Actually, you might be surprised at how large it would be. F=It/V. 10mS write time for a sector, 2S for a file.
You don't need enough power to finish the OS-level task...you only need enough to write out the data in the drive's RAM cache. Since that is 256MB or less on most current SSD drives (512MB is found on some drives greater than 500GB), it's not as much as you estimate.
Then, too, when you use the correct timings (10ms for a sector is about 200 times too long), you see that even the slowest SSD takes only about 5ms to write out 1MB (with an average around 3ms), that's around 0.75 seconds to flush the whole cache, resulting in about 1/3 the power you estimate. And, that's assuming that the whole cache needs to be flushed, as it's possible that only a few blocks need to be written.
You can also estimate the time in another way, in that these drives can sustain 300-500MB/sec write rates. That means that copying from the drive's RAM cache to flash must be at least that fast, which gives between 0.5 and 1.7 seconds to flush. This time is similar to my estimate above, so I suspect that 0.75 seconds isn't far off from correct.
For example, assuming browsers incorporate the capability to decompress it, lowering the bandwidth of Youtube by ~3% is an achievement.
I don't know why people keep mentioning Youtube, since all videos are already compressed in such a way that pretty much no external compression is going to gain anything.
Although when compressing a video Zopfli might result in a smaller file compared to gzip, that doesn't mean either will be smaller than the original. All H.264 files should be using CABAC after the motion, macroblock, psychovisual, DCT, etc. stages, and that pretty much means that the resulting files have as much entropy per bit as possible. At that point, nothing can compress them further.
And secondly, they require the popup to be signed so you cannot say "I didn't know" further down the line.
I'm not sure how you "sign" a popup, but what happens if someone other than the account holder does whatever it takes to dismiss the popup? It could be someone else in the house or it could be someone leeching off your WiFi (either open or not). Or, what if you are just typing along in a web page when this popup displays, and the next key you happen to hit is "Enter" and away goes the popup?
And what if I challenge the ISP to show me proof for the notice.
There's no way they can, as it's obvious that there are too many ways for the popup to be gone without the account holder knowing it ever was there.
All I could do to prove it was a mistake is all that anyone who is falsely accused of doing something wrong can do to prove their innocence, which is to tell the truth.
Unfortunately, you're being too logical. Everything you say would work really well if you were allowed to somehow challenge the accuracy of the accusation. But, you aren't.
And, here's where it gets really fun. I was under the impression (as I think we all were) that you could ask for a review on any "alert". Not so with Verizon:
Our Copyright Alert Program allows you to seek review by an independent reviewer, the American Arbitration Association (AAA), of the alerts you have received. The opportunity to seek review is available once you have received a fifth or sixth copyright alert from us.
This means that if something flags your account (maybe the DHCP logs are wrong, maybe somebody is spoofing other IPs to the tracker just to get other people in trouble, etc.), you won't even be able to ask for a review until it's already at the point that they will be intercepting your browser requests to "educate" you.
For several years, copyright owners have been using certain automated techniques to identify Internet users who they believe are engaged in possible infringing activity via P2P networks. When a copyright owner identifies an instance of possible infringement, it sends the user's ISP a notice that contains the name of the copyrighted work, the date and time of the alleged infringement and the Internet Protocol (IP) address of the computer that is alleged to be sharing copyrighted content. The ISP matches the IP address sent by the copyright owner to the specific customer to whom that IP address was assigned at that time, and then forwards the information provided by the content owner to that user without identifying the user to the copyright owner(s).
The evidence will include your account, IP, their DHCP logs showing you were in control of the IP at that time, and statistics on how much uploading and downloading of the torrent you performed.
No, it won't:
The Alerts, include the date, time, time zone and title of the copyrighted content
You don't blow up a satellite, you simply push it out of orbit far enough that it can't correct back. Absolute destruction isn't the only option, it's just the one that Hollywood things of most often.
Although total destruction is far overkill, a "mission kill" of a satellite is most easily achieved by hitting it with something that will very likely break pieces off. That might also push it out of orbit, but you really are trying to put holes in the sensitive parts, not just move it around.
E.g. ever wonder how long ago slapping a woman was acceptable as a movie scene? Do you remember "The Man with the Golden Gun" (1974)? What about the "Airplane!" (1980)?)
Even today, slapping a woman is quite acceptable when done for comedic effect, as in Airplane!, so I don't think that's a good example.
A couple minutes is enough to confirm there's some copyrighted content
Not for many formats, as there won't be anything usable unless you download a very large chunk.
no matter that the other 90% of the torrent might theoretically be something other than the rest of that GoT episode. Then you use a custom client that requests just those chunks known to contain copyrighted material.
If that's the method they use, then I can argue "fair use", and they won't be able to refute it since they never downloaded the entire torrent, nor will they know if I did.
Every peer who hands you one of those chunks is sharing copyrighted content.
Even better. If they accuse based on one chunk of a torrent, then then only have about 0.1% of the total, and that's certainly fair use, as it's only around 3 seconds of an hour long show.
The key here is that once somebody ponies up the $35 and raises these sorts of challenges ("show me the bits you downloaded from me, personally", etc.), if those challenges work, then it will be posted somewhere and everybody will use the same argument. The **AA won't be able to keep paying for the reviews, and will have to do some actual work to proves infringement. Either way, it will cost them more money. The more likely alternative is that all **AA "evidence" is assumed to be 100% valid regardless of method and the **AA will win all "reviews". Once this comes to light, then I suspect that class-action lawsuits will start.
First of all, I don't have an open wifi... and secondly, I don't infringe on copyright, so how is it that you propose that software which allegedly doesn't make mistakes would link my IP to an infringement, exactly?
No, I said "the monitoring software is assumed to never make mistakes". In other words, everybody who is ruling on your guilt is under the assumption that if you got an alert, it's not a mistake. In your case, it's probably a mistake (although even non-open WiFi can be hacked, and if your account was used, then it is not a mistake by the definition of this program), but your ISP and the "independant review board" will assume that it is not.
FYI: "Guilty till proven innocent" is a concept of Criminal Case Law
Not in the US, it isn't.
and has no role in Civil Litigation, where you must only show "probable cause".
Civil lawsuit outcomes are based on "preponderance of evidence"...in layman's terms, who is "most right", not "probable cause". Regardless of how the determination of liability at the end of the trial is done, you can still sue anybody for anything with no real evidence that they did anything wrong. With far less than "probable cause", civil litigation can reach the discovery stage, at which point you better have deep pockets if you want to continue a defense.
The major difference in a lawsuit and this system is that evidence is not assumed to be valid in a lawsuit...it must be proven. For example, if the plaintiff claims to have evidence of infringement on a computer, the defendant would be able to get to see that computer (usually via a neutral third party) to verify that the alleged evidence does exist and was collected in a proper manner. This discovery might allow the defendant to show that the evidence is faulty in some way.
All this is absent from the "six strikes" system, which assumes that all evidence against the alleged infringer is valid with no possibility for a mistake. Because of this, the term "guilty until proven innocent" applies. Unfortunately, there is really no way to prove innocence once you start with the assumption that all evidence indicating guilt is valid and not subject to rebuttal.
If it's that easy to neutralize drones (making them head back to their designated landing zone, if your human-piloted fighters don't take them out while they're flying autonomously), then they're really quite useless against a technologically advanced opponent.
You can use very similar tactics against planes with pilots, too. If you cut off a pilot from the AWACS data, or communications from front-line forces, a lot of missions would have to be scrubbed.
The only real difference comes when you are repeatedly attacking the same area, and then the pilot could have learned enough to not need much electronic support.
.... that the $35 fee is refunded to consumers who pay it, and are not found to have actually committed any copyright infringement.
OK, you've just gotten an "alert". Prove that it was not issued correctly. Remember that the monitoring software is assumed to never make mistakes, so if they have a list with your IP address, a date and time, and a torrent name, then it that is proof that that IP address downloaded and uploaded that torrent at that date and time. The ISP has furnished a DHCP log that shows your account was holding that IP address at that date and time.
Note that you, personally, are not being accused. Your account is being accused. So, now you have to provide proof that your account is not to blame for something that did happen (according to the "guilty until proven innocent" system in place). You can play the "open WiFi" card once. If you get another "alert" after using the "open WiFi" defense, you can't use it again.
Once they see a title being made available on the peer-to-peer network, they confirm that it is, in fact, copyrighted content.
Riiiight... 'cuz, we all know, ISPs and the MAFIAA are certainly trustworthy entities, who would never misuse people's personal information, or god forbid, lie to support their goals.
Can you imagine how much bandwidth and disk space such a content checking system must have to "confirm copyrighted content"?
A search for "Game of Thrones" on Pirate Bay lists 5 pages (at 30 entries per page) before the torrent size is less than 5GB, and 22 pages before the size is less than 0.5GB. That's over 1TB they would need to download just to verify every torrent for a single title. If they don't download the entire torrent, they can't be sure that it's really is something they hold the copyright to.
With 5 million torrents on TPB, even if only half (likely too low a guess) are copyrighted by companies that belong to **AA, and if the average size is 250MB, that's 625TB of data.
The *AA are paying DtecNet to send the strikes in the first place, and if you appeal it will cost them more than $35 to put together the paperwork to argue with you.
I sincerely doubt that printing out an IP address, date and time, and torrent name will cost $35. This is all the "evidence" they have, and it's unknown whether you will get to see it unless you ask for a review. The "alert" might be nothing more than "you've been a bad boy" with no details.
It's not necessarily a waste of $35, since if they are unable to verify the alleged infringement, you get the $35 refunded, and the "strike" is removed.
Just how many claims do you think will be overturned by a group of people that were picked by the same people who made the claim?
In particular, when the entire evidence of infringement for these claims consist of an IP address, a date and time, and the name of a torrent, exactly how are you going to prove that you didn't infringe? When the party making the claim of infringement doesn't even have to prove that the named torrent was their copyrighted content (much less any of the other things needed for infringement, like proving uploading), how can you possibly defend yourself? Even if you aren't running any bittorrent software at all, how can you prove that? And, if you are running bittorrent software, how can you prove that you were not part of that torrent at that time?
This is one of the many problems with "guilty until proven innocent"...often you have to prove a negative, and sometimes that can't be done.
Hollywood *rarely* gets technical issues right, but the speech in Top Gun where they were talking about pilots becoming reliant on missiles in Korea was actually true, and the basic principle should still be true today. Dogfighting specifically doesn't really happen any more, but the basic evasive agility skills that it's based on are still applicable.
It's definitely true that US pilots had been losing such skills over time and something needed to be done, but it's somewhat ironic that the platform with pretty much the ultimate "fire and forget" missle was used in the Navy training.
Your global satellite system is no match for China's anti-satellite missiles.
Destruction of a satellite isn't going to happen unless it's an all-out shooting war, as the chance for collateral damage to a "friendly" satellite is pretty high. Although there might not be immediate damage, spreading debris in the same orbit height as your own satellites is not good for long-term stability.
This drone stuff only works because all the opponents are decades behind the US technologically, so they have little ability (yet) to block the radio signals needed to keep these aircraft under control.
If all you want to do is cut off the signal to/from the drone, it's pretty easy as long as you have a general idea of the frequency used and have a directional antenna and a big enough transmitter (and can track the drone well enough to keep your antenna pointed at it).
This could protect you from drone strikes, but the failsafe systems in the drone would mean that it would just turn and head for the designated landing zone.
Blocking GPS signals is something well beyond the capabilities of some Taliban fighters living in a cave and carrying nothing more advanced than AK47s, however it's well within China's abilities.
Again, this is actually pretty easy to do, but gains you nothing, as the drones also have intertial navigation systems which allow them to at least head for the failsafe landing zone. Spoofing GPS signals, on the other hand, is very hard.
I can get the UAC for network connection though. If you download a office suite say and it tries to connect to the internet you might be suspicious.
Since there is now a lot of legitimate software that requires a network connection to "phone home", it just gets to the point that a user will blindly click the "make this dialog go away" button without reading. In addition, all the dialog gives you is the program name, and malware can have names that seem legitimate, while I sometimes have to google an EXE or DLL to see if it is OK or not.
Also, Internet Explorer is whitelisted, so if the malware creates an IE instance (which doesn't require a visible window), then the user will never know. Between being annoyed by UAC for permissions to open the firewall plus the fact that the firewall isn't really useful against much of what a user should be concerned about (things like a keylogger phoning home where a POST to some random control server somewhere probably wouldn't raise a red flag in IE) makes UAC just more security theater...it's annoying, doesn't really protect you, but allows Microsoft to say "we did something".
Although I suggested this, too, I now realize it has the same problem as plus addressing, in that either the initial sign-up or the unsubscribe might use a broken test when trying to determine if the e-mail address is legitimate.
I have personally dealt with co-workers who assumed that the domain name must be of the format "word.exactly3characters". New top-level domains have at least killed the thought that all TLDs are 3 characters long, but the "only one period in the domain name" belief is still there for some people.
And change site name.mydomain to their mail servers! : D
I find your ideas intriguing, and wish to subscribe to your newsletter.
The problem is for 90% of users a UNC prompt or its mac/linux equivalent pops up and they click ok. To most users the fingers go in the ears as soon as you try to explain the risks and what is happening and they just ask "So what do I need to click to continue?"
If common tasks didn't require the user to answer these pop-ups, then they would see them as "unusual" and wouldn't be as likely to just "click to continue".
There are quite a few control panel settings in Windows 7 that require answering a UAC prompt just to see the settings. Any software that tries to make a network connection and isn't on the Windows firewall "approved" list generates a UAC prompt. Then there are some settings (like "Adjust Visual Effects" in the "Perfomance Information and Tools" control panel page) that shouldn't require UAC at all, as they are merely personal preferences that are no different from desktop wallpaper as far as security or system stability is concerned.
The FCC ruled that Verizon's blocking tethering without paying extra is illegal.
They don't block general tethering...they only block using your phone as a WiFi hotspot, since that allows multiple other devices to use the connection at the same time. This is what the watchdog app is for.
I can one-to-one tether with no problem, and I have a grandfathered unlimited data plan.
How does that work when you send e-mail from half a dozen different systems, including Outlook, pine, Android mail, sendmail, and in a pinch, even telnet to port 25 or openssl to port 465/587?
These are one-off addresses tied to each company and are used for preventing spam to a personal e-mail address, and most of them aren't ever used to send e-mail. The few times you need to, it's also usually not critical that the e-mail be answered right now (unlike a business e-mail), so you can wait a bit until you are at one of your machines with the correct software (because you aren't going to be doing this from random machines, ever, as it's still your personal e-mail).
If you are really desperate, though, you just run some remote access software to get to a machine with the right software.
I (not the submittor) frequently use <myname>+<site name>@<mydomain>.
One of the issues with this is that <myname>@<mydomain> will be delivered, too. And, if that's your "real" e-mail address, then it's now out there for spammers to hit.
If you instead use something that doesn't rely on special address parsing (like <myname><site name>@<mydomain> or <myname>@<site name>.<mydomain>), you can just ditch the e-mail address once it is compromised. There are a couple of companies that I had to do this to simply because their "you've done business with us, which we consider an opt-in" mailing list has no reasonable way to unsubscribe.
You can't legally charge anyone to listen to your CD
Maybe you can't charge people to listen to it, but you can legally charge people to rent the CD from you. Once those people rent it, they can choose to listen to it if they want. The "license" printed on most media these days that implies you cannot rent it is not legally binding.
and you certainly can't play it for a crowd of strangers even if they don't pay you a dime.
I wish you'd tell that to all the idiots who drive by my house with their stereo cranked to "any occupants of this car will be deaf by next year". Also, if I grab random people off the street and ask them if the want to come to my house to listen to the latest insert artist name here CD, it's perfectly legal. Even though once I get enough people, they are a crowd, and they are strangers, it's not a public performance.
Slashdot Mirror
Actually, you might be surprised at how large it would be. F=It/V. 10mS write time for a sector, 2S for a file.
You don't need enough power to finish the OS-level task...you only need enough to write out the data in the drive's RAM cache. Since that is 256MB or less on most current SSD drives (512MB is found on some drives greater than 500GB), it's not as much as you estimate.
Then, too, when you use the correct timings (10ms for a sector is about 200 times too long), you see that even the slowest SSD takes only about 5ms to write out 1MB (with an average around 3ms), that's around 0.75 seconds to flush the whole cache, resulting in about 1/3 the power you estimate. And, that's assuming that the whole cache needs to be flushed, as it's possible that only a few blocks need to be written.
You can also estimate the time in another way, in that these drives can sustain 300-500MB/sec write rates. That means that copying from the drive's RAM cache to flash must be at least that fast, which gives between 0.5 and 1.7 seconds to flush. This time is similar to my estimate above, so I suspect that 0.75 seconds isn't far off from correct.
For example, assuming browsers incorporate the capability to decompress it, lowering the bandwidth of Youtube by ~3% is an achievement.
I don't know why people keep mentioning Youtube, since all videos are already compressed in such a way that pretty much no external compression is going to gain anything.
Although when compressing a video Zopfli might result in a smaller file compared to gzip, that doesn't mean either will be smaller than the original. All H.264 files should be using CABAC after the motion, macroblock, psychovisual, DCT, etc. stages, and that pretty much means that the resulting files have as much entropy per bit as possible. At that point, nothing can compress them further.
And secondly, they require the popup to be signed so you cannot say "I didn't know" further down the line.
I'm not sure how you "sign" a popup, but what happens if someone other than the account holder does whatever it takes to dismiss the popup? It could be someone else in the house or it could be someone leeching off your WiFi (either open or not). Or, what if you are just typing along in a web page when this popup displays, and the next key you happen to hit is "Enter" and away goes the popup?
And what if I challenge the ISP to show me proof for the notice.
There's no way they can, as it's obvious that there are too many ways for the popup to be gone without the account holder knowing it ever was there.
All I could do to prove it was a mistake is all that anyone who is falsely accused of doing something wrong can do to prove their innocence, which is to tell the truth.
Unfortunately, you're being too logical. Everything you say would work really well if you were allowed to somehow challenge the accuracy of the accusation. But, you aren't.
And, here's where it gets really fun. I was under the impression (as I think we all were) that you could ask for a review on any "alert". Not so with Verizon:
This means that if something flags your account (maybe the DHCP logs are wrong, maybe somebody is spoofing other IPs to the tracker just to get other people in trouble, etc.), you won't even be able to ask for a review until it's already at the point that they will be intercepting your browser requests to "educate" you.
Remember this is done by the ISP and so will carry more authority than third-party torrent tracking.
No, it's not. It's being done using the same third party tracking:
The evidence will include your account, IP, their DHCP logs showing you were in control of the IP at that time, and statistics on how much uploading and downloading of the torrent you performed.
No, it won't:
You don't blow up a satellite, you simply push it out of orbit far enough that it can't correct back. Absolute destruction isn't the only option, it's just the one that Hollywood things of most often.
Although total destruction is far overkill, a "mission kill" of a satellite is most easily achieved by hitting it with something that will very likely break pieces off. That might also push it out of orbit, but you really are trying to put holes in the sensitive parts, not just move it around.
E.g. ever wonder how long ago slapping a woman was acceptable as a movie scene? Do you remember "The Man with the Golden Gun" (1974)? What about the "Airplane!" (1980)?)
Even today, slapping a woman is quite acceptable when done for comedic effect, as in Airplane!, so I don't think that's a good example.
A couple minutes is enough to confirm there's some copyrighted content
Not for many formats, as there won't be anything usable unless you download a very large chunk.
no matter that the other 90% of the torrent might theoretically be something other than the rest of that GoT episode. Then you use a custom client that requests just those chunks known to contain copyrighted material.
If that's the method they use, then I can argue "fair use", and they won't be able to refute it since they never downloaded the entire torrent, nor will they know if I did.
Every peer who hands you one of those chunks is sharing copyrighted content.
Even better. If they accuse based on one chunk of a torrent, then then only have about 0.1% of the total, and that's certainly fair use, as it's only around 3 seconds of an hour long show.
The key here is that once somebody ponies up the $35 and raises these sorts of challenges ("show me the bits you downloaded from me, personally", etc.), if those challenges work, then it will be posted somewhere and everybody will use the same argument. The **AA won't be able to keep paying for the reviews, and will have to do some actual work to proves infringement. Either way, it will cost them more money. The more likely alternative is that all **AA "evidence" is assumed to be 100% valid regardless of method and the **AA will win all "reviews". Once this comes to light, then I suspect that class-action lawsuits will start.
First of all, I don't have an open wifi... and secondly, I don't infringe on copyright, so how is it that you propose that software which allegedly doesn't make mistakes would link my IP to an infringement, exactly?
No, I said "the monitoring software is assumed to never make mistakes". In other words, everybody who is ruling on your guilt is under the assumption that if you got an alert, it's not a mistake. In your case, it's probably a mistake (although even non-open WiFi can be hacked, and if your account was used, then it is not a mistake by the definition of this program), but your ISP and the "independant review board" will assume that it is not.
So, prove it was a mistake.
FYI: "Guilty till proven innocent" is a concept of Criminal Case Law
Not in the US, it isn't.
and has no role in Civil Litigation, where you must only show "probable cause".
Civil lawsuit outcomes are based on "preponderance of evidence"...in layman's terms, who is "most right", not "probable cause". Regardless of how the determination of liability at the end of the trial is done, you can still sue anybody for anything with no real evidence that they did anything wrong. With far less than "probable cause", civil litigation can reach the discovery stage, at which point you better have deep pockets if you want to continue a defense.
The major difference in a lawsuit and this system is that evidence is not assumed to be valid in a lawsuit...it must be proven. For example, if the plaintiff claims to have evidence of infringement on a computer, the defendant would be able to get to see that computer (usually via a neutral third party) to verify that the alleged evidence does exist and was collected in a proper manner. This discovery might allow the defendant to show that the evidence is faulty in some way.
All this is absent from the "six strikes" system, which assumes that all evidence against the alleged infringer is valid with no possibility for a mistake. Because of this, the term "guilty until proven innocent" applies. Unfortunately, there is really no way to prove innocence once you start with the assumption that all evidence indicating guilt is valid and not subject to rebuttal.
If it's that easy to neutralize drones (making them head back to their designated landing zone, if your human-piloted fighters don't take them out while they're flying autonomously), then they're really quite useless against a technologically advanced opponent.
You can use very similar tactics against planes with pilots, too. If you cut off a pilot from the AWACS data, or communications from front-line forces, a lot of missions would have to be scrubbed.
The only real difference comes when you are repeatedly attacking the same area, and then the pilot could have learned enough to not need much electronic support.
.... that the $35 fee is refunded to consumers who pay it, and are not found to have actually committed any copyright infringement.
OK, you've just gotten an "alert". Prove that it was not issued correctly. Remember that the monitoring software is assumed to never make mistakes, so if they have a list with your IP address, a date and time, and a torrent name, then it that is proof that that IP address downloaded and uploaded that torrent at that date and time. The ISP has furnished a DHCP log that shows your account was holding that IP address at that date and time.
Note that you, personally, are not being accused. Your account is being accused. So, now you have to provide proof that your account is not to blame for something that did happen (according to the "guilty until proven innocent" system in place). You can play the "open WiFi" card once. If you get another "alert" after using the "open WiFi" defense, you can't use it again.
So, prove your innocence.
Riiiight... 'cuz, we all know, ISPs and the MAFIAA are certainly trustworthy entities, who would never misuse people's personal information, or god forbid, lie to support their goals.
Can you imagine how much bandwidth and disk space such a content checking system must have to "confirm copyrighted content"?
A search for "Game of Thrones" on Pirate Bay lists 5 pages (at 30 entries per page) before the torrent size is less than 5GB, and 22 pages before the size is less than 0.5GB. That's over 1TB they would need to download just to verify every torrent for a single title. If they don't download the entire torrent, they can't be sure that it's really is something they hold the copyright to.
With 5 million torrents on TPB, even if only half (likely too low a guess) are copyrighted by companies that belong to **AA, and if the average size is 250MB, that's 625TB of data.
The *AA are paying DtecNet to send the strikes in the first place, and if you appeal it will cost them more than $35 to put together the paperwork to argue with you.
I sincerely doubt that printing out an IP address, date and time, and torrent name will cost $35. This is all the "evidence" they have, and it's unknown whether you will get to see it unless you ask for a review. The "alert" might be nothing more than "you've been a bad boy" with no details.
It's not necessarily a waste of $35, since if they are unable to verify the alleged infringement, you get the $35 refunded, and the "strike" is removed.
Just how many claims do you think will be overturned by a group of people that were picked by the same people who made the claim?
In particular, when the entire evidence of infringement for these claims consist of an IP address, a date and time, and the name of a torrent, exactly how are you going to prove that you didn't infringe? When the party making the claim of infringement doesn't even have to prove that the named torrent was their copyrighted content (much less any of the other things needed for infringement, like proving uploading), how can you possibly defend yourself? Even if you aren't running any bittorrent software at all, how can you prove that? And, if you are running bittorrent software, how can you prove that you were not part of that torrent at that time?
This is one of the many problems with "guilty until proven innocent"...often you have to prove a negative, and sometimes that can't be done.
Hollywood *rarely* gets technical issues right, but the speech in Top Gun where they were talking about pilots becoming reliant on missiles in Korea was actually true, and the basic principle should still be true today. Dogfighting specifically doesn't really happen any more, but the basic evasive agility skills that it's based on are still applicable.
It's definitely true that US pilots had been losing such skills over time and something needed to be done, but it's somewhat ironic that the platform with pretty much the ultimate "fire and forget" missle was used in the Navy training.
Your global satellite system is no match for China's anti-satellite missiles.
Destruction of a satellite isn't going to happen unless it's an all-out shooting war, as the chance for collateral damage to a "friendly" satellite is pretty high. Although there might not be immediate damage, spreading debris in the same orbit height as your own satellites is not good for long-term stability.
This drone stuff only works because all the opponents are decades behind the US technologically, so they have little ability (yet) to block the radio signals needed to keep these aircraft under control.
If all you want to do is cut off the signal to/from the drone, it's pretty easy as long as you have a general idea of the frequency used and have a directional antenna and a big enough transmitter (and can track the drone well enough to keep your antenna pointed at it).
This could protect you from drone strikes, but the failsafe systems in the drone would mean that it would just turn and head for the designated landing zone.
Blocking GPS signals is something well beyond the capabilities of some Taliban fighters living in a cave and carrying nothing more advanced than AK47s, however it's well within China's abilities.
Again, this is actually pretty easy to do, but gains you nothing, as the drones also have intertial navigation systems which allow them to at least head for the failsafe landing zone. Spoofing GPS signals, on the other hand, is very hard.
I can get the UAC for network connection though. If you download a office suite say and it tries to connect to the internet you might be suspicious.
Since there is now a lot of legitimate software that requires a network connection to "phone home", it just gets to the point that a user will blindly click the "make this dialog go away" button without reading. In addition, all the dialog gives you is the program name, and malware can have names that seem legitimate, while I sometimes have to google an EXE or DLL to see if it is OK or not.
Also, Internet Explorer is whitelisted, so if the malware creates an IE instance (which doesn't require a visible window), then the user will never know. Between being annoyed by UAC for permissions to open the firewall plus the fact that the firewall isn't really useful against much of what a user should be concerned about (things like a keylogger phoning home where a POST to some random control server somewhere probably wouldn't raise a red flag in IE) makes UAC just more security theater...it's annoying, doesn't really protect you, but allows Microsoft to say "we did something".
Use myname@site name.mydomain
Although I suggested this, too, I now realize it has the same problem as plus addressing, in that either the initial sign-up or the unsubscribe might use a broken test when trying to determine if the e-mail address is legitimate.
I have personally dealt with co-workers who assumed that the domain name must be of the format "word.exactly3characters". New top-level domains have at least killed the thought that all TLDs are 3 characters long, but the "only one period in the domain name" belief is still there for some people.
And change site name.mydomain to their mail servers! : D
I find your ideas intriguing, and wish to subscribe to your newsletter.
The problem is for 90% of users a UNC prompt or its mac/linux equivalent pops up and they click ok. To most users the fingers go in the ears as soon as you try to explain the risks and what is happening and they just ask "So what do I need to click to continue?"
If common tasks didn't require the user to answer these pop-ups, then they would see them as "unusual" and wouldn't be as likely to just "click to continue".
There are quite a few control panel settings in Windows 7 that require answering a UAC prompt just to see the settings. Any software that tries to make a network connection and isn't on the Windows firewall "approved" list generates a UAC prompt. Then there are some settings (like "Adjust Visual Effects" in the "Perfomance Information and Tools" control panel page) that shouldn't require UAC at all, as they are merely personal preferences that are no different from desktop wallpaper as far as security or system stability is concerned.
The FCC ruled that Verizon's blocking tethering without paying extra is illegal.
They don't block general tethering...they only block using your phone as a WiFi hotspot, since that allows multiple other devices to use the connection at the same time. This is what the watchdog app is for.
I can one-to-one tether with no problem, and I have a grandfathered unlimited data plan.
How does that work when you send e-mail from half a dozen different systems, including Outlook, pine, Android mail, sendmail, and in a pinch, even telnet to port 25 or openssl to port 465/587?
These are one-off addresses tied to each company and are used for preventing spam to a personal e-mail address, and most of them aren't ever used to send e-mail. The few times you need to, it's also usually not critical that the e-mail be answered right now (unlike a business e-mail), so you can wait a bit until you are at one of your machines with the correct software (because you aren't going to be doing this from random machines, ever, as it's still your personal e-mail).
If you are really desperate, though, you just run some remote access software to get to a machine with the right software.
I (not the submittor) frequently use <myname>+<site name>@<mydomain>.
One of the issues with this is that <myname>@<mydomain> will be delivered, too. And, if that's your "real" e-mail address, then it's now out there for spammers to hit.
If you instead use something that doesn't rely on special address parsing (like <myname><site name>@<mydomain> or <myname>@<site name>.<mydomain>), you can just ditch the e-mail address once it is compromised. There are a couple of companies that I had to do this to simply because their "you've done business with us, which we consider an opt-in" mailing list has no reasonable way to unsubscribe.
You can't legally charge anyone to listen to your CD
Maybe you can't charge people to listen to it, but you can legally charge people to rent the CD from you. Once those people rent it, they can choose to listen to it if they want. The "license" printed on most media these days that implies you cannot rent it is not legally binding.
and you certainly can't play it for a crowd of strangers even if they don't pay you a dime.
I wish you'd tell that to all the idiots who drive by my house with their stereo cranked to "any occupants of this car will be deaf by next year". Also, if I grab random people off the street and ask them if the want to come to my house to listen to the latest insert artist name here CD, it's perfectly legal. Even though once I get enough people, they are a crowd, and they are strangers, it's not a public performance.