Alternatively couldn't we just put an IRS electric meter on your car and have people pay an electric tax for electricty-as-fuel usage? Since both the electricity and gasoline are already easily metered, and such meters reveal little if anything about your actual driving habits, it seems both straightforward and relatively unobtrusive.
I realize electric meters and gas pumps don't cover every possible form of fuel for a vehicle, but they cover the two types currently in use, and those basic types are unlikely to change more rapidly than we could reasonable keep up with in legislation.
This is what we do for sales tax, and it seems to work just fine -- there are many cases where sales tax might not apply, and it's up to the purchaser to have the appropriate paperwork to avoid being charged. There are methods both to certify and exemption in an attempt to avoid being charged, and to obtain a refund in the case that tax was improperly collected.
It's also more or less the same system we have for income tax; the IRS assumes that all of your income is subject to taxation unless you provide specific evidence to the contrary.
I'm not saying it would be fun, but it's a pretty standard way to collect taxes -- assume everything is taxable and let individual taxpayers try to justify the reason that some of their activities should not be taxable.
Couldn't we just distribute an open-source virus to Windows users who are RARing things in the first place? I have trouble coming up with any good reason to use RAR anymore, and I know from experience that a high proportion of usage is totally inappropriate (segmenting torrent files, for example).
I don't understand how SMS is different than email, other than mobile carriers are big fans of non-interoperable systems because such systems fit their business model. Yes, it's length limited, but it's not like your email client would stop working if you only wrote short messages -- if you and your friend were both sending email from a phone presumably you'd send the same sort of short messages you would if you were using SMS.
But I'm sure I'm just a crusty old man who doesn't understand the appeal of using monitored, archived, unencrypted messages that can only be sent to other mobile handsets versus sending the same message over a secure(able) channel that can connect to the same mobile handsets as well as any computer.
But you can't use patented methods even in free (in either sense) software either, so I don't see how making it hard to make for-profit software has anything to do creating more not-for-profit software.
He compared a fine against a US company to EVERY OTHER FINE EVER LEVIED. How is that not presenting both sides? Or are you suggesting that the EU has only ever fined US companies?
And that same hospital will take 30% off the top of your bill if you ask and offer to pay them on net-30 terms. You're free to negotiate the same rates as insurance companies -- you have less buying power, so you probably won't get quite as good a rate (just like processors are cheaper by the 1000), but there's no rule that says you have to pay the listed rack rate just because you're paying cash.
Presumable the person who built a linux-controlled/monitored sump pump did benefit from it. Just because you don't have an application for it doesn't mean that none exists. And if you've ever had a failed sump pump, or more incoming water than the sump could drain, you might understand the benefit of having a "smart" pump that could alert you to failures.
At the very least there's the benefit of having done something he enjoyed and produced a demonstrable product -- it's not any different than someone who likes woodworking and builds birdhouses/furniture/whatever that they don't strictly need.
They're not assuming that *life* is earth-like, they're assuming that *chemistry* works in other places like it does in the bits of the universe we've observed so far -- water is fairly common as non-elemental substances goes, and liquid water very conducive to chemistry, even if that chemistry doesn't involve carbon-based lifeforms. As my chemistry prof used to say anytime someone asked for help -- "draw a beaker and put some water in it" -- because the reaction you're trying to model probably happens in liquid water.
Now it's plausible that life based on non-chemical reactions exists -- a nuclear-powered martian would be kind of cool -- but it's very unlikely that we'd recognize such life even if it where here on Earth, let alone if it were 25 million miles from here, so there's not much point in looking for it.
Trees are mostly water -- it's just the outside that's dry. The entire nutrient transport system is totally dependent on water, not to mention the photosynthesis thing.
And lichen does very, very dormant when it's dry, undergoing a fairly significant transformation when it gets wet again; left dry for sufficiently long it will also die, as the algae component again needs liquid water for photosynthesis.
That's not to say you necessary can't have life without liquids, but on Earth there are not a lot of examples.
If you're going to FedEx drives to Amazon, couldn't you just FedEx the Backblaze around instead?
I'm also wondering how S3 gets you data at geographically diverse locations any faster than overnight shipping a box full of disks -- if you're moving large amounts of data from the US to Russia FedEx is probably the fastest, cheapest solution no matter how you host the data once it gets there.
What failure rate are you using to "virtually guarantee" that you'll get data corruption with 45 drives? What failure rate in your RAM, CPU, and motherboard are you using to guarantee that the ZFS checksum are not themselves corrupted? Not to mention the high possibility of bugs in a younger file system, and the different performance characteristics among FSes.
I'm not say ZFS is a bad plan, at least if you're running enough spindles, but if you're going to "virtually guarantee" silent corruption with less than 100 drives I'd like to see some documentation for the the non-detectable failure rates you're expecting.
It's also worth noting that in a lot of data, a small amount of bit-flips might not be worth protecting against at all. Or they might be better protected at the application level instead of the block level -- for example, if the data will be transmitted to another system before it is consumed, as would be typical for a disk-host like this, a single checksum of the entire file (think md5sum) could be computed at the end-use system, rather than computing a per-block checksum at the disk host and then just assuming the file makes it across the network and through the other system's I/O stack without error.
You probably can't write an employment contract that requires me to keep working after you've stopped paying me. Even if you could, a violation of such a contract would be purely a civil matter, unless there was some actual crime underlying the breach.
You're still assuming that "killing the host" happens some non-trivial amount of time after "infecting the host" -- at some point the host won't be able to travel far enough to infect anyone who wasn't also at risk from the original point of infection. Given modern travel speeds that threshold is pretty small, but it certainly still exists; a virus that killed people within 1 minute of infection would never make it out of the building where it was first encountered.
I've never tried the "magic update" process, but with a FireWire or Ethernet cable and a few clicks to turn on file sharing he could pretty easily copy his entire home folder with one drag-and-drop (or my personal recommendation, by creating a disk image). If the old machine can be powered down a target-disk-mode boot would likely be even faster and similarly easy.
I mean, it would be nice if "magic update" worked as described, but it's not like Apple makes it hard to copy files, or hides your personal data in 1000 different folders.
Your smart card has a unique key, and the per-channel decryption keys sent to your device are encrypted against that key. If you could duplicate that key you could decrypt the per-channel keys on any device, though you'd still need to have a valid subscription for the original card; you could share a subscription, but not avoid the bill altogether. But it's not trivial to to pull private keys out of the smart card -- it can usually be done, but it's not easy -- that's why they use the smart card in the first place.
The per-channel authorization keys expire after some days (I'm thinking 14 days, but it's been a while since I checked); if you leave your receiver disconnected for that long it cannot tune again until it is re-authorized. There may be some more rapidly changing key in place for actual media streams, but it would have to be something you can decrypt with the multi-day authorization key -- disconnecting your receiver for a few minutes/hours does not prevent you from tuning when reconnected.
In short, it's not a simple key shared among all devices, nor is it a system where cracking a single key grants you access to all future keys; it's probably feasible to crack, but not cheap and easy, and therefore it's indifferent from impossible unless the data itself is very valuable.
There are hardware hacks that will output the decrypted but still compressed video stream from DirecTV and DishNetwork systems via FireWire, at least for their MPEG-2 streams. Last time I checked their newer streams were still unavailable, but things could have changed since then.
9 days is a pretty old figure. It's still more than 2 minutes, but it's not very long if you're willing to buy the hardware: http://www.copacobana.org/
They don't. They just send threatening letters to everyone who didn't pay the TV tax under the assumption that essentially everyone owns at least one TV.
I don't see why watching a show live means I like it more than watching it delayed, or how a week delay reduces the effectiveness of (most) advertisements. Frankly I'd only watch a show in realtime if I *didn't* like it, because without the ability to rewind and pause I'm likely to miss at least some of the show.
And while streaming flash video is a handy alternative it's not at all like having a real recording. For one thing it requires using your computer, which in most cases is not hooked up to the same display you would otherwise use for the show. For another the video quality is terrible compared to what you get OTA or via cable. Finally the UI controls and seek capabilities are usually poor and impossible to integrate with a limited-interface control (like an IR remote).
NAT provides exactly the same security as a connection-tracking firewall -- there is no further benefit to address translation over a dynamic firewall with the same rules. Dropping the NAT part makes it about 11,000 times easier to run services on the inside, particularly if they use multiple connections (e.g. FTP, SIP) in the course of a session, and it removes the "only 1 person can run a service on the default port" limitation introduced when you put more than one system behind a single address.
I don't believe it. What exactly does the telescope "focus" in the laser output? Focusing uncollimated light might make it more useful -- the sun through a magnifying glass -- but it's not such a great plan for lasers. It also greatly increases operational complexity, since focused light is only focused at a certain distance from the source, so you'd have to track the distance to the plan and continually re-focus to keep on-target.
Owning a TV is no invasion of privacy. Even using it isn't. But I suppose that isn't relevant to your holier-than-thou rant.
Putting other people in charge of your media distribution may well be a threat to your privacy, but that's hardly unavoidable nor is it unique to TV -- brick and mortar and online media outlets have been collecting whatever customer information they can for years, from credit card info, points/loyalty cards, etc. Unless you're buying anonymously with cash, or only consuming media from an institution that values privacy (like a library) there's a good chance someone is already tracking your media consumption habits, be they video, book, or otherwise.
A book of random numbers is great for statistics. If that's your use there's no need to do anything else, and RAND's book is still a good choice.
But the value of random numbers in things like cryptography is that they are unpredictable. If everyone is using the same list the numbers are entirely predictable and therefore useless. A typical example is the hybrid cryptosystem used in public-key encryption -- the computer picks a random number for use as the secret key for the shared-secret cypher, encrypts the data with that key, then encrypts the key itself using the much slower and more restrictive public-key algorithm. If you could guess the random number used as the key in step 1 you wouldn't need to break they encryption in stages 2 or 3 to read the message because you'd already have the password.
Slashdot Mirror
Alternatively couldn't we just put an IRS electric meter on your car and have people pay an electric tax for electricty-as-fuel usage? Since both the electricity and gasoline are already easily metered, and such meters reveal little if anything about your actual driving habits, it seems both straightforward and relatively unobtrusive.
I realize electric meters and gas pumps don't cover every possible form of fuel for a vehicle, but they cover the two types currently in use, and those basic types are unlikely to change more rapidly than we could reasonable keep up with in legislation.
This is what we do for sales tax, and it seems to work just fine -- there are many cases where sales tax might not apply, and it's up to the purchaser to have the appropriate paperwork to avoid being charged. There are methods both to certify and exemption in an attempt to avoid being charged, and to obtain a refund in the case that tax was improperly collected.
It's also more or less the same system we have for income tax; the IRS assumes that all of your income is subject to taxation unless you provide specific evidence to the contrary.
I'm not saying it would be fun, but it's a pretty standard way to collect taxes -- assume everything is taxable and let individual taxpayers try to justify the reason that some of their activities should not be taxable.
Couldn't we just distribute an open-source virus to Windows users who are RARing things in the first place? I have trouble coming up with any good reason to use RAR anymore, and I know from experience that a high proportion of usage is totally inappropriate (segmenting torrent files, for example).
I don't understand how SMS is different than email, other than mobile carriers are big fans of non-interoperable systems because such systems fit their business model. Yes, it's length limited, but it's not like your email client would stop working if you only wrote short messages -- if you and your friend were both sending email from a phone presumably you'd send the same sort of short messages you would if you were using SMS.
But I'm sure I'm just a crusty old man who doesn't understand the appeal of using monitored, archived, unencrypted messages that can only be sent to other mobile handsets versus sending the same message over a secure(able) channel that can connect to the same mobile handsets as well as any computer.
But you can't use patented methods even in free (in either sense) software either, so I don't see how making it hard to make for-profit software has anything to do creating more not-for-profit software.
He compared a fine against a US company to EVERY OTHER FINE EVER LEVIED. How is that not presenting both sides? Or are you suggesting that the EU has only ever fined US companies?
And that same hospital will take 30% off the top of your bill if you ask and offer to pay them on net-30 terms. You're free to negotiate the same rates as insurance companies -- you have less buying power, so you probably won't get quite as good a rate (just like processors are cheaper by the 1000), but there's no rule that says you have to pay the listed rack rate just because you're paying cash.
Presumable the person who built a linux-controlled/monitored sump pump did benefit from it. Just because you don't have an application for it doesn't mean that none exists. And if you've ever had a failed sump pump, or more incoming water than the sump could drain, you might understand the benefit of having a "smart" pump that could alert you to failures.
At the very least there's the benefit of having done something he enjoyed and produced a demonstrable product -- it's not any different than someone who likes woodworking and builds birdhouses/furniture/whatever that they don't strictly need.
They're not assuming that *life* is earth-like, they're assuming that *chemistry* works in other places like it does in the bits of the universe we've observed so far -- water is fairly common as non-elemental substances goes, and liquid water very conducive to chemistry, even if that chemistry doesn't involve carbon-based lifeforms. As my chemistry prof used to say anytime someone asked for help -- "draw a beaker and put some water in it" -- because the reaction you're trying to model probably happens in liquid water.
Now it's plausible that life based on non-chemical reactions exists -- a nuclear-powered martian would be kind of cool -- but it's very unlikely that we'd recognize such life even if it where here on Earth, let alone if it were 25 million miles from here, so there's not much point in looking for it.
Trees are mostly water -- it's just the outside that's dry. The entire nutrient transport system is totally dependent on water, not to mention the photosynthesis thing.
And lichen does very, very dormant when it's dry, undergoing a fairly significant transformation when it gets wet again; left dry for sufficiently long it will also die, as the algae component again needs liquid water for photosynthesis.
That's not to say you necessary can't have life without liquids, but on Earth there are not a lot of examples.
If you're going to FedEx drives to Amazon, couldn't you just FedEx the Backblaze around instead?
I'm also wondering how S3 gets you data at geographically diverse locations any faster than overnight shipping a box full of disks -- if you're moving large amounts of data from the US to Russia FedEx is probably the fastest, cheapest solution no matter how you host the data once it gets there.
What failure rate are you using to "virtually guarantee" that you'll get data corruption with 45 drives?
What failure rate in your RAM, CPU, and motherboard are you using to guarantee that the ZFS checksum are not themselves corrupted? Not to mention the high possibility of bugs in a younger file system, and the different performance characteristics among FSes.
I'm not say ZFS is a bad plan, at least if you're running enough spindles, but if you're going to "virtually guarantee" silent corruption with less than 100 drives I'd like to see some documentation for the the non-detectable failure rates you're expecting.
It's also worth noting that in a lot of data, a small amount of bit-flips might not be worth protecting against at all. Or they might be better protected at the application level instead of the block level -- for example, if the data will be transmitted to another system before it is consumed, as would be typical for a disk-host like this, a single checksum of the entire file (think md5sum) could be computed at the end-use system, rather than computing a per-block checksum at the disk host and then just assuming the file makes it across the network and through the other system's I/O stack without error.
You probably can't write an employment contract that requires me to keep working after you've stopped paying me. Even if you could, a violation of such a contract would be purely a civil matter, unless there was some actual crime underlying the breach.
You're still assuming that "killing the host" happens some non-trivial amount of time after "infecting the host" -- at some point the host won't be able to travel far enough to infect anyone who wasn't also at risk from the original point of infection. Given modern travel speeds that threshold is pretty small, but it certainly still exists; a virus that killed people within 1 minute of infection would never make it out of the building where it was first encountered.
I've never tried the "magic update" process, but with a FireWire or Ethernet cable and a few clicks to turn on file sharing he could pretty easily copy his entire home folder with one drag-and-drop (or my personal recommendation, by creating a disk image). If the old machine can be powered down a target-disk-mode boot would likely be even faster and similarly easy.
I mean, it would be nice if "magic update" worked as described, but it's not like Apple makes it hard to copy files, or hides your personal data in 1000 different folders.
Your smart card has a unique key, and the per-channel decryption keys sent to your device are encrypted against that key. If you could duplicate that key you could decrypt the per-channel keys on any device, though you'd still need to have a valid subscription for the original card; you could share a subscription, but not avoid the bill altogether. But it's not trivial to to pull private keys out of the smart card -- it can usually be done, but it's not easy -- that's why they use the smart card in the first place.
The per-channel authorization keys expire after some days (I'm thinking 14 days, but it's been a while since I checked); if you leave your receiver disconnected for that long it cannot tune again until it is re-authorized. There may be some more rapidly changing key in place for actual media streams, but it would have to be something you can decrypt with the multi-day authorization key -- disconnecting your receiver for a few minutes/hours does not prevent you from tuning when reconnected.
In short, it's not a simple key shared among all devices, nor is it a system where cracking a single key grants you access to all future keys; it's probably feasible to crack, but not cheap and easy, and therefore it's indifferent from impossible unless the data itself is very valuable.
There are hardware hacks that will output the decrypted but still compressed video stream from DirecTV and DishNetwork systems via FireWire, at least for their MPEG-2 streams. Last time I checked their newer streams were still unavailable, but things could have changed since then.
9 days is a pretty old figure. It's still more than 2 minutes, but it's not very long if you're willing to buy the hardware:
http://www.copacobana.org/
They don't. They just send threatening letters to everyone who didn't pay the TV tax under the assumption that essentially everyone owns at least one TV.
I don't see why watching a show live means I like it more than watching it delayed, or how a week delay reduces the effectiveness of (most) advertisements. Frankly I'd only watch a show in realtime if I *didn't* like it, because without the ability to rewind and pause I'm likely to miss at least some of the show.
And while streaming flash video is a handy alternative it's not at all like having a real recording. For one thing it requires using your computer, which in most cases is not hooked up to the same display you would otherwise use for the show. For another the video quality is terrible compared to what you get OTA or via cable. Finally the UI controls and seek capabilities are usually poor and impossible to integrate with a limited-interface control (like an IR remote).
NAT provides exactly the same security as a connection-tracking firewall -- there is no further benefit to address translation over a dynamic firewall with the same rules. Dropping the NAT part makes it about 11,000 times easier to run services on the inside, particularly if they use multiple connections (e.g. FTP, SIP) in the course of a session, and it removes the "only 1 person can run a service on the default port" limitation introduced when you put more than one system behind a single address.
I don't know about you, but I fully expect to be able to speak my mind even if it does involve national security.
I don't believe it. What exactly does the telescope "focus" in the laser output? Focusing uncollimated light might make it more useful -- the sun through a magnifying glass -- but it's not such a great plan for lasers. It also greatly increases operational complexity, since focused light is only focused at a certain distance from the source, so you'd have to track the distance to the plan and continually re-focus to keep on-target.
But I think you'd have a hard time raping someone through an online social network.
Owning a TV is no invasion of privacy. Even using it isn't. But I suppose that isn't relevant to your holier-than-thou rant.
Putting other people in charge of your media distribution may well be a threat to your privacy, but that's hardly unavoidable nor is it unique to TV -- brick and mortar and online media outlets have been collecting whatever customer information they can for years, from credit card info, points/loyalty cards, etc. Unless you're buying anonymously with cash, or only consuming media from an institution that values privacy (like a library) there's a good chance someone is already tracking your media consumption habits, be they video, book, or otherwise.
A book of random numbers is great for statistics. If that's your use there's no need to do anything else, and RAND's book is still a good choice.
But the value of random numbers in things like cryptography is that they are unpredictable. If everyone is using the same list the numbers are entirely predictable and therefore useless. A typical example is the hybrid cryptosystem used in public-key encryption -- the computer picks a random number for use as the secret key for the shared-secret cypher, encrypts the data with that key, then encrypts the key itself using the much slower and more restrictive public-key algorithm. If you could guess the random number used as the key in step 1 you wouldn't need to break they encryption in stages 2 or 3 to read the message because you'd already have the password.