It doesn't matter whether they did or not. The fact that they effectively control distribution of a number of essential apps means that they can then use this to force phone vendors to install other Google apps. Both US and EU antitrust law agree on this point: It doesn't matter how you came by your first monopoly, you aren't allowed to use it to gain a second.
Let's take a web browser as an example. Chromium has a number of sandboxing strategies that provide different levels of protection. On Linux it can use chroot (which can't restrict network access), SELinux or seccomp-bpf (which aren't great because they don't differentiate between different instances of the same program, so one sandbox can access anything that any sandbox can access). On OS X, it uses the TrustedBSD-based sandboxing framework. On Windows it uses kernel ACLs. On FreeBSD, it can use Capsicum, which provides the best isolation (though this support isn't fully upstreamed, it's in progress because Google is porting Capsicum to Linux for the ChromeBook). On OpenBSD, it can only use chroot, because the OpenBSD developers believe that complexity is the enemy of security and don't implement security features for userspace software to use (they did have systrace, but it was shown to be trivially vulnerable to timing attacks and never fixed).
Bottom line: the features that a kernel provides can have a big impact on overall system security, and OpenBSD lacks a lot of the security features that you'd expect from a modern system.
Campaign contributions, on the other hand, are not speech.
The problem is that this distinction is abused. You may not be allowed to donate to a person's campaign, but you can pay large amounts to have ads run on a particular issue that just so happens to be one of the core parts of a particular candidate's campaign platform.
You seem to fundamentally misunderstand how antitrust regulations work. You need to have a sufficiently large market share that your actions distort the market to be considered a problem. I could release a smartphone tomorrow that had a single app store and was completely obnoxious in every single way that people have complained about Microsoft, Apple, and Google, but I would be exempt from any antitrust exemptions because no one would buy my CrapPhone and it would have no impact on the market.
I thought that the Android location stuff let any map application register for handling positions and addresses? I've certainly had the calendar app open OSMAnd on my phone from an address in an appointment. Are there other map APIs that don't support different service providers?
Try doing mobile banking without Google. My bank provides their Android App through the Google Play store and via no other channels. The same is true for a lot of other banks. Google has intentionally made sure that they control the app distribution channel, so useful devices need to have the Google Play app installed. Of course, the contract to have Google Play preinstalled means that you then have to have all of the other Google apps preinstalled on devices that you sell.
The advantage of the segway over a bike as a tourist is that it requires a lot less concentration and is easier to stop and start. It's much easier to look at your surroundings on a segway and stop to take a closer look (and to turn sharp corners).
Most likely because they aren't doing any of the firmware signature verification in hardware and so releasing specs that would allow people to write drivers to load arbitrary firmware would make it trivial for malware to do the same (not that a bit of reverse engineering won't do that eventually, but companies do love security by obscurity).
Power management that didn't suck, a sane capability-oriented security model, a realtime nanokernel that allowed you to run the baseband and application stack on the same core, to name a few.
Symbian still runs circles around Linux as a kernel for mobile devices, the problem was failing to update the userland APIs to something that didn't constrain developers heavily so that code could scale down to devices with under 2MB of total memory when every phone started shipping with 256-512MB as a minimum. The systematic problem was that Nokia had at least four separate projects to replace the Symbian userspace with something modern, all handled by teams that did their best to sabotage the efforts of the others - very successfully, I might add. You can't run a company in a competitive market when your middle management is more interested in competing with other groups within the company than with other companies.
While the company is burning down around these guys are making cash up front from the contract signed
And that's the problem. Don't pay sales guys a bonus for signing a contract, pay them a big bonus related to the profit that the contract brings in once you ship it.
If you're not expecting users to remember them, why use passwords at all? There are standard HTML facilities for generating keypairs in the browser. If you want to make it easy to share logins with mobile devices, then something that turns a private key into a QR code for easily copying to the mobile device would work fine.
Nokia is a huge telecoms company that is most well known by consumers for making mobile phone handsets, though this was a relatively small part of their total product line. Microsoft bought the mobile handset division. Then remainder of the company has a market cap of around $30bn. This means that, including stock, they easily have enough capital to buy another company for $16bn (the $7bn in cash from MS probably helped though).
It sounds like someone at Nokia realised that mobile phones were in a race to the bottom and the profit is in the back-end infrastructure.
Safari has had this feature for quite a while. The dialog box has two buttons, 'allow' and 'don't allow', which is pretty poor UI. They ought to be 'No', and 'No, why the fuck would I ever want that?'
There's nothing preventing them from driving the tapes to a safety-deposit box every week.
And that adds to the cost of tape.
And you trust them why?
Because you're giving them money and because it's not that hard to arrange a second source. Try building a business that doesn't rely on at least one other company for some core part of its infrastructure.
And how much does that cost compared to a tape drive? Even gigabit ethernet isn't that fast to a single random internet site, and is pretty slow compared to the speed a tape drive operates at. Seriously, how is some little company with 20 employees supposed to afford this? $2k for a tape drive and some flunky IT employee to run backups every week is not a big cost for a small business.
For 20 employees, a central file server that backs up to a rotating set of external hard disks that the managing director takes home in the evening is much cheaper than tape and provides off-site backup. That's the kind of arrangement I've seen in such companies. Tape just isn't economical for them because, outside of a few industries, they're not producing enough data to justify more than a couple of hard disks worth of backups. Oh, and you really think 20-employee businesses have dedicated IT people? They'll occasionally hire a consultant to set things up, but most of the time IT will be done by the boss's son or someone else who seems to know a bit about computers. Cheap off-the-shelf backup programs (including the built-in Windows backup utility) can do nightly backups to an external disk and then it's just a matter of rotating them.
That $2K tape drive isn't economical for anyone. The companies that are still buying tape are not buying $2K drives for backup, they're buying them so that they can restore from accidental deletions without interrupting a backup - they're buying $20-30K tape robots for real backup. If you're a company that can afford a $30K tape robot (and to keep it fed with tape), then you're not far off wanting to have, if not your own datacentre, a rack or two in a colo, and at that point tape stops being interesting.
On-site backup is not backup, it's just wishful thinking for a business.
How do you know they're backing up your data?
They are the backup.
How do you know they're not selling your data to someone?
Because you encrypt it first. And because, if you're a business, you solve this in the same way that you solve the other issues: a contract with steep penalty clauses.
And that's a lot of bandwidth too.
There's a reason that business broadband exists and provides 1+Gb/s speeds...
The first instruction in the Intel Architecture Reference (Part 2: Instruction Set Reference) is AAA, which is named after the noise made by people forced to read x86 assembly. It is short for 'ASCII Adjust After Addition' (yes, that should be AAAA, but that would be too consistent for Intel). This instruction exists to convert the result of binary addition into the result of the corresponding BCD addition.
Or, to put it in simpler terms: Anyone who thinks x86 assembly is not that difficult to understand is certifiably batshit insane.
$2K may be nothing in absolute terms, but the question is what that's compared to. $2K will buy you a file server off-site somewhere with snapshots. $2K only buys you a tape drive, not any kind of tape robot, so also factor in the cost of someone's time inserting the tapes and putting them in secure storage. By the time that's added together, you can easily buy two machines in off-site data centres for off-site backups to hard disks with snapshots and easy recovery (or, if you're not a tech-focussed company, you buy an off-the-shelf solution for remote backups that's cheaper than managing your own tapes). And that $2K was $1K for the previous generation of tapes. If you're serious about tape backups, you need at least $20-30K of tape robot and that's a much bigger investment than most small companies want.
Tape hasn't gone away, but (as I said in the last two posts), the set companies in the middle for whom it makes sense is gradually shrinking. It's far more expensive than the alternatives unless you're in the sweet spot (around 200TB-10PB), but that sweet spot is shrinking from both ends.
Here's the list, though a few are mis-filed (the arbitrary code execution from this year is actually in Flash, no idea why it appears here), but most of the privilege elevation ones from this year and most of the arbitrary code execution vulnerabilities are real (though several seem to be in Logitech HID drivers).
The problem is that every other attempt to focus on the high-end of the computing market has failed. You either let economies of scale work for you, or you let them work for a competing technology. Tape used to be the backup medium for SMEs. If you were backing up 5-10 computers worth of data on a regular basis then you wanted tape. Now, the point at which tape makes sense is 1PB, or around 1,000 workstations.
The problem with this is that tape doesn't make sense for people building their own datacentres, so you're trapped in the market where your customers are very big companies, that are investing a lot in IT infrastructure, but aren't the likes of Google or Amazon (who would help you build the economies of scale). The size of company where it makes sense to start using tape is growing, but the size where it stops is shrinking. As the market shrinks, the prices go up, and that shrinks the size even faster. Eventually no one who has a need for tape will be able to afford it. And, by eventually, I mean really soon.
Everyone who understands the economics of tech. Show me one technology that has gone from being cheap enough to include in high-end consumer desktops to being out of the price range of small to medium businesses and then become a big commercial success.
The problem is that tapes are going in the wrong direction. I remember tape drives costing about the same as a hard disk of a similar capacity and the tapes costing 10-20% the cost of the drive. That made it feasible to stick tape drives in high-end desktop computers so that people who cared about their data had a good backup system. The tape drive added 10-20% to the total cost of the machine (less if it included a decent monitor). Now, a tape drive with the same capacity as a hard drive costs around ten times as much as the disk and more than doubles the cost of the computer it goes in. It really doesn't make sense to buy tape drives: if you have enough data for tapes to be useful then you're going to be buying big tape robots. Tapes left consumer pricing a long time ago and have now left small business pricing too. That's the trajectory for a technology on its way to obsolescence.
If you buy drives from NetApp, they're insanely expensive compared to just buying the drives. One of the things you get for that is a guarantee that you don't have two drives from the same batch in the same array. Just buying two drives does not double the reliability: if they're from the same batch and subject to the same stresses then there's a good chance that they'll fail at the same time. When a drive fails in a RAID set, you want to have a long to resilver before the next one dies.
Slashdot Mirror
It doesn't matter whether they did or not. The fact that they effectively control distribution of a number of essential apps means that they can then use this to force phone vendors to install other Google apps. Both US and EU antitrust law agree on this point: It doesn't matter how you came by your first monopoly, you aren't allowed to use it to gain a second.
Paying for someone else's bills is always welcome in my book
In antitrust circles, this is called 'dumping' and can have very bad effects on the health of the market.
And I'd rather pay $20/month for 1Gb/s. What? If we're making up totally unrealistic markets then I can have the one that I want too!
Let's take a web browser as an example. Chromium has a number of sandboxing strategies that provide different levels of protection. On Linux it can use chroot (which can't restrict network access), SELinux or seccomp-bpf (which aren't great because they don't differentiate between different instances of the same program, so one sandbox can access anything that any sandbox can access). On OS X, it uses the TrustedBSD-based sandboxing framework. On Windows it uses kernel ACLs. On FreeBSD, it can use Capsicum, which provides the best isolation (though this support isn't fully upstreamed, it's in progress because Google is porting Capsicum to Linux for the ChromeBook). On OpenBSD, it can only use chroot, because the OpenBSD developers believe that complexity is the enemy of security and don't implement security features for userspace software to use (they did have systrace, but it was shown to be trivially vulnerable to timing attacks and never fixed).
Bottom line: the features that a kernel provides can have a big impact on overall system security, and OpenBSD lacks a lot of the security features that you'd expect from a modern system.
Campaign contributions, on the other hand, are not speech.
The problem is that this distinction is abused. You may not be allowed to donate to a person's campaign, but you can pay large amounts to have ads run on a particular issue that just so happens to be one of the core parts of a particular candidate's campaign platform.
You seem to fundamentally misunderstand how antitrust regulations work. You need to have a sufficiently large market share that your actions distort the market to be considered a problem. I could release a smartphone tomorrow that had a single app store and was completely obnoxious in every single way that people have complained about Microsoft, Apple, and Google, but I would be exempt from any antitrust exemptions because no one would buy my CrapPhone and it would have no impact on the market.
I thought that the Android location stuff let any map application register for handling positions and addresses? I've certainly had the calendar app open OSMAnd on my phone from an address in an appointment. Are there other map APIs that don't support different service providers?
Try doing mobile banking without Google. My bank provides their Android App through the Google Play store and via no other channels. The same is true for a lot of other banks. Google has intentionally made sure that they control the app distribution channel, so useful devices need to have the Google Play app installed. Of course, the contract to have Google Play preinstalled means that you then have to have all of the other Google apps preinstalled on devices that you sell.
The advantage of the segway over a bike as a tourist is that it requires a lot less concentration and is easier to stop and start. It's much easier to look at your surroundings on a segway and stop to take a closer look (and to turn sharp corners).
Most likely because they aren't doing any of the firmware signature verification in hardware and so releasing specs that would allow people to write drivers to load arbitrary firmware would make it trivial for malware to do the same (not that a bit of reverse engineering won't do that eventually, but companies do love security by obscurity).
Power management that didn't suck, a sane capability-oriented security model, a realtime nanokernel that allowed you to run the baseband and application stack on the same core, to name a few.
Symbian still runs circles around Linux as a kernel for mobile devices, the problem was failing to update the userland APIs to something that didn't constrain developers heavily so that code could scale down to devices with under 2MB of total memory when every phone started shipping with 256-512MB as a minimum. The systematic problem was that Nokia had at least four separate projects to replace the Symbian userspace with something modern, all handled by teams that did their best to sabotage the efforts of the others - very successfully, I might add. You can't run a company in a competitive market when your middle management is more interested in competing with other groups within the company than with other companies.
While the company is burning down around these guys are making cash up front from the contract signed
And that's the problem. Don't pay sales guys a bonus for signing a contract, pay them a big bonus related to the profit that the contract brings in once you ship it.
If you're not expecting users to remember them, why use passwords at all? There are standard HTML facilities for generating keypairs in the browser. If you want to make it easy to share logins with mobile devices, then something that turns a private key into a QR code for easily copying to the mobile device would work fine.
Nokia is a huge telecoms company that is most well known by consumers for making mobile phone handsets, though this was a relatively small part of their total product line. Microsoft bought the mobile handset division. Then remainder of the company has a market cap of around $30bn. This means that, including stock, they easily have enough capital to buy another company for $16bn (the $7bn in cash from MS probably helped though).
It sounds like someone at Nokia realised that mobile phones were in a race to the bottom and the profit is in the back-end infrastructure.
Safari has had this feature for quite a while. The dialog box has two buttons, 'allow' and 'don't allow', which is pretty poor UI. They ought to be 'No', and 'No, why the fuck would I ever want that?'
There's nothing preventing them from driving the tapes to a safety-deposit box every week.
And that adds to the cost of tape.
And you trust them why?
Because you're giving them money and because it's not that hard to arrange a second source. Try building a business that doesn't rely on at least one other company for some core part of its infrastructure.
And how much does that cost compared to a tape drive? Even gigabit ethernet isn't that fast to a single random internet site, and is pretty slow compared to the speed a tape drive operates at. Seriously, how is some little company with 20 employees supposed to afford this? $2k for a tape drive and some flunky IT employee to run backups every week is not a big cost for a small business.
For 20 employees, a central file server that backs up to a rotating set of external hard disks that the managing director takes home in the evening is much cheaper than tape and provides off-site backup. That's the kind of arrangement I've seen in such companies. Tape just isn't economical for them because, outside of a few industries, they're not producing enough data to justify more than a couple of hard disks worth of backups. Oh, and you really think 20-employee businesses have dedicated IT people? They'll occasionally hire a consultant to set things up, but most of the time IT will be done by the boss's son or someone else who seems to know a bit about computers. Cheap off-the-shelf backup programs (including the built-in Windows backup utility) can do nightly backups to an external disk and then it's just a matter of rotating them.
That $2K tape drive isn't economical for anyone. The companies that are still buying tape are not buying $2K drives for backup, they're buying them so that they can restore from accidental deletions without interrupting a backup - they're buying $20-30K tape robots for real backup. If you're a company that can afford a $30K tape robot (and to keep it fed with tape), then you're not far off wanting to have, if not your own datacentre, a rack or two in a colo, and at that point tape stops being interesting.
The off-site storage option is high-risk
On-site backup is not backup, it's just wishful thinking for a business.
How do you know they're backing up your data?
They are the backup.
How do you know they're not selling your data to someone?
Because you encrypt it first. And because, if you're a business, you solve this in the same way that you solve the other issues: a contract with steep penalty clauses.
And that's a lot of bandwidth too.
There's a reason that business broadband exists and provides 1+Gb/s speeds...
The first instruction in the Intel Architecture Reference (Part 2: Instruction Set Reference) is AAA, which is named after the noise made by people forced to read x86 assembly. It is short for 'ASCII Adjust After Addition' (yes, that should be AAAA, but that would be too consistent for Intel). This instruction exists to convert the result of binary addition into the result of the corresponding BCD addition.
Or, to put it in simpler terms: Anyone who thinks x86 assembly is not that difficult to understand is certifiably batshit insane.
$2K may be nothing in absolute terms, but the question is what that's compared to. $2K will buy you a file server off-site somewhere with snapshots. $2K only buys you a tape drive, not any kind of tape robot, so also factor in the cost of someone's time inserting the tapes and putting them in secure storage. By the time that's added together, you can easily buy two machines in off-site data centres for off-site backups to hard disks with snapshots and easy recovery (or, if you're not a tech-focussed company, you buy an off-the-shelf solution for remote backups that's cheaper than managing your own tapes). And that $2K was $1K for the previous generation of tapes. If you're serious about tape backups, you need at least $20-30K of tape robot and that's a much bigger investment than most small companies want.
Tape hasn't gone away, but (as I said in the last two posts), the set companies in the middle for whom it makes sense is gradually shrinking. It's far more expensive than the alternatives unless you're in the sweet spot (around 200TB-10PB), but that sweet spot is shrinking from both ends.
Here's the list, though a few are mis-filed (the arbitrary code execution from this year is actually in Flash, no idea why it appears here), but most of the privilege elevation ones from this year and most of the arbitrary code execution vulnerabilities are real (though several seem to be in Logitech HID drivers).
The problem is that every other attempt to focus on the high-end of the computing market has failed. You either let economies of scale work for you, or you let them work for a competing technology. Tape used to be the backup medium for SMEs. If you were backing up 5-10 computers worth of data on a regular basis then you wanted tape. Now, the point at which tape makes sense is 1PB, or around 1,000 workstations.
The problem with this is that tape doesn't make sense for people building their own datacentres, so you're trapped in the market where your customers are very big companies, that are investing a lot in IT infrastructure, but aren't the likes of Google or Amazon (who would help you build the economies of scale). The size of company where it makes sense to start using tape is growing, but the size where it stops is shrinking. As the market shrinks, the prices go up, and that shrinks the size even faster. Eventually no one who has a need for tape will be able to afford it. And, by eventually, I mean really soon.
Everyone who understands the economics of tech. Show me one technology that has gone from being cheap enough to include in high-end consumer desktops to being out of the price range of small to medium businesses and then become a big commercial success.
The problem is that tapes are going in the wrong direction. I remember tape drives costing about the same as a hard disk of a similar capacity and the tapes costing 10-20% the cost of the drive. That made it feasible to stick tape drives in high-end desktop computers so that people who cared about their data had a good backup system. The tape drive added 10-20% to the total cost of the machine (less if it included a decent monitor). Now, a tape drive with the same capacity as a hard drive costs around ten times as much as the disk and more than doubles the cost of the computer it goes in. It really doesn't make sense to buy tape drives: if you have enough data for tapes to be useful then you're going to be buying big tape robots. Tapes left consumer pricing a long time ago and have now left small business pricing too. That's the trajectory for a technology on its way to obsolescence.
If you buy drives from NetApp, they're insanely expensive compared to just buying the drives. One of the things you get for that is a guarantee that you don't have two drives from the same batch in the same array. Just buying two drives does not double the reliability: if they're from the same batch and subject to the same stresses then there's a good chance that they'll fail at the same time. When a drive fails in a RAID set, you want to have a long to resilver before the next one dies.