is much easier to effect the stock on 3-month timeframes when you just silently agree to a truce with your competition and work on reducing internal costs and using accounting tricks to increase profit.
The thought that net neutrality will be an issue for the 2020 presidential election instead of being put to bed this year is very discouraging. The speed increase I get when using a VPN connection with streaming services is steadily increasing.
On a related note, I'm having frequent connection issues, up to a few times a week now for around an hour at a time, using multiple major public DNS systems including Cloudflare, Google, and OpenDNS. Interestingly, I can get to them through the VPN when I can't get to them directly. I believe my provider is interfering with competitive DNS systems that are taking away information that my provider sells.
Pretty much what I came to say. It would fit with the Adobe culture to have their product rely on a custom instruction set on a device in a USB dongle to execute using the excuse that it is a critical "accelerator".
Do you mean "books" or the more inclusive experience of "books + chair + lamp + reading glasses"?
I've probably been using gmail as long as you. Might have even got the invitation from the same place.
I'm fairly positive that all spam goes to the spam folder. I can't remember the last time I had to mark a spam message. I do look at the spam folder periodically for its entertainment value and to keep up with the flavor of the current scams. I haven't had a false positive in many years.