The sale of small parts declined when electronics moved to robotic assembly and surface mount parts. Modern circuit boards are too densely packed to be easily repaired; for the most part even the companies don't try, they just throw them away. (Expensive boards like computer motherboards do get repaired, but they ship them to China to do it so the work can be done with inexpensive labor.) And it takes specialized tools and quality magnifiers to be able to work on them at all, so the cost of entry is non-trivial. The amount of time that a repair would take means that a small independent shop really can't make a living at fixing boards; they can swap main assemblies for new ones, fix connections between boards, and replace a few large parts like connectors, and that's about it.
The same trends pretty much killed kits. To put together a non-trivial kit using surface mount parts you have to do a lot of work to package it appropriately, because some of the parts have no markings at all, so they cannot be identified without testing once you have removed them from their spool. (Ceramic capacitors are a prime offender.) And after you have gone to all that trouble, your market is limited because most electronic hobbyists aren't interested in building a surface mount kit. It is becoming increasingly difficult to design a kit that uses old fashioned through-hole parts; many components now cost far more in that form, are no longer available, or were never made in through-hole versions at all. Small kit companies still exist to sell to electronic hobbyists and amateur radio operators, but it's because those people like to build things and will buy kits even if they don't save any money.
Fighting against those trends meant that Radio Shack's business of selling components would have declined no matter what they did. But they worsened the problem by their own actions. They hired the wrong people for that business, gave them the wrong training, and offered the wrong pay structure. This is all assuming that the company actually wanted to sell any parts rather than merely keep them around as a vestige of their old days.
The pay structure matters because the salaries of the people on the Radio Shack were poor; the only way to make any money at the job was on sales commissions, and those were only significant for the large ticket items. (In the last few years that has mostly meant phones; farther back it would have also included things like CB radios and stereo components.) Sales people don't want to spend any time helping you find parts because there is nothing in it for them, and it takes them away from possibly serving a customer who might buy something that they will earn some commissions on. Getting a reputation in the store for being knowledgeable about the parts could be counterproductive; the other salespeople would then foist all those customers onto you while they served the more lucrative phone buyers.
If Radio Shack had wanted to have people on the floor who actually had a clue about electronics and electronic parts, they would have had to pay them more. And it probably would have helped to take the further step of eliminating commissions and simply giving people a good salary. Commissions have lots of problems, including the perverse incentives that I mentioned above, and the fact that they make the salespeople enemies of each other rather than allies. But I think the core problem is that they make the self interest of the salesperson be directly opposed to the self interest of the customer. The sales person wants to just sell you as much crap as possible as quickly as possible because that maximizes their commission. Given the lousy pay, they're probably not planning to stay in the job long enough to care about repeat business.
Putting the parts in drawers didn't help. It made the parts business less visible. It made it difficult for hobbyists to browse the parts and perhaps find something they wanted it to buy, making it something that was only useful if you were looking for a specific thing.
Capacitor values still matter in analog circuits. If you're building audio or radio gear, you'll need some that aren't bypass or power filtering values. Digital circuits now rarely contain analog timing circuits or filters (earlier ones did have delay lines) so you're rarely going to need capacitors in values that aren't powers of 10.
So they're doing a refresh to bring their systems a bit more up to date. The real world benefits are small; they'll be about 10% faster than the models they replace and will have slightly longer battery life. Streaming fans who hook up their laptops to an external display will get one more benefit: the new systems will presumably support 4K streaming because they'll have the required Kaby Lake processors.
Updating the Air is a bit more of a step forward since it's currently still using a 5th generation processor; it might get a 20% speed improvement. Apple probably was thinking that it would go end of life by now, but they still have a lot of presence in the higher education market and the Air is their main product there. The ultrathin MacBook is a bit too limited, and I don't think Apple is prepared to cut the price to the under-$1000 level that they need to reach. The MacBook Pro is too expensive; the CS and media studies majors will buy it because they need the extra power but most students will not.
The question if they do update the Air: will they just shove in a newer CPU and leave the rest of the design as it is, or will they decide that the product has a longer term place in their product lineup and give it a more substantial refresh? The relatively low display resolution stands out as the #1 thing they could improve, though that would bring up the question of cannibalizing their more expensive products. It's also likely that a new version would move to USB-C for charging and connectivity - but would they go all-in as they have on their other products, or offer a mix of USB-C and legacy USB ports as most Windows laptops have?
The ultrathin MacBook might see some design changes as well, though not as major. The most popular thing Apple could do with that system would be to shoehorn in at least one more USB port; that would allow the user to connect something and charge the system at the same time without the need for yet another dongle. The MacBook Pro design is less than a year old, so I don't expect any changes there other than the CPU and perhaps some tuning of the available configurations (amounts of RAM and flash memory, etc).
The author isn't saying that electric cars will replace internal combustion cars as something that most people own. His actual claim is that private car ownership will mostly go away, with autonomous electric cars being the technology that makes that possible. A very different premise, and one that deserves analysis.
First up is the question of whether autonomous car technology will be ready quickly enough for his time frame. To meet the 2024 time point, autonomous cars have to be ready for mass production by 2020 at the latest and receive regulatory approval to be sold and driven autonomously. I think that's an overly optimistic date, especially for the legal part of the equation; that technology will come but I think it will take a bit longer than he believes
It will take quite a few years to replace the entire installed base of vehicles. Currently there are about 250 million small vehicles (cars, SUVs, pickup trucks) registered in the US and 17.5 million sold each year. At those rates it would take about 12 years to replace them all. Moving toward shared autonomous cars will reduce the total number needed, but it's still likely to take more than four years to make enough of them.
Another big question: will people be willing to forgo car ownership and do most of their travelling by using services such as Uber and/or public transit? (Uber itself won't necessarily be one of the important players or even survive, but if not somebody else will rise up to replace them.) In cities I think the case is strong; the economics of car ownership are unappealing, driving and parking are challenging and costly, and a ride in an autonomous vehicle has the potential to be much less expensive than taxis or the existing Uber because of eliminating labor costs for the drivers.
But outside the city things get murkier. Maintaining a viable car sharing service requires a high density of use, which means dense population. As you leave the city and go into the suburbs that density disappears, and the viability of a shared infrastructure of autonomous cars drops. A further problem is the fact that the places where people live and the places where they work are well separated, so the pattern of use doesn't suit the model; people will want to take their cars to their workplace, where they will sit unused or underused for most of the day until the workers are ready to return home. Meanwhile, there are few cars left in the bedroom communities for the people who are still there and want rides.
Eventually you reach rural areas, and I don't see the shared vehicle model taking hold there any time soon. There simply aren't enough people living there, and the distances people routinely travel are a barrier for electric cars.
A final question is whether people are ready for the psychological shift of giving up car ownership. I suspect that for many people, the answer is no. They see their cars as personal space, much like their homes are, and won't want to give that up. The extent to which that is true is also likely to correlate with where they live. Many people who live in the city are already using public transit and are comfortable with travelling in vehicles that they do now own or control, but people who have moved out of cities do not. I have talked to people who find the idea of using public transit horrifying for that reason, and they also choose to drive many hundreds of miles on long trips rather than using a bus or train because of their preference for private travel. (Flying has become its own horror show and I can understand people wanting to avoid that.)
There is also likely to be some resistance to autonomous vehicles themselves, aside from the question of ownership. Some people who value independence and private travel will not be prepared to let a machine drive for them because it involves giving up control; the fact that the autonomous car is likely to be safer won't sway them. Eventually some jurisdictions may move toward banning human driving for safety reasons and that will set off a firestorm of controversy.
Those calculators have no network connectivity so the amount of data that the calculator company gets is limited. If you buy one at a campus bookstore it's possible that the store passes along some demographic info.
And you can't use them during tests because you might use the smartphone to cheat. You're required to use one of a few models of dedicated calculator that the test givers know do not have any programmable features. Sometimes the list is limited to TI; other schools have slightly more inclusive lists that include HP and Casio models.
No, not lying. The current version of the TI-84 is just under $110 on Amazon: https://www.amazon.com/Texas-I... If your kid buys one in a brick and mortar store it's probably a bit more expensive.
It's not the number of copies. It's that the OEM license only authorizes a bare iron install. The only Microsoft licenses that authorize virtual machine installations are full retail copies and certain forms of enterprise licensing, plus the exception of XP Mode in Windows 7 Professional which gave you a license for one copy of XP in a virtual machine on the system that was running W7. Upgrade licenses can be used for a virtual machine install IF they are used as an upgrade of a full retail license; they retain the same rights that the upgraded copy had.
Even then, the licensing terms are unreasonably restrictive in that they only authorize ONE virtual machine. Not one running at a time, one period. (The licensing terms for virtual machines under enterprise licensing are more reasonable and do allow multiple VMs.) The net effect is to eliminate nearly all reasonable uses of virtual machines for home and small office users. And if you want both Windows and Linux on your system simultaneously you're forced to run a Windows host and a Linux VM, because doing it the other way around will cost you a bunch of additional money.
I will grant that Microsoft has legitimate business reasons for some of the restrictions. A case they're trying to prevent is somebody running a hosting company or a remote terminal server on the cheap. If they allowed you to run unlimited VMs, somebody could buy a big honking piece of server-grade hardware with hundreds of gigabytes of RAM and run dozens of clients in virtual machines while only using one Windows license. I think a more reasonable restriction would be to stipulate that you can run as many VMs as you like with one license, so long as they are all used by the same human being at any given time. They could not be used to provide a UI to somebody else, nor to offer services to other people or computers.
One problem is that it would also put an effective kill date on the computers themselves. Sometimes upgrading isn't an option: the computer itself or something that is installed in it or connected to it is not supported by the newer OS.
Upgrading computers from versions of Windows earlier than Windows 7 to a current version also costs money - and a lot more of it than the $20 that Apple charges for upgrades from some old versions of macOS. Upgrading from 7 or 8 to 10 also costs money now but there are ways around it.
Microsoft doesn't care about the XP systems. The reason they felt the need to push an XP update this time is because this piece of malware propagates peer to peer, and thus infected XP systems threaten the systems that Microsoft DOES care about.
And before anybody gets on my case about a similar disagreement: I am talking about the word hierarchies, which is a singular thing, not the hierarchies themselves which are plural. Thus "which is plural" is correct.
Comments 6a and 6b are incorrect. The second part of that sentence is referencing hierarchies, which is plural. What is being referenced is unclear from structure but clear from context; nobody except the most extreme anarchist believes that society could possibly be going away, but some people would advocate the elimination of hierarchies.
One deterrent to doing that is that it will cost you extra. The Windows license that comes with your computer licenses one copy on that computer. A virtual Windows machine ON THE SAME COMPUTER does not qualify, so you would have to buy another license for that.
It was actually Scott McNealy, co-founder of Sun Microsystems, who said that. More accurately, the quote was "You have zero privacy anyway. Get over it." If Eric Schmidt ever said something similar he would have been referencing McNealy.
Chrome on x86 is not yet 64 bit only, and won't be for some time - there are still too many machines running the 32 bit version of Windows. What changed recently is that they automatically install the 64 bit version on compatible systems, rather than people specifically having to download it, and have also started to offer to upgrade installations of 32-bit Chrome to the 64 bit version. It probably won't be long until they REQUIRE the upgrade to 64 bit on 64 bit Windows and only allow you to run the 32 bit version on 32 bit Windows. Google will eventually drop support of 32 bit Windows but it's probably a couple of years away.
Chrome on macOS and Linux is 64 bit only. That does mean that the small number of people still running 32 bit versions of those OSes can't run current versions of Chrome. On the Mac that only applies to ancient versions of macOS that are no longer supported anyway. But many current Linux distributions still offer 32 bit versions because lots of Linux users run legacy hardware; part of the appeal of Linux is that you can run current versions on hardware that is no longer supported by other OSes. Chrome won't work on those, though its open source relative, Chromium, can still be built for 32 bit Linux for now.
The Micro USB connector has some of the same problems, because it may have add-on capabilities like USB On-The-Go and MHL. But the USB-C connector is far worse. On the plus side, the connector is more physically robust than the fragile Micro USB connector.
That's actually part of the problem. There are a bunch of ports that look exactly like USB-C ports but they don't all do the same thing. You see something on your computer or phone that looks like a USB-C port but it could be any number of things:
1. A bog standard USB-C port that implements Superspeed Gen 2 USB (10Gbps)
2. A USB 3.0 standard port (5Gbps)
3. A USB 2.0 standard port (480Mbps) - these are mostly on phones
4. A Thunderbolt 3 port that implements some unknown number of PCIe lanes
5. A USB-C port that implements HDMI Alternate Mode (USB-C to HDMI with no dongle)
6. A USB-C port that implements DisplayPort Alternate Mode (USB-C to DP with no dongle)
7. A USB-C port that will charge the device
What's the problem? You can't tell which set of capabilities a USB-C port has by looking at it. There are no standard device markings or color coding to tell the various kinds of ports apart. The only thing that will help is a spec sheet for the device. A computer with more than one USB-C port may not have the same capabilities on all of them, so whether a device will work depends on which port you plug it into.
In some cases, Apple explicitly bakes in a home team advantage. The notable example is web browsers, where you are required to use the HTML rendering and Javascript engines from iOS rather than including your own. Those things are essentially the same ones that were in previous versions of Safari, so you are guaranteed that your alternative browser will offer worse performance than Apple's browser. You can attempt to compete with improvements in features and user interface, but it's a hard sell when your core performance is hobbled by Apple's restrictions.
FreeBSD has never gotten the same level of desktop adoption as GNU/Linux. So the GUIs are less refined and there are fewer desktop applications available in the repositories. For most people, a suitable distribution of GNU/Linux is a better choice on a desktop client.
However, FreeBSD is a great server OS. It has a very high performance network stack and a more fully developed and stable version of the ZFS file system. I have used it successfully on servers, both in its basic form and in FreeNAS, a distribution for storage servers that is based on FreeBSD. If you're choosing your server OS for function rather than ideology, FreeBSD in the server room is worth a look.
I would not describe FreeBSD as "more free" than GPL software. Rather, it protects a different set of rights. It's impossible to protect ALL types of freedom; some are in conflict. Stallman believes that the rights that the GPL protects are the more important ones.
Two rights that the GPL protects that the BSD license does not: a developer's ability to use and adapt all code that is based on their code, and the right of end users to use all code that is based on a given code base.
On the other hand, the BSD license allows code to be reused in non-free software, something that the GPL prohibits. Stallman does not consider that important because it means that access to that derivative code can be denied, which is a right that he considers more important. But some people consider that right to proprietary adaptation to be an important one.
Conflicting freedoms and rights are frequently seen. One simple expression of it: the principle that "your right to freely swing your arms about ends at my face". My right to not be hit by your hands is a more important one than your right to put your arms wherever you wish. Smoking is an issue where public opinion has changed over time; the right to breathe smoke-free air is now considered more important than the right to smoke in most situations.
The Pentium Pro was released in November 1995. Systems became readily available in 1996. It was not the success that Intel had hoped for in desktop applications because it had a serious performance issue when running 16 bit legacy code. (It was more popular on servers, which had migrated to 32 bit code by then.) Referencing both the AL register (the lower half of AX) and the AX register in the same code sequence caused a complete pipeline flush, as did loading a segment register. Both of these things were rather common in 16 bit code.
Desktop adoption of Intel's new P6 microarchitecture didn't really take off until the introduction of the Pentium II in May 1997; it included a segment register cache and improved handling of register aliasing to solve those two problems so it performed much better on old code. It was also much cheaper for Intel to manufacture because it eliminated the multi-die packaging that the Pentium Pro used to incorporate cache and instead used a pluggable CPU module that contained the processor and the cache RAM. That was succeeded by the Pentium III, which started with the same SECC modules as Pentium II but later moved to on-die RAM when it became possible to put more transistors on one chip.
After THAT we got the failure that was Netburst. But that's a story for another day.
Except for late night talk shows there will be little immediate effect. Any scripted show that will premiere during the summer is already written, if not already filmed. But some fall shows are currently in production and may not have all their scripts in hand; if the strike drags on they could be affected.
We're not at the limit of potential natural gas production. But as more extreme fracking becomes necessary to extract it, the cost goes up. The cost of wind and solar continues to come down. So we're probably approaching peak natural gas production because it won't be cost effective to produce more.
The demand for fossil fuels will gradually decline rather than falling off a cliff because there is a large installed base. All those internal combustion vehicles and power plants won't be replaced overnight. We're at the point where it is no longer cost effective to build a new coal power plant and will soon reach the point where it is not cost effective to build a new natural gas plant. We're approaching the point when it is no longer cost effective to operate an existing coal plant where the cost of construction is a sunk cost, but that tipping point for natural gas is still some years off.
Electric cars are not yet cost effective without the help of tax subsidies, and the charging infrastructure for long range travel is still lacking. Battery prices are dropping, but the price of a long range electric car like the Chevy Bolt (and upcoming Tesla Model 3, Nissan Leaf version 2, Hyundai Ioniq, etc.) will have to drop by another $10,000 to make the lifetime cost competitive with gasoline. At that point they'll still cost more than gas-powered cars but the reduced fuel and maintenance costs will make up the difference.
Demand for oil and natural gas won't fall to zero in the near future but demand will eventually drop dramatically. Even if we stop burning them, which would be a good idea for the environment, they are useful for manufacturing petrochemicals. Oil has uses as a lubricant. And we might find it useful to maintain a limited number of fossil fuel power plants as backups in case there is a long stretch when solar and wind sources underperform due to adverse weather conditions.
Slashdot Mirror
The sale of small parts declined when electronics moved to robotic assembly and surface mount parts. Modern circuit boards are too densely packed to be easily repaired; for the most part even the companies don't try, they just throw them away. (Expensive boards like computer motherboards do get repaired, but they ship them to China to do it so the work can be done with inexpensive labor.) And it takes specialized tools and quality magnifiers to be able to work on them at all, so the cost of entry is non-trivial. The amount of time that a repair would take means that a small independent shop really can't make a living at fixing boards; they can swap main assemblies for new ones, fix connections between boards, and replace a few large parts like connectors, and that's about it.
The same trends pretty much killed kits. To put together a non-trivial kit using surface mount parts you have to do a lot of work to package it appropriately, because some of the parts have no markings at all, so they cannot be identified without testing once you have removed them from their spool. (Ceramic capacitors are a prime offender.) And after you have gone to all that trouble, your market is limited because most electronic hobbyists aren't interested in building a surface mount kit. It is becoming increasingly difficult to design a kit that uses old fashioned through-hole parts; many components now cost far more in that form, are no longer available, or were never made in through-hole versions at all. Small kit companies still exist to sell to electronic hobbyists and amateur radio operators, but it's because those people like to build things and will buy kits even if they don't save any money.
Fighting against those trends meant that Radio Shack's business of selling components would have declined no matter what they did. But they worsened the problem by their own actions. They hired the wrong people for that business, gave them the wrong training, and offered the wrong pay structure. This is all assuming that the company actually wanted to sell any parts rather than merely keep them around as a vestige of their old days.
The pay structure matters because the salaries of the people on the Radio Shack were poor; the only way to make any money at the job was on sales commissions, and those were only significant for the large ticket items. (In the last few years that has mostly meant phones; farther back it would have also included things like CB radios and stereo components.) Sales people don't want to spend any time helping you find parts because there is nothing in it for them, and it takes them away from possibly serving a customer who might buy something that they will earn some commissions on. Getting a reputation in the store for being knowledgeable about the parts could be counterproductive; the other salespeople would then foist all those customers onto you while they served the more lucrative phone buyers.
If Radio Shack had wanted to have people on the floor who actually had a clue about electronics and electronic parts, they would have had to pay them more. And it probably would have helped to take the further step of eliminating commissions and simply giving people a good salary. Commissions have lots of problems, including the perverse incentives that I mentioned above, and the fact that they make the salespeople enemies of each other rather than allies. But I think the core problem is that they make the self interest of the salesperson be directly opposed to the self interest of the customer. The sales person wants to just sell you as much crap as possible as quickly as possible because that maximizes their commission. Given the lousy pay, they're probably not planning to stay in the job long enough to care about repeat business.
Putting the parts in drawers didn't help. It made the parts business less visible. It made it difficult for hobbyists to browse the parts and perhaps find something they wanted it to buy, making it something that was only useful if you were looking for a specific thing.
Capacitor values still matter in analog circuits. If you're building audio or radio gear, you'll need some that aren't bypass or power filtering values. Digital circuits now rarely contain analog timing circuits or filters (earlier ones did have delay lines) so you're rarely going to need capacitors in values that aren't powers of 10.
So they're doing a refresh to bring their systems a bit more up to date. The real world benefits are small; they'll be about 10% faster than the models they replace and will have slightly longer battery life. Streaming fans who hook up their laptops to an external display will get one more benefit: the new systems will presumably support 4K streaming because they'll have the required Kaby Lake processors.
Updating the Air is a bit more of a step forward since it's currently still using a 5th generation processor; it might get a 20% speed improvement. Apple probably was thinking that it would go end of life by now, but they still have a lot of presence in the higher education market and the Air is their main product there. The ultrathin MacBook is a bit too limited, and I don't think Apple is prepared to cut the price to the under-$1000 level that they need to reach. The MacBook Pro is too expensive; the CS and media studies majors will buy it because they need the extra power but most students will not.
The question if they do update the Air: will they just shove in a newer CPU and leave the rest of the design as it is, or will they decide that the product has a longer term place in their product lineup and give it a more substantial refresh? The relatively low display resolution stands out as the #1 thing they could improve, though that would bring up the question of cannibalizing their more expensive products. It's also likely that a new version would move to USB-C for charging and connectivity - but would they go all-in as they have on their other products, or offer a mix of USB-C and legacy USB ports as most Windows laptops have?
The ultrathin MacBook might see some design changes as well, though not as major. The most popular thing Apple could do with that system would be to shoehorn in at least one more USB port; that would allow the user to connect something and charge the system at the same time without the need for yet another dongle. The MacBook Pro design is less than a year old, so I don't expect any changes there other than the CPU and perhaps some tuning of the available configurations (amounts of RAM and flash memory, etc).
The author isn't saying that electric cars will replace internal combustion cars as something that most people own. His actual claim is that private car ownership will mostly go away, with autonomous electric cars being the technology that makes that possible. A very different premise, and one that deserves analysis.
First up is the question of whether autonomous car technology will be ready quickly enough for his time frame. To meet the 2024 time point, autonomous cars have to be ready for mass production by 2020 at the latest and receive regulatory approval to be sold and driven autonomously. I think that's an overly optimistic date, especially for the legal part of the equation; that technology will come but I think it will take a bit longer than he believes
It will take quite a few years to replace the entire installed base of vehicles. Currently there are about 250 million small vehicles (cars, SUVs, pickup trucks) registered in the US and 17.5 million sold each year. At those rates it would take about 12 years to replace them all. Moving toward shared autonomous cars will reduce the total number needed, but it's still likely to take more than four years to make enough of them.
Another big question: will people be willing to forgo car ownership and do most of their travelling by using services such as Uber and/or public transit? (Uber itself won't necessarily be one of the important players or even survive, but if not somebody else will rise up to replace them.) In cities I think the case is strong; the economics of car ownership are unappealing, driving and parking are challenging and costly, and a ride in an autonomous vehicle has the potential to be much less expensive than taxis or the existing Uber because of eliminating labor costs for the drivers.
But outside the city things get murkier. Maintaining a viable car sharing service requires a high density of use, which means dense population. As you leave the city and go into the suburbs that density disappears, and the viability of a shared infrastructure of autonomous cars drops. A further problem is the fact that the places where people live and the places where they work are well separated, so the pattern of use doesn't suit the model; people will want to take their cars to their workplace, where they will sit unused or underused for most of the day until the workers are ready to return home. Meanwhile, there are few cars left in the bedroom communities for the people who are still there and want rides.
Eventually you reach rural areas, and I don't see the shared vehicle model taking hold there any time soon. There simply aren't enough people living there, and the distances people routinely travel are a barrier for electric cars.
A final question is whether people are ready for the psychological shift of giving up car ownership. I suspect that for many people, the answer is no. They see their cars as personal space, much like their homes are, and won't want to give that up. The extent to which that is true is also likely to correlate with where they live. Many people who live in the city are already using public transit and are comfortable with travelling in vehicles that they do now own or control, but people who have moved out of cities do not. I have talked to people who find the idea of using public transit horrifying for that reason, and they also choose to drive many hundreds of miles on long trips rather than using a bus or train because of their preference for private travel. (Flying has become its own horror show and I can understand people wanting to avoid that.)
There is also likely to be some resistance to autonomous vehicles themselves, aside from the question of ownership. Some people who value independence and private travel will not be prepared to let a machine drive for them because it involves giving up control; the fact that the autonomous car is likely to be safer won't sway them. Eventually some jurisdictions may move toward banning human driving for safety reasons and that will set off a firestorm of controversy.
Those calculators have no network connectivity so the amount of data that the calculator company gets is limited. If you buy one at a campus bookstore it's possible that the store passes along some demographic info.
And you can't use them during tests because you might use the smartphone to cheat. You're required to use one of a few models of dedicated calculator that the test givers know do not have any programmable features. Sometimes the list is limited to TI; other schools have slightly more inclusive lists that include HP and Casio models.
No, not lying. The current version of the TI-84 is just under $110 on Amazon: https://www.amazon.com/Texas-I... If your kid buys one in a brick and mortar store it's probably a bit more expensive.
It's not the number of copies. It's that the OEM license only authorizes a bare iron install. The only Microsoft licenses that authorize virtual machine installations are full retail copies and certain forms of enterprise licensing, plus the exception of XP Mode in Windows 7 Professional which gave you a license for one copy of XP in a virtual machine on the system that was running W7. Upgrade licenses can be used for a virtual machine install IF they are used as an upgrade of a full retail license; they retain the same rights that the upgraded copy had.
Even then, the licensing terms are unreasonably restrictive in that they only authorize ONE virtual machine. Not one running at a time, one period. (The licensing terms for virtual machines under enterprise licensing are more reasonable and do allow multiple VMs.) The net effect is to eliminate nearly all reasonable uses of virtual machines for home and small office users. And if you want both Windows and Linux on your system simultaneously you're forced to run a Windows host and a Linux VM, because doing it the other way around will cost you a bunch of additional money.
I will grant that Microsoft has legitimate business reasons for some of the restrictions. A case they're trying to prevent is somebody running a hosting company or a remote terminal server on the cheap. If they allowed you to run unlimited VMs, somebody could buy a big honking piece of server-grade hardware with hundreds of gigabytes of RAM and run dozens of clients in virtual machines while only using one Windows license. I think a more reasonable restriction would be to stipulate that you can run as many VMs as you like with one license, so long as they are all used by the same human being at any given time. They could not be used to provide a UI to somebody else, nor to offer services to other people or computers.
One problem is that it would also put an effective kill date on the computers themselves. Sometimes upgrading isn't an option: the computer itself or something that is installed in it or connected to it is not supported by the newer OS.
Upgrading computers from versions of Windows earlier than Windows 7 to a current version also costs money - and a lot more of it than the $20 that Apple charges for upgrades from some old versions of macOS. Upgrading from 7 or 8 to 10 also costs money now but there are ways around it.
Microsoft doesn't care about the XP systems. The reason they felt the need to push an XP update this time is because this piece of malware propagates peer to peer, and thus infected XP systems threaten the systems that Microsoft DOES care about.
And before anybody gets on my case about a similar disagreement: I am talking about the word hierarchies, which is a singular thing, not the hierarchies themselves which are plural. Thus "which is plural" is correct.
Comments 6a and 6b are incorrect. The second part of that sentence is referencing hierarchies, which is plural. What is being referenced is unclear from structure but clear from context; nobody except the most extreme anarchist believes that society could possibly be going away, but some people would advocate the elimination of hierarchies.
One deterrent to doing that is that it will cost you extra. The Windows license that comes with your computer licenses one copy on that computer. A virtual Windows machine ON THE SAME COMPUTER does not qualify, so you would have to buy another license for that.
It was actually Scott McNealy, co-founder of Sun Microsystems, who said that. More accurately, the quote was "You have zero privacy anyway. Get over it." If Eric Schmidt ever said something similar he would have been referencing McNealy.
Chrome on x86 is not yet 64 bit only, and won't be for some time - there are still too many machines running the 32 bit version of Windows. What changed recently is that they automatically install the 64 bit version on compatible systems, rather than people specifically having to download it, and have also started to offer to upgrade installations of 32-bit Chrome to the 64 bit version. It probably won't be long until they REQUIRE the upgrade to 64 bit on 64 bit Windows and only allow you to run the 32 bit version on 32 bit Windows. Google will eventually drop support of 32 bit Windows but it's probably a couple of years away.
Chrome on macOS and Linux is 64 bit only. That does mean that the small number of people still running 32 bit versions of those OSes can't run current versions of Chrome. On the Mac that only applies to ancient versions of macOS that are no longer supported anyway. But many current Linux distributions still offer 32 bit versions because lots of Linux users run legacy hardware; part of the appeal of Linux is that you can run current versions on hardware that is no longer supported by other OSes. Chrome won't work on those, though its open source relative, Chromium, can still be built for 32 bit Linux for now.
Literally, the Golden Palm. That's palm as in tree, not as in part of the hand.
The Micro USB connector has some of the same problems, because it may have add-on capabilities like USB On-The-Go and MHL. But the USB-C connector is far worse. On the plus side, the connector is more physically robust than the fragile Micro USB connector.
That's actually part of the problem. There are a bunch of ports that look exactly like USB-C ports but they don't all do the same thing. You see something on your computer or phone that looks like a USB-C port but it could be any number of things:
1. A bog standard USB-C port that implements Superspeed Gen 2 USB (10Gbps)
2. A USB 3.0 standard port (5Gbps)
3. A USB 2.0 standard port (480Mbps) - these are mostly on phones
4. A Thunderbolt 3 port that implements some unknown number of PCIe lanes
5. A USB-C port that implements HDMI Alternate Mode (USB-C to HDMI with no dongle)
6. A USB-C port that implements DisplayPort Alternate Mode (USB-C to DP with no dongle)
7. A USB-C port that will charge the device
What's the problem? You can't tell which set of capabilities a USB-C port has by looking at it. There are no standard device markings or color coding to tell the various kinds of ports apart. The only thing that will help is a spec sheet for the device. A computer with more than one USB-C port may not have the same capabilities on all of them, so whether a device will work depends on which port you plug it into.
I foresee a surge in routings via Canada and Mexico.
In some cases, Apple explicitly bakes in a home team advantage. The notable example is web browsers, where you are required to use the HTML rendering and Javascript engines from iOS rather than including your own. Those things are essentially the same ones that were in previous versions of Safari, so you are guaranteed that your alternative browser will offer worse performance than Apple's browser. You can attempt to compete with improvements in features and user interface, but it's a hard sell when your core performance is hobbled by Apple's restrictions.
FreeBSD has never gotten the same level of desktop adoption as GNU/Linux. So the GUIs are less refined and there are fewer desktop applications available in the repositories. For most people, a suitable distribution of GNU/Linux is a better choice on a desktop client.
However, FreeBSD is a great server OS. It has a very high performance network stack and a more fully developed and stable version of the ZFS file system. I have used it successfully on servers, both in its basic form and in FreeNAS, a distribution for storage servers that is based on FreeBSD. If you're choosing your server OS for function rather than ideology, FreeBSD in the server room is worth a look.
I would not describe FreeBSD as "more free" than GPL software. Rather, it protects a different set of rights. It's impossible to protect ALL types of freedom; some are in conflict. Stallman believes that the rights that the GPL protects are the more important ones.
Two rights that the GPL protects that the BSD license does not: a developer's ability to use and adapt all code that is based on their code, and the right of end users to use all code that is based on a given code base.
On the other hand, the BSD license allows code to be reused in non-free software, something that the GPL prohibits. Stallman does not consider that important because it means that access to that derivative code can be denied, which is a right that he considers more important. But some people consider that right to proprietary adaptation to be an important one.
Conflicting freedoms and rights are frequently seen. One simple expression of it: the principle that "your right to freely swing your arms about ends at my face". My right to not be hit by your hands is a more important one than your right to put your arms wherever you wish. Smoking is an issue where public opinion has changed over time; the right to breathe smoke-free air is now considered more important than the right to smoke in most situations.
The Pentium Pro was released in November 1995. Systems became readily available in 1996. It was not the success that Intel had hoped for in desktop applications because it had a serious performance issue when running 16 bit legacy code. (It was more popular on servers, which had migrated to 32 bit code by then.) Referencing both the AL register (the lower half of AX) and the AX register in the same code sequence caused a complete pipeline flush, as did loading a segment register. Both of these things were rather common in 16 bit code.
Desktop adoption of Intel's new P6 microarchitecture didn't really take off until the introduction of the Pentium II in May 1997; it included a segment register cache and improved handling of register aliasing to solve those two problems so it performed much better on old code. It was also much cheaper for Intel to manufacture because it eliminated the multi-die packaging that the Pentium Pro used to incorporate cache and instead used a pluggable CPU module that contained the processor and the cache RAM. That was succeeded by the Pentium III, which started with the same SECC modules as Pentium II but later moved to on-die RAM when it became possible to put more transistors on one chip.
After THAT we got the failure that was Netburst. But that's a story for another day.
Except for late night talk shows there will be little immediate effect. Any scripted show that will premiere during the summer is already written, if not already filmed. But some fall shows are currently in production and may not have all their scripts in hand; if the strike drags on they could be affected.
We're not at the limit of potential natural gas production. But as more extreme fracking becomes necessary to extract it, the cost goes up. The cost of wind and solar continues to come down. So we're probably approaching peak natural gas production because it won't be cost effective to produce more.
The demand for fossil fuels will gradually decline rather than falling off a cliff because there is a large installed base. All those internal combustion vehicles and power plants won't be replaced overnight. We're at the point where it is no longer cost effective to build a new coal power plant and will soon reach the point where it is not cost effective to build a new natural gas plant. We're approaching the point when it is no longer cost effective to operate an existing coal plant where the cost of construction is a sunk cost, but that tipping point for natural gas is still some years off.
Electric cars are not yet cost effective without the help of tax subsidies, and the charging infrastructure for long range travel is still lacking. Battery prices are dropping, but the price of a long range electric car like the Chevy Bolt (and upcoming Tesla Model 3, Nissan Leaf version 2, Hyundai Ioniq, etc.) will have to drop by another $10,000 to make the lifetime cost competitive with gasoline. At that point they'll still cost more than gas-powered cars but the reduced fuel and maintenance costs will make up the difference.
Demand for oil and natural gas won't fall to zero in the near future but demand will eventually drop dramatically. Even if we stop burning them, which would be a good idea for the environment, they are useful for manufacturing petrochemicals. Oil has uses as a lubricant. And we might find it useful to maintain a limited number of fossil fuel power plants as backups in case there is a long stretch when solar and wind sources underperform due to adverse weather conditions.