When I was younger I used to think that as well. But reality and watching many companies I've worked at go through rounds of layoffs just doesn't bear that out.
Although it often doesn't work that way, the problem is that people perceive that it works that way, no matter how many times any manager tells them that it doesn't. Thus, when workers see other workers working long hours, they will usually feel like they are expected to do likewise. And in the grand scheme of things, there is no difference between actually being expected to work long hours and misperceiving that they are expected to do so.
Worse, in some fields, there is a significant penalty caused by additional communication as the number of people increases. Thus, in the short term, it is also significantly more efficient to work two people 60 hours per week than three people at 40 hours per week. That works up until they start burning out, their health starts to suffer, they miss a play/soccer game/piano recital/wedding anniversary and the resulting problems at home start to bleed into their work performance, or any number of other problems arising out of being massively overworked.
You say they want to "destabilize" the system? Well so do I. That's why I voted for Trump [youtube.com], because the system fucking sucks.
Yes, the system sucks. But throwing Trump into the system is like saying "San Francisco has a problem with homeless poop on the streets. Let's drop a nuke on it."
There's are a few right ways and a great many wrong ways to cause real political change. Trump was quite obviously the latter. Instead of merely upsetting the status quo and getting new, less corrupt blood into power, the right wing is clinging to his coattails and agreeing with whatever borderline insane ideas he comes up with, and from all indications, most people on both sides of the aisle are likely to retain their power, which means nothing really changed meaningfully, except that the rudder is now being controlled by a random number generator.
Worse, some of the things he is doing are so utterly bats**t crazy that he is making it harder for candidates who aren't part of the current legislative elite to win elections; too many voters are afraid that they'll become another Trump. The entrenched powers are becoming even harder to unseat, thanks to Trump. This isn't how to create change; rather it is how to create chaos that cements the power of those who already have it.
Now if you were to make a statement that alienates a good chunk of the population. The government can't persecute you for being an asshole towards other people if you don't break any particular law. But other citizens may exercise their right to free speech calling for a boycott of your business. This may harm your business so bad that you'd have to shut it down.
This kind of misses the point, though. The right is still broader than just preventing the government itself from restraining your speech. It also prevents the government from acting in ways that would support others restraining your speech. For example, libel is illegal, but someone suing for libel generally cannot use the courts to obtain prior restraint on that speech. Once the speech has happened, though, the consequences of that speech are largely a separate issue, e.g. the person you libeled can sue you.
No, everyone should have flexibility in when they work. But that's not the same thing as flexibility in how long they work.
If you allow workers to work longer hours, three things tend to happen:
First, they appear to be working harder, so those folks get promoted while the people whose outside lives (family commitments, for example) don't give them the flexibility to work longer hours slowly see their salaries slip, and may eventually even get laid off. This often results in age discrimination, gender discrimination, etc.
Second, other people start to try to mimic that behavior, until you end up with a whole team that is working 60+-hour weeks. And performance begins to suffer, because it turns out there really are limits to how much intellectual work you can usefully do in a week.
Third, schedules start to slip, which causes people to think that it must be because they aren't working long enough hours. This leads to further quality loss, resulting in a vicious cycle.
It really is necessary for employers to limit their workers based on the number of hours that a typical employee can manage to function usefully. Employers that don't do so eventually pay the price for that mistake, but usually not before the clueless people who proposed longer hours disappear, held aloft by their golden parachutes while everybody else falls.
Yeah, I figured that you were typing on a phone, what with the plant reference in the first sentence.:-)
I think that won't be enough for the upgrades they will need, particularly when they realize they need to add lidar and self-cleaning capability for the cameras. Wouldn't want your Tesla to get stuck 2000 miles away because it picked up some dirt from a passing truck.
I don't think they need LIDAR, necessarily. LIDAR just reduces the complexity of the image recognition by giving you a depth map for free (computationally) and, if you have sufficient historical data, a very accurate notion of your car's position. But even if I'm wrong, it wouldn't be that expensive or difficult to install one of the newer, smaller LIDAR units (e.g. Quanergy) in lieu of Tesla's existing RADAR hardware, just in front of the front grille.
And as long as the windshield wipers can clear at least one of the three front cameras, it should be able to at least limp home at a slow speed even if the other cameras are thoroughly blocked. With two cameras, you have depth info, and it shouldn't even be limping.
I expect the charge port will be upgraded too, to help it plug in automatically. The snake thing they showed off was totally impractical.
I don't think that any charge port design would help much with that. The ports are at different heights on different models of car, and even with different tires on the same model of car. And you'll never get a car to line up the charge port to within a fraction of an inch where you could do something simple. It is far easier for the arm to just find the charge port and line itself up. The snake design is pretty bizarre, though. You could do it just as easily with a basic motorized arm or, for that matter, a pair of motorized sliding rails mounted at 90 degrees with a telescoping arm attached, and a single rotational motor on the end. *shrugs*
Make excessive overtime illegal (or enforce existing laws). If you miss a deadline the scheduling manager is at fault.
This. Crunch overtime shouldn't be optional. As soon as you allow anyone to do insane extra amounts of work, you create an environment where that becomes expected. And then, because there were no negative consequences from the poor planning, nobody learns, and the next time, it is even worse. Pretty soon, you end up in a situation where you're all-crunch, all the time.
At least the MySQL stuff has a valid reason. The old functions were hard to use in ways that didn't introduce security bugs. Non-parameterized SQL queries are one of the most common causes of security holes, largely because they're really hard to get right. I'm all in favor of killing them. With fire.
It's the other thirteen pages of changes from PHP 5 to PHP 7 that are the problem — little, subtle things where variables get populated in the opposite order depending on what version of PHP you're running, or the order of operations changes in ways that require new parentheses. These sorts of changes are hard to spot in large code bases, and could potentially result in unexplained failures. If your code doesn't have amazing unit test coverage (which PHP usually doesn't)... good luck.
They Prunella probably realised that the upgrades they need to add to the cars are going to cost more than 2k.
You Vulgaris seem to be a little confused. The hardware upgrade might cost more than $2k, but the $2k was the discount over the "add it later" price, which is $5k. So the cost is $3,000, all of which can ostensibly be used to pay for the replacement AP hardware, because that's the only difference. I would be surprised if the AP computer were really that expensive.
The FSD package was offered at a discount because it was a long way from being done, and it was used to help fund the development of the feature. Now that the first FSD-only features are about to start rolling out, albeit in a limited form, it arguably doesn't make sense to offer it at a discount anymore.
Also, this will likely significantly reduce the number of Model 3 owners trying to get upgrades to their AP hardware during the first few months after they start upgrading them early next year, thus reducing the chaos a bit.
That's not entirely true. The President nominates someone to head the Federal Reserve, which is a position that wields a lot of power over the economy. Fortunately, Trump seems to have accidentally nominated somebody who he doesn't really agree with, so we're okay there.:-D
The general consensus is that the economy takes a long time to change direction, and as such, any economic motion during the first year or two of any presidency is primarily due to momentum.
Or, to put it another way, we blamed the weak economy during Obama's first year or two on Bush. Why should Trump be held to a different standard just because the economy is going in a positive direction?
I am, of course, referring to the areas that have recently thawed, which are presumably the areas of actual interest, as opposed to areas that have always been accessible by boat, which one would assume have been mapped for a long time.
Well that's a bold statement considering we know nothing about the sea floor there today, what underwater volcanos might be doing, geologic movements, etc.
If there were active underwater volcanos, I'd expect holes in the ice above it (what with heat rising and all). I'm not saying there's no geological activity, but I'd expect it to move at a glacial pace, so to speak.
IMO, the cheese grater really is too big (3044 ci). It was designed in an era when the G5 required insane amounts of cooling, and it got tweaked only slightly during the Intel transition. You could easily build something with four drives and a high-end Xeon CPU in an enclosure half that size, and possibly a fourth that size.
The two main problems with the current design are that A. Apple decided to remove SATA entirely, and B. they built the product in an enclosure that was barely a tenth that size (338 ci). As a result, with the exception of people who use them as software build machines, pretty much everybody who owns one has this unstackable round thing sitting beside a rectangular Thunderbolt RAID, because the Mac Pro is limited to a paltry 1 TB of storage — less than my laptop.
So Apple does need to build something bigger and more pro-capable (six drive bays would be ideal), but a horizontal design with drives above the logic board would be much better (smaller). I'd settle for a vertical design with drives beside the logic board. Either way.
And a good design should have removable handles that can be replaced with rack ears so that you can rack mount it in 2U or 3U of rack space, rather than the 4U used by the trashcan or the 5U for the cheese grater (after grinding off the handles).
Only Apple could be so arrogant as to remove the only low-latency audio output from their iPad, and then announce that neutered product at an academy of music.
Not to mention that most ships are following trade routes.
Yes, because that is where you need the most accuracy anyway, and they would be mapped first. That is the most win/win scenario I've seen for.a while.
And that's *all* that will be mapped. Once you've done this with a few ships, you've mapped those areas. It is uninteresting to keep mapping the same routes over and over. And yes, the routes change somewhat, but they're still going to the same endpoints, which kind of limits how much they can change.
I think that a better approach would be to have commercial ships toss out solar-powered, motorized buoy-drones with mapping hardware at random points along their trip, and let them float/boat around in a spiral pattern or something, moving further and further from the shipping lanes.
It's the plague of locusts. They've been waiting in the ice for countless millions of years — waiting for the right moment to wipe out all our food crops and begin the nuclear resource war to end all wars.
No, I was approaching it from the perspective of a small business that didn't want to back up to the cloud. Most businesses don't need data to be available after a decade. The probability of something still having any value after it gets deleted rapidly approaches zero after a couple of months, and if it existed within the last couple of months, it should be in at least a couple of full backups and possibly incremental backups.
More to the point, $150 will buy you a basic 8 TB external hard drive these days. So in the worst-case compression scenario, tape is only about 2/3rds the price of storing the data on hard drives, assuming you have sufficient physical space to store the hard drives, and ignoring the cost of the tape drive. (More to the point, if compression can let you store more data on tape, that same compression would also let you store more on a hard drive, so for a fair comparison, only the native capacity matters.)
As soon as you add in the cost of the tape drive, you can compute a break-even point where tape becomes cheaper. The break-even point is where the cost of hard drives, x, equals the cost of tapes, which is the cost of the tape drive ($4k) plus 2/3 * x. So x = $4000 +.66666x. Solve for x, and you get x/3 = $4000, so x = $12000.
At $150 for 8 TB, that comes to 640 TB. Realistically, that ought to be enough for anybody.:-)
But I guess if you have to store more than that, tape might be a good deal.
Also, LTO-8 has only 12 TB of native capacity. Ostensibly, you can get up to 30 TB of storage per tape, but that's the best-case scenario. Realistically, you need to assume that you'll get 12 TB per tape, and if you get more, yay.
That's why every freeway wreck is bigger in Texas.
But seriously, Section 545.061 is usually interpreted to mean that on any road with four or more lanes, the existing traffic has the right of way, not the merging traffic. In fact, even vehicles in the left lane have more right to use the right lane than someone merging in from an exit lane — forget somebody in the right lane.
The lack of a physical keyboard makes many types of content creation much less efficient (typing, video editing, etc.).
Which is irrelevant if the content you are creating isn't related to typing.
Actually, it isn't. That's the point I was trying to make when I mentioned video editing. A LOT of creative tools make use of the keyboard. In Photoshop, for example, there are very different behaviors in the selection tool, depending on whether I'm holding down shift (add to selection), command (remove from selection), etc. In video editing tools, you hold down keys to rapidly change tools or change the behavior of tools, and you actually use a large percentage of the keyboard regularly, even for fairly simple video editors.
Yes, it is possible to do those things without the keyboard, but it is a LOT clumsier. It turns out that having an array of a hundred physical buttons lets you get away with an order of magnitude fewer clicks or taps.
I/O is limited (a single Lightning port for charging and USB-2.0-speed I/O). This will go away when Apple adopts USB-C ports on their iDevices.
This is a good approximation of a non-issue. You seem to be forgetting about WiFi, Bluetooth, NFC, etc. There are innumerable tasks that aren't particular I/O intensive that tablets can handle just fine. It's not like one needs gigabit ethernet to write notes or take pictures etc. I agree that Apple should move to USB-C but I/O speed isn't the primary reason for that.
I *wish* I could forget about those things. Wi-Fi is only really viable in the presence of an infrastructure network, because in both IBSS and peer-to-peer modes, you're limited to 802.11g speeds. Bluetooth is the single most unreliable technology I have ever used, and I used to use Netscape 2.x on the Mac. NFC doesn't do anything useful at all for you I/O wise other than make Wi-Fi or Bluetooth pairing with other hardware easier. And none of those do anything at all to fix the primary complaint, which is a lack of bandwidth. Try importing a few dozen 4 GB video files over 802.11g. Go ahead. I'll wait. No, actually, I won't. It would take too long.
There's a reason I use my laptop for everything and never bothered to upgrade my tablet hardware, which is now sitting unused. Even right now, I cannot do even 1% of the things I do on my laptop while using a tablet — not because the software isn't there, but because the hardware couldn't support the software even if it existed. And even if the hardware got better, iOS still wouldn't be viable as a laptop replacement because of the touch interface. When you use something all day, having to choose between a sore neck and gorilla arm just doesn't make sense. Trackpads and keyboards work much, much better.
So basically the entire market for "pro" use of a tablet is limited to:
People who travel a lot and can't or won't leave their work at home, but don't want to carry a laptop for some reason.
People who use pro apps infrequently and do not care that everything takes several times as long.
People who use very basic apps that don't do much, and thus don't need very many controls.
For everyone else, tablets just can't cut it for *most* types of content creation, and I don't foresee that changing in the next decade, given that in the last decade, there has been approximately zero progress at closing the gap between laptops and tablets with the sole exception of their CPU performance (and even then, not under sustained use).
Tablets have insufficient RAM to work with large projects without constant paging (which would significantly reduce hardware life expectancy unless very carefully managed).
A) "Large" is an ambiguous term and tablets have more than enough memory to do a huge variety of useful tasks right now. B) It is a trivial proposition to
Slashdot Mirror
Although it often doesn't work that way, the problem is that people perceive that it works that way, no matter how many times any manager tells them that it doesn't. Thus, when workers see other workers working long hours, they will usually feel like they are expected to do likewise. And in the grand scheme of things, there is no difference between actually being expected to work long hours and misperceiving that they are expected to do so.
Worse, in some fields, there is a significant penalty caused by additional communication as the number of people increases. Thus, in the short term, it is also significantly more efficient to work two people 60 hours per week than three people at 40 hours per week. That works up until they start burning out, their health starts to suffer, they miss a play/soccer game/piano recital/wedding anniversary and the resulting problems at home start to bleed into their work performance, or any number of other problems arising out of being massively overworked.
Yes, the system sucks. But throwing Trump into the system is like saying "San Francisco has a problem with homeless poop on the streets. Let's drop a nuke on it."
There's are a few right ways and a great many wrong ways to cause real political change. Trump was quite obviously the latter. Instead of merely upsetting the status quo and getting new, less corrupt blood into power, the right wing is clinging to his coattails and agreeing with whatever borderline insane ideas he comes up with, and from all indications, most people on both sides of the aisle are likely to retain their power, which means nothing really changed meaningfully, except that the rudder is now being controlled by a random number generator.
Worse, some of the things he is doing are so utterly bats**t crazy that he is making it harder for candidates who aren't part of the current legislative elite to win elections; too many voters are afraid that they'll become another Trump. The entrenched powers are becoming even harder to unseat, thanks to Trump. This isn't how to create change; rather it is how to create chaos that cements the power of those who already have it.
This kind of misses the point, though. The right is still broader than just preventing the government itself from restraining your speech. It also prevents the government from acting in ways that would support others restraining your speech. For example, libel is illegal, but someone suing for libel generally cannot use the courts to obtain prior restraint on that speech. Once the speech has happened, though, the consequences of that speech are largely a separate issue, e.g. the person you libeled can sue you.
No, everyone should have flexibility in when they work. But that's not the same thing as flexibility in how long they work.
If you allow workers to work longer hours, three things tend to happen:
It really is necessary for employers to limit their workers based on the number of hours that a typical employee can manage to function usefully. Employers that don't do so eventually pay the price for that mistake, but usually not before the clueless people who proposed longer hours disappear, held aloft by their golden parachutes while everybody else falls.
Yeah, I figured that you were typing on a phone, what with the plant reference in the first sentence. :-)
I don't think they need LIDAR, necessarily. LIDAR just reduces the complexity of the image recognition by giving you a depth map for free (computationally) and, if you have sufficient historical data, a very accurate notion of your car's position. But even if I'm wrong, it wouldn't be that expensive or difficult to install one of the newer, smaller LIDAR units (e.g. Quanergy) in lieu of Tesla's existing RADAR hardware, just in front of the front grille.
And as long as the windshield wipers can clear at least one of the three front cameras, it should be able to at least limp home at a slow speed even if the other cameras are thoroughly blocked. With two cameras, you have depth info, and it shouldn't even be limping.
I don't think that any charge port design would help much with that. The ports are at different heights on different models of car, and even with different tires on the same model of car. And you'll never get a car to line up the charge port to within a fraction of an inch where you could do something simple. It is far easier for the arm to just find the charge port and line itself up. The snake design is pretty bizarre, though. You could do it just as easily with a basic motorized arm or, for that matter, a pair of motorized sliding rails mounted at 90 degrees with a telescoping arm attached, and a single rotational motor on the end. *shrugs*
This. Crunch overtime shouldn't be optional. As soon as you allow anyone to do insane extra amounts of work, you create an environment where that becomes expected. And then, because there were no negative consequences from the poor planning, nobody learns, and the next time, it is even worse. Pretty soon, you end up in a situation where you're all-crunch, all the time.
At least the MySQL stuff has a valid reason. The old functions were hard to use in ways that didn't introduce security bugs. Non-parameterized SQL queries are one of the most common causes of security holes, largely because they're really hard to get right. I'm all in favor of killing them. With fire.
It's the other thirteen pages of changes from PHP 5 to PHP 7 that are the problem — little, subtle things where variables get populated in the opposite order depending on what version of PHP you're running, or the order of operations changes in ways that require new parentheses. These sorts of changes are hard to spot in large code bases, and could potentially result in unexplained failures. If your code doesn't have amazing unit test coverage (which PHP usually doesn't)... good luck.
You Vulgaris seem to be a little confused. The hardware upgrade might cost more than $2k, but the $2k was the discount over the "add it later" price, which is $5k. So the cost is $3,000, all of which can ostensibly be used to pay for the replacement AP hardware, because that's the only difference. I would be surprised if the AP computer were really that expensive.
The FSD package was offered at a discount because it was a long way from being done, and it was used to help fund the development of the feature. Now that the first FSD-only features are about to start rolling out, albeit in a limited form, it arguably doesn't make sense to offer it at a discount anymore.
Also, this will likely significantly reduce the number of Model 3 owners trying to get upgrades to their AP hardware during the first few months after they start upgrading them early next year, thus reducing the chaos a bit.
That's not entirely true. The President nominates someone to head the Federal Reserve, which is a position that wields a lot of power over the economy. Fortunately, Trump seems to have accidentally nominated somebody who he doesn't really agree with, so we're okay there. :-D
The general consensus is that the economy takes a long time to change direction, and as such, any economic motion during the first year or two of any presidency is primarily due to momentum.
Or, to put it another way, we blamed the weak economy during Obama's first year or two on Bush. Why should Trump be held to a different standard just because the economy is going in a positive direction?
Eh. They just measured a stack of a hundred of them. It happens to everybody.
I am, of course, referring to the areas that have recently thawed, which are presumably the areas of actual interest, as opposed to areas that have always been accessible by boat, which one would assume have been mapped for a long time.
If there were active underwater volcanos, I'd expect holes in the ice above it (what with heat rising and all). I'm not saying there's no geological activity, but I'd expect it to move at a glacial pace, so to speak.
IMO, the cheese grater really is too big (3044 ci). It was designed in an era when the G5 required insane amounts of cooling, and it got tweaked only slightly during the Intel transition. You could easily build something with four drives and a high-end Xeon CPU in an enclosure half that size, and possibly a fourth that size.
The two main problems with the current design are that A. Apple decided to remove SATA entirely, and B. they built the product in an enclosure that was barely a tenth that size (338 ci). As a result, with the exception of people who use them as software build machines, pretty much everybody who owns one has this unstackable round thing sitting beside a rectangular Thunderbolt RAID, because the Mac Pro is limited to a paltry 1 TB of storage — less than my laptop.
So Apple does need to build something bigger and more pro-capable (six drive bays would be ideal), but a horizontal design with drives above the logic board would be much better (smaller). I'd settle for a vertical design with drives beside the logic board. Either way.
And a good design should have removable handles that can be replaced with rack ears so that you can rack mount it in 2U or 3U of rack space, rather than the 4U used by the trashcan or the 5U for the cheese grater (after grinding off the handles).
Only Apple could be so arrogant as to remove the only low-latency audio output from their iPad, and then announce that neutered product at an academy of music.
Ignorance of your audience FTW!
I don't see any part of this machine that looks 30 inches long. What, and how, were they measuring?
They missed a decimal point. It is 0.30" thick.
Not to mention that most ships are following trade routes.
Yes, because that is where you need the most accuracy anyway, and they would be mapped first. That is the most win/win scenario I've seen for.a while.
And that's *all* that will be mapped. Once you've done this with a few ships, you've mapped those areas. It is uninteresting to keep mapping the same routes over and over. And yes, the routes change somewhat, but they're still going to the same endpoints, which kind of limits how much they can change.
I think that a better approach would be to have commercial ships toss out solar-powered, motorized buoy-drones with mapping hardware at random points along their trip, and let them float/boat around in a spiral pattern or something, moving further and further from the shipping lanes.
It's the plague of locusts. They've been waiting in the ice for countless millions of years — waiting for the right moment to wipe out all our food crops and begin the nuclear resource war to end all wars.
Hear that whooshing sound?
No, I was approaching it from the perspective of a small business that didn't want to back up to the cloud. Most businesses don't need data to be available after a decade. The probability of something still having any value after it gets deleted rapidly approaches zero after a couple of months, and if it existed within the last couple of months, it should be in at least a couple of full backups and possibly incremental backups.
More to the point, $150 will buy you a basic 8 TB external hard drive these days. So in the worst-case compression scenario, tape is only about 2/3rds the price of storing the data on hard drives, assuming you have sufficient physical space to store the hard drives, and ignoring the cost of the tape drive. (More to the point, if compression can let you store more data on tape, that same compression would also let you store more on a hard drive, so for a fair comparison, only the native capacity matters.)
As soon as you add in the cost of the tape drive, you can compute a break-even point where tape becomes cheaper. The break-even point is where the cost of hard drives, x, equals the cost of tapes, which is the cost of the tape drive ($4k) plus 2/3 * x. So x = $4000 + .66666x. Solve for x, and you get x/3 = $4000, so x = $12000.
At $150 for 8 TB, that comes to 640 TB. Realistically, that ought to be enough for anybody. :-)
But I guess if you have to store more than that, tape might be a good deal.
*shrugs*
Also, LTO-8 has only 12 TB of native capacity. Ostensibly, you can get up to 30 TB of storage per tape, but that's the best-case scenario. Realistically, you need to assume that you'll get 12 TB per tape, and if you get more, yay.
That's why every freeway wreck is bigger in Texas.
But seriously, Section 545.061 is usually interpreted to mean that on any road with four or more lanes, the existing traffic has the right of way, not the merging traffic. In fact, even vehicles in the left lane have more right to use the right lane than someone merging in from an exit lane — forget somebody in the right lane.
The lack of a physical keyboard makes many types of content creation much less efficient (typing, video editing, etc.).
Which is irrelevant if the content you are creating isn't related to typing.
Actually, it isn't. That's the point I was trying to make when I mentioned video editing. A LOT of creative tools make use of the keyboard. In Photoshop, for example, there are very different behaviors in the selection tool, depending on whether I'm holding down shift (add to selection), command (remove from selection), etc. In video editing tools, you hold down keys to rapidly change tools or change the behavior of tools, and you actually use a large percentage of the keyboard regularly, even for fairly simple video editors.
Yes, it is possible to do those things without the keyboard, but it is a LOT clumsier. It turns out that having an array of a hundred physical buttons lets you get away with an order of magnitude fewer clicks or taps.
I/O is limited (a single Lightning port for charging and USB-2.0-speed I/O). This will go away when Apple adopts USB-C ports on their iDevices.
This is a good approximation of a non-issue. You seem to be forgetting about WiFi, Bluetooth, NFC, etc. There are innumerable tasks that aren't particular I/O intensive that tablets can handle just fine. It's not like one needs gigabit ethernet to write notes or take pictures etc. I agree that Apple should move to USB-C but I/O speed isn't the primary reason for that.
I *wish* I could forget about those things. Wi-Fi is only really viable in the presence of an infrastructure network, because in both IBSS and peer-to-peer modes, you're limited to 802.11g speeds. Bluetooth is the single most unreliable technology I have ever used, and I used to use Netscape 2.x on the Mac. NFC doesn't do anything useful at all for you I/O wise other than make Wi-Fi or Bluetooth pairing with other hardware easier. And none of those do anything at all to fix the primary complaint, which is a lack of bandwidth. Try importing a few dozen 4 GB video files over 802.11g. Go ahead. I'll wait. No, actually, I won't. It would take too long.
There's a reason I use my laptop for everything and never bothered to upgrade my tablet hardware, which is now sitting unused. Even right now, I cannot do even 1% of the things I do on my laptop while using a tablet — not because the software isn't there, but because the hardware couldn't support the software even if it existed. And even if the hardware got better, iOS still wouldn't be viable as a laptop replacement because of the touch interface. When you use something all day, having to choose between a sore neck and gorilla arm just doesn't make sense. Trackpads and keyboards work much, much better.
So basically the entire market for "pro" use of a tablet is limited to:
For everyone else, tablets just can't cut it for *most* types of content creation, and I don't foresee that changing in the next decade, given that in the last decade, there has been approximately zero progress at closing the gap between laptops and tablets with the sole exception of their CPU performance (and even then, not under sustained use).
Tablets have insufficient RAM to work with large projects without constant paging (which would significantly reduce hardware life expectancy unless very carefully managed).
A) "Large" is an ambiguous term and tablets have more than enough memory to do a huge variety of useful tasks right now. B) It is a trivial proposition to