#1: obey the law and require the government to actually take the trivial steps required to get warrants in a FISA court. You protect the public's rights, protect your own backside, and force government to follow its own rules.
#2: break the law and become criminals. You break the law and encourage the intelligence community to be lazy (get everything and sort through it later).
How is this a difficult choice again? #2 really doesn't help anyone. The only explanation I can think of for companies bending so easily is if they did so in exchange for the government looking the other way about something else. I'd be interested to find out what.:-)
Four words: piercing the corporate veil. Once the company is opened up to civil liability, lawsuits will be filed, and during discovery, those lawsuits will likely uncover information about who knew what and when. At that point, depending on what they turn up, criminal charges might be filed against some of them.
Facebook account: Free.
iPhone: $199
Logging into Facebook from your iPhone instead of the victim's computer so you don't get arrested for burglary: priceless.
There are some things money can't buy. For everything else, there's Mastercard.
It's actually pretty simple to get most of the way there:
The entire OS is write protected.
No code allowed in the kernel unless signed by a key from a company whose key is signed by the OS vendor, with a command-line software switch to disable the check for geeks who want to tinker, confident in the knowledge that people are more terrified of the command line than they are of anything else, including malware.
All applications and plug-ins must be self-contained bundles of files. No outside helpers may be installed anywhere.
Code outside an application or plug-in bundle is not executable except while the "I'm a geek/coder" switch mentioned above is set.
All applications and plug-ins must be signed unless the "I'm a geek flag" is set.
All applications and plug-ins immediately become write protected upon drop. The kernel must ensure that the drop operation is atomic from the perspective of applications.
Applications and plug-ins must use a system-provided service for self-updating, and must include a list of valid source domains in a property list within the app or plug-in bundle to prevent plug-ins from being able to masquerade as the app and overwrite it.
A central plug-in manager UI should show each application and what plug-ins are installed, allowing you to enable and disable them at will.
The plug-in manager must not use any APIs whose behavior can be altered by any system-level plug-in.
All applications that support plug-ins must provide a matching dictionary to indicate the type of plug-ins it supports. Plug-ins must provide a matching dictionary to indicate the services they vend. An OS-provided service then provides a list of matching plug-ins to the application upon request, and tells the application which plug-ins the user has enabled/disabled.
A plug-in shall be disabled until explicitly enabled by the user in the plug-in manager. The application may, however, open a URI that brings up the correct pane in the PIM to simplify the user experience.
All background daemons and cron jobs must be manually enabled by the user.
All reads and writes to user files (all files outside of the preferences or caches folders) can only occur after the file is opened through a standard file dialog, and only until the application closes the file.
All applications must provide a list of exported symbols against which a plug-in can link. Access to other symbols will fail.
All plug-ins should be run out of process by marshaling the exported function calls across IPC. Attempts to read arbitrary pointers will fail. The OS should provide an mallocForPlugIn() function to allocate a memory region shared with plug-ins.
With those relatively minor changes, arbitrary code execution bugs (unless you can find a vulnerability in a system service that runs as root) can only succeed in destroying any currently open user file, and can neither permanently inflict harm on the OS or apps, nor persist after the app quits in any significant form. And to the extent that malware could cause harm by tricking the user into installing a system-level plug-in that causes bad behavior, removing such malware would require at most one click in the plug-in manager. At that point, 99.999% of the malware problem goes away. Sadly, these ideas are pretty similar to the ones I came up with when I was still in elementary school. They're beyond obvious....
I can't imagine any store charging for listening to previews. Well, maybe somebody might, but most would probably roll it into the cost of doing business and raise prices as needed. I suppose you'd end up paying for it either way, but I wouldn't expect it to be explicit.
Presumably, if I had songs on iTunes, then yes, I'd get some fraction of a cent or something. ASCAP and BMI don't collect for everybody. You're thinking of Sound Exchange. ASCAP collects only for its members, as do BMI and SESAC. For any composers or publishers who aren't a member of one of those three performing rights societies (or any of the other similar groups in other countries), the person making the recording or airing it or whatever generally has to track down the composer and publisher through some other means.
The net effect of this sort of thing, if passed, would probably end up being (at most) a fraction of a cent per play of such a snippet, which translates to almost nothing for all but the biggest artists, but if it ends up causing previews to be less accessible, would translate to a huge loss for all but the biggest artists (who probably wouldn't see much difference). Thus, I'd expect the biggest artists to want these laws and everybody else to think there's not enough crack in the world.:-)
I like Time Machine because you don't have to think about it much. When you're talking about backing up a non-server machine, being able to pull down a menu and choose "Back Up Now" is a real advantage. Imaging a drive every so often certainly isn't a bad idea, of course, but a disk image is to Time Machine as a hydrogen bomb is to a flyswatter....
No need to drop the music. Just stop making previews available. Then stand back, watch the impact on their sales, and wait for them to pull their heads out of their asses. As an ASCAP member myself, one of these days, I should write a letter or something to tell them to stop being idiots. *sigh* There's never enough time to correct all the idiots in the world. They are simply too great in number and proliferate too quickly.
In this case, for the Mac OS X installation, my answer would be the same as any other user (or at least the client side portion is the same):
Time Machine to an XServe attached to a giant hardware RAID 5 array via fibre channel. In other words, the same way I back up my laptop except with a serious server providing the disk instead of an ABS....
You should be able to back up the office file server to a Mac OS X Server box just as easily as you could back up to a Linux box, but the reverse isn't true. Backing up a Mac OS X installation with resource forks, extended attributes, etc. to a Linux box is nontrivial at best.... Been there, tried that, never got it working reliably.
Oh, yeah, and I've already had to get its transmission rebuilt back at about 85,000 miles. And people wonder why I will no longer buy American cars. Although foreign cars are probably no better at owning up to design flaws, at least they don't seem to have nearly as many serious design flaws....
Yes, they should. And IMHO Ford owes me a couple hundred in parts and about three or four hours of my time paid at their dealer's' obscene $200/hour repair rate for the design flaw in their valve cover that forced me to replace six port seals, I think eight isolator bolts, and some major gasket between the layers of the intake, scrub the intake, flush out the EGR hose, and replace the valve cover. I'm pretty sure I'll die before Ford covers those costs, though. It should have been an emissions recall, but instead, the local Ford dealer (Peninsula Ford of Sunnyvale, bless their thieving, lying souls who claimed that my <100,000 mile car probably needed to be replaced or at least have a new engine after they wasted almost five days and over $1500 in unnecessary and redundant repairs replacing parts that were almost brand new while trying to find a simple low-temperature-only coolant leak caused by a split in a metal line in the TOP section of the intake manifold and who only fixed my car after I tore their service manager a new asshole) said it would cost $1500 for them to put in that couple hundred bucks in parts. I did it myself in an afternoon with no car repair experience whatsoever. It's 40,000 miles later and my car is still working just fine. Go f**k yourself, Peninsula Ford of Sunnyvale.
Then again, the fact that the same Ford dealer did a half-assed copout fix of blowing out the carbon deposits instead of fixing the real problem just outside of warranty when it failed the first time (at which point Ford actually WOULD have been willing to cover the parts cost for the repair, or so I'm told) really takes the cake. After those two service experiences, I've been desperately hoping Ford would decide to drop that dealership in the whole financial mess, as they give Ford a really bad name. But I digress.
The fact that I had the problem in the first place and that Ford left me and tens of thousands of other 1999 Windstar owners footing the entire repair bill for what was clearly and beyond any doubt a design flaw in the engine is a pretty clear indication that the car companies won't ever own up to mistakes unless you can prove that it's a safety problem, and even then, if they calculate that the cost of the lawsuits from wrongful death will cost less than a recall, they still won't fix them. These sorts of corporations make me sick.
Whatever happened to companies standing behind their products? Isn't that the whole benefit of buying a product from a major company instead of just group buy importing it from a fab shop in China somewhere? Who wants to organize the automobile group buy?
Reprogramming the ECU with updated firmware is a good example of vehicle-specific behavior, but I think that it is more than reasonable for independent shops to be able to get that information. In many cases, reprogramming the ECU is required to fully fix certain DTCs because they have both a software and a hardware root cause---particularly when it comes to how certain sensors interact with fuel mixture on aging vehicles and other similar issues.
As someone who has dealt with OBD-II a bit, allow me to shed some light on the subject. The OBD-II standards specify a series of families of error codes. Codes within a certain range are chassis codes, another range for powertrain codes, another range for emissions control crap, etc. The precise details of what non-powertrain codes mean, however, are specific to each vehicle. For example, in most Chevy cars, C1780 means "Loss of Steering Position Signal". In Ford vehicles, it means "Temperature select failure". And IIRC, there are even some variations between specific models, though I don't have time to hunt for specific examples.
And even within the powertrain codes, the root cause can be vehicle-specific. For example, P0171 and P0174 are both codes for engine banks running lean on a '99 Windstar. They usually mean a vacuum leak caused by oil breakdown of port seals coupled with carbon deposits in the intake due to a flaw in the front valve cover. That's something the code number can't tell you.
Finally, some cars have multiple ECUs at multiple addresses. The chassis and brake codes might be in a different ECU, and AFAIK, that info is completely undocumented.
In short, generic scan tools generally give you a reasonable view of the powertrain codes and nothing more. Although it's better than nothing, it isn't a complete picture.
Or---and here's a brilliant idea---implement this only for commercial vehicles whose drivers already have to log their hours, routes, etc., have no expectation of privacy, and (I think) do disproportionately more damage to the roads that ordinary cars and light trucks per unit of fuel consumed. Let the people with alternative fuel and low fuel vehicles pay less in taxes because they already paid a huge chunk of extra change up front and are helping make our air cleaner. Call it a tax break for environmental conservation if you want to justify it.
I might accept the GPS idea if and only if I could be assured that the devices report only provide an annual report of the number of miles travelled per city or county as a summary. For tax purposes, no further detail is needed or useful, and as such, no further information should be allowable. By limiting it to annual reporting and cumulative mileage, this allows road funds to be distributed more accurately without providing any real info about who went where and when. That is the absolute most information I would be willing to tolerate without filing a lawsuit, however, and only then if they could prove beyond a reasonable doubt (with verifiable, auditable, mathematically proven software and firmware) that the devices *cannot* provide any other information.
This also means that roadside gathering of information is not acceptable because the mere fact that your car sent the data to a particular gathering station provides information that is greater than what is described above---specifically, your presence on that road at that time. In short, it must involve someone explicitly going to a center and getting scanned or plugging a thumb drive into it or unplugging it from the car and plugging it into a computer or... some explicit action by the user to start the data transfer.
Yes. If I cloned my phone and had two of them making calls on an unlimited plan, I would expect to be able to make 2880 minutes. Unless, of course, the plan explicitly forbids SIM cloning.
It's simply not true that a doctor can't do anything for you. We've had antivirals that are highly effective against influenza for a while now. However, it is important to limit the use of these antivirals to the most severe cases because of the risk of emergence of resistant strains. Thus, in general, doctors will not do anything for you unless there is reason to believe that you might not get over it on your own.
Unless the energy density is a huge win, at $2 per watt and lasting only a few months, that's pretty uninteresting, particularly when you consider that silicon is the second most abundant elements in the earth's crust. If it's pricey, there's something very wrong. I can pick up a handful of the dioxide form for free just a few blocks from here. Of course, there are usually cigarette butts and shell fragments in it....
Now if they were replacing germanium, selenium, gallium or indium arsenide, or some of the other rare elements/compounds that can be used in electronics, I could understand, but replacing sand? I just don't get it. It's not the silicon that makes solar panels expensive. It's all the other stuff.
It's an incomplete story. What we need to know to evaluate cost is A. life expectancy, and B. W/m^2. A solar panel that produces the same wattage for a price comparable to some of the higher density solar panels (IIRC) is cool if it lasts at least as long and has similar density. Otherwise, the replacement costs or the shipping costs and installation footprint make it more expensive, respectively.
Agreed. A boss doesn't have to be technical, nor "right" all the time, so long as the boss acknowledges what he or she doesn't know and doesn't try to pretend he/she knows something that he/she doesn't.
It is when a boss thinks he or she knows everything but actually knows nothing that errors are made. A boss can be completely clueless as long as he/she defers to your expertise. The ones we really can't stand are the ones who are clueless but don't know it. They give bad advice that leads their underlings repeatedly down wrong paths, then ding the underlings on salary reviews for listening to them.
Only slightly better are the bosses that let their employees graze and don't give them any guidance about what they are trying to accomplish. Neither type of boss is particularly effective, and both are, sadly, far more common than good bosses.
Only anecdotal evidence from having witnessed doctors prescribing drugs whose side effects are worse than the condition they try (and usually fail) to fix followed by prescribing drugs to try to fix those side effects, eventually leading to drug-induced dementia or other serious side effects. I've seen severe overmedication often enough to be disturbed by the problem, and even CNN has done stories on the problem, mostly as it applies to the elderly and to school-age children. I'm not aware of any widespread studies to quantify the problem, though.
Slashdot Mirror
#1: obey the law and require the government to actually take the trivial steps required to get warrants in a FISA court. You protect the public's rights, protect your own backside, and force government to follow its own rules.
#2: break the law and become criminals. You break the law and encourage the intelligence community to be lazy (get everything and sort through it later).
How is this a difficult choice again? #2 really doesn't help anyone. The only explanation I can think of for companies bending so easily is if they did so in exchange for the government looking the other way about something else. I'd be interested to find out what. :-)
Four words: piercing the corporate veil. Once the company is opened up to civil liability, lawsuits will be filed, and during discovery, those lawsuits will likely uncover information about who knew what and when. At that point, depending on what they turn up, criminal charges might be filed against some of them.
So no need to transport the iceberg, then?
Facebook account: Free.
iPhone: $199
Logging into Facebook from your iPhone instead of the victim's computer so you don't get arrested for burglary: priceless.
There are some things money can't buy. For everything else, there's Mastercard.
It's actually pretty simple to get most of the way there:
With those relatively minor changes, arbitrary code execution bugs (unless you can find a vulnerability in a system service that runs as root) can only succeed in destroying any currently open user file, and can neither permanently inflict harm on the OS or apps, nor persist after the app quits in any significant form. And to the extent that malware could cause harm by tricking the user into installing a system-level plug-in that causes bad behavior, removing such malware would require at most one click in the plug-in manager. At that point, 99.999% of the malware problem goes away. Sadly, these ideas are pretty similar to the ones I came up with when I was still in elementary school. They're beyond obvious....
I can't imagine any store charging for listening to previews. Well, maybe somebody might, but most would probably roll it into the cost of doing business and raise prices as needed. I suppose you'd end up paying for it either way, but I wouldn't expect it to be explicit.
Presumably, if I had songs on iTunes, then yes, I'd get some fraction of a cent or something. ASCAP and BMI don't collect for everybody. You're thinking of Sound Exchange. ASCAP collects only for its members, as do BMI and SESAC. For any composers or publishers who aren't a member of one of those three performing rights societies (or any of the other similar groups in other countries), the person making the recording or airing it or whatever generally has to track down the composer and publisher through some other means.
The net effect of this sort of thing, if passed, would probably end up being (at most) a fraction of a cent per play of such a snippet, which translates to almost nothing for all but the biggest artists, but if it ends up causing previews to be less accessible, would translate to a huge loss for all but the biggest artists (who probably wouldn't see much difference). Thus, I'd expect the biggest artists to want these laws and everybody else to think there's not enough crack in the world. :-)
I like Time Machine because you don't have to think about it much. When you're talking about backing up a non-server machine, being able to pull down a menu and choose "Back Up Now" is a real advantage. Imaging a drive every so often certainly isn't a bad idea, of course, but a disk image is to Time Machine as a hydrogen bomb is to a flyswatter....
No need to drop the music. Just stop making previews available. Then stand back, watch the impact on their sales, and wait for them to pull their heads out of their asses. As an ASCAP member myself, one of these days, I should write a letter or something to tell them to stop being idiots. *sigh* There's never enough time to correct all the idiots in the world. They are simply too great in number and proliferate too quickly.
In this case, for the Mac OS X installation, my answer would be the same as any other user (or at least the client side portion is the same):
Time Machine to an XServe attached to a giant hardware RAID 5 array via fibre channel. In other words, the same way I back up my laptop except with a serious server providing the disk instead of an ABS....
You should be able to back up the office file server to a Mac OS X Server box just as easily as you could back up to a Linux box, but the reverse isn't true. Backing up a Mac OS X installation with resource forks, extended attributes, etc. to a Linux box is nontrivial at best.... Been there, tried that, never got it working reliably.
Oh, yeah, and I've already had to get its transmission rebuilt back at about 85,000 miles. And people wonder why I will no longer buy American cars. Although foreign cars are probably no better at owning up to design flaws, at least they don't seem to have nearly as many serious design flaws....
Yes, they should. And IMHO Ford owes me a couple hundred in parts and about three or four hours of my time paid at their dealer's' obscene $200/hour repair rate for the design flaw in their valve cover that forced me to replace six port seals, I think eight isolator bolts, and some major gasket between the layers of the intake, scrub the intake, flush out the EGR hose, and replace the valve cover. I'm pretty sure I'll die before Ford covers those costs, though. It should have been an emissions recall, but instead, the local Ford dealer (Peninsula Ford of Sunnyvale, bless their thieving, lying souls who claimed that my <100,000 mile car probably needed to be replaced or at least have a new engine after they wasted almost five days and over $1500 in unnecessary and redundant repairs replacing parts that were almost brand new while trying to find a simple low-temperature-only coolant leak caused by a split in a metal line in the TOP section of the intake manifold and who only fixed my car after I tore their service manager a new asshole) said it would cost $1500 for them to put in that couple hundred bucks in parts. I did it myself in an afternoon with no car repair experience whatsoever. It's 40,000 miles later and my car is still working just fine. Go f**k yourself, Peninsula Ford of Sunnyvale.
Then again, the fact that the same Ford dealer did a half-assed copout fix of blowing out the carbon deposits instead of fixing the real problem just outside of warranty when it failed the first time (at which point Ford actually WOULD have been willing to cover the parts cost for the repair, or so I'm told) really takes the cake. After those two service experiences, I've been desperately hoping Ford would decide to drop that dealership in the whole financial mess, as they give Ford a really bad name. But I digress.
The fact that I had the problem in the first place and that Ford left me and tens of thousands of other 1999 Windstar owners footing the entire repair bill for what was clearly and beyond any doubt a design flaw in the engine is a pretty clear indication that the car companies won't ever own up to mistakes unless you can prove that it's a safety problem, and even then, if they calculate that the cost of the lawsuits from wrongful death will cost less than a recall, they still won't fix them. These sorts of corporations make me sick.
Whatever happened to companies standing behind their products? Isn't that the whole benefit of buying a product from a major company instead of just group buy importing it from a fab shop in China somewhere? Who wants to organize the automobile group buy?
Reprogramming the ECU with updated firmware is a good example of vehicle-specific behavior, but I think that it is more than reasonable for independent shops to be able to get that information. In many cases, reprogramming the ECU is required to fully fix certain DTCs because they have both a software and a hardware root cause---particularly when it comes to how certain sensors interact with fuel mixture on aging vehicles and other similar issues.
As someone who has dealt with OBD-II a bit, allow me to shed some light on the subject. The OBD-II standards specify a series of families of error codes. Codes within a certain range are chassis codes, another range for powertrain codes, another range for emissions control crap, etc. The precise details of what non-powertrain codes mean, however, are specific to each vehicle. For example, in most Chevy cars, C1780 means "Loss of Steering Position Signal". In Ford vehicles, it means "Temperature select failure". And IIRC, there are even some variations between specific models, though I don't have time to hunt for specific examples.
And even within the powertrain codes, the root cause can be vehicle-specific. For example, P0171 and P0174 are both codes for engine banks running lean on a '99 Windstar. They usually mean a vacuum leak caused by oil breakdown of port seals coupled with carbon deposits in the intake due to a flaw in the front valve cover. That's something the code number can't tell you.
Finally, some cars have multiple ECUs at multiple addresses. The chassis and brake codes might be in a different ECU, and AFAIK, that info is completely undocumented.
In short, generic scan tools generally give you a reasonable view of the powertrain codes and nothing more. Although it's better than nothing, it isn't a complete picture.
And I'm guessing the fix involved putting the key on a hard, flat surface and hitting it with a hammer....
Or---and here's a brilliant idea---implement this only for commercial vehicles whose drivers already have to log their hours, routes, etc., have no expectation of privacy, and (I think) do disproportionately more damage to the roads that ordinary cars and light trucks per unit of fuel consumed. Let the people with alternative fuel and low fuel vehicles pay less in taxes because they already paid a huge chunk of extra change up front and are helping make our air cleaner. Call it a tax break for environmental conservation if you want to justify it.
I might accept the GPS idea if and only if I could be assured that the devices report only provide an annual report of the number of miles travelled per city or county as a summary. For tax purposes, no further detail is needed or useful, and as such, no further information should be allowable. By limiting it to annual reporting and cumulative mileage, this allows road funds to be distributed more accurately without providing any real info about who went where and when. That is the absolute most information I would be willing to tolerate without filing a lawsuit, however, and only then if they could prove beyond a reasonable doubt (with verifiable, auditable, mathematically proven software and firmware) that the devices *cannot* provide any other information.
This also means that roadside gathering of information is not acceptable because the mere fact that your car sent the data to a particular gathering station provides information that is greater than what is described above---specifically, your presence on that road at that time. In short, it must involve someone explicitly going to a center and getting scanned or plugging a thumb drive into it or unplugging it from the car and plugging it into a computer or... some explicit action by the user to start the data transfer.
There's definitely a full moon joke here somewhere.
Yes. If I cloned my phone and had two of them making calls on an unlimited plan, I would expect to be able to make 2880 minutes. Unless, of course, the plan explicitly forbids SIM cloning.
Well, that's an easy one. Use precisely guided rockets and explosives to carve her face on the surface of the moon....
And I thought for a second you were talking about someone's astronomer aunt.
It's simply not true that a doctor can't do anything for you. We've had antivirals that are highly effective against influenza for a while now. However, it is important to limit the use of these antivirals to the most severe cases because of the risk of emergence of resistant strains. Thus, in general, doctors will not do anything for you unless there is reason to believe that you might not get over it on your own.
Unless the energy density is a huge win, at $2 per watt and lasting only a few months, that's pretty uninteresting, particularly when you consider that silicon is the second most abundant elements in the earth's crust. If it's pricey, there's something very wrong. I can pick up a handful of the dioxide form for free just a few blocks from here. Of course, there are usually cigarette butts and shell fragments in it....
Now if they were replacing germanium, selenium, gallium or indium arsenide, or some of the other rare elements/compounds that can be used in electronics, I could understand, but replacing sand? I just don't get it. It's not the silicon that makes solar panels expensive. It's all the other stuff.
It's an incomplete story. What we need to know to evaluate cost is A. life expectancy, and B. W/m^2. A solar panel that produces the same wattage for a price comparable to some of the higher density solar panels (IIRC) is cool if it lasts at least as long and has similar density. Otherwise, the replacement costs or the shipping costs and installation footprint make it more expensive, respectively.
Agreed. A boss doesn't have to be technical, nor "right" all the time, so long as the boss acknowledges what he or she doesn't know and doesn't try to pretend he/she knows something that he/she doesn't.
It is when a boss thinks he or she knows everything but actually knows nothing that errors are made. A boss can be completely clueless as long as he/she defers to your expertise. The ones we really can't stand are the ones who are clueless but don't know it. They give bad advice that leads their underlings repeatedly down wrong paths, then ding the underlings on salary reviews for listening to them.
Only slightly better are the bosses that let their employees graze and don't give them any guidance about what they are trying to accomplish. Neither type of boss is particularly effective, and both are, sadly, far more common than good bosses.
Only anecdotal evidence from having witnessed doctors prescribing drugs whose side effects are worse than the condition they try (and usually fail) to fix followed by prescribing drugs to try to fix those side effects, eventually leading to drug-induced dementia or other serious side effects. I've seen severe overmedication often enough to be disturbed by the problem, and even CNN has done stories on the problem, mostly as it applies to the elderly and to school-age children. I'm not aware of any widespread studies to quantify the problem, though.