"No, what I'm saying is that, at the moment, the government(s) largely prop up one organisation's development: microsoft's."
Actually, the US government contracts with Microsoft, Red Hat, Novell, Sun, IBM, and various other companies. It is actually considered a problem, and one of the things Obama seeks to fix, because different branches and government agencies use different software and different file formats, which impedes communication between other agencies and the people.
"Redhat could be contracted to provide government with OS, but Ubuntu might be preferred for a future upgrade. Currently, we don't have this freedom."
What makes you think that switching from Red Hat to Canonical would be so easy? We are not talking about a single machine here, we are talking about thousands (millions?) of computers at a time, with various regulations on data retention. If the government standardized on RHEL, it is unlikely that a migration to another Linux, let alone a different kinda of OS altogether, would be a simple or easy move.
Check the license agreement, if you can find it. I could not find it, but other licenses from SAS indicate that they provide no guarantees of the software's fitness for any purpose. You can be that a simulation error in SAS would not matter one bit if it caused an airplane crash, the company would be in the clear.
I love how she uses the term, "freeware," rather than "open source" or "free software."
Matlab is a programming language. Let me guess, you also think that Rexx and AWK are not programming languages, because they have weird syntax and are specialized for certain tasks?
You could compile an R program if you wanted to, and if you were willing to write a compiler. Same with Matlab. Your argument is like saying that AWK is not a programming language, because it is interpreted.
"Our taxes should not prop up individual corporations -- especially when that monopolizes their power and cripples other parts of the economy."
I agreed with everything else in your post, but here is where I have to disagree. Suppose the US government standardizes on free software -- do you really want tax dollars spent creating a massive government IT force, or would you rather the government contracted with Red Hat or Novell? It is not inherently bad for the government to contract with corporations -- in fact, this is why corporations were created in the first place. We should focus on the real issue, which is that software used by the government should be free-libre, except in certain cases involving military systems (missile control, RADAR, etc.).
Actually, those two lines were only shut down in a single area, and very briefly while new parts were installed. The rest of the system was operational, and there were no stations where service was shut down -- the stations in the area where those lines were shut down were being serviced by other subway lines. The system was robust enough to allow for a very quick reroute of trains from West 4 street (three stations north of the fire) to Jay street (two stations south of the fire). Not only that, but the relay room that was destroyed by the fire controlled only two of the four tracks to that station (Chambers St./World Trade Center), and the other two tracks, which were used for relaying the third line that services that station (and terminates there) remained operational because they were controlled by another relay room.
Around the same period, the MTA attempted to bring the first phase of the new, computerized central control system online; this was the radio coordination system, which is currently running equipment that was installed in the 1960s. The computerized system crashed when three simultaneous emergency calls were made -- a typical situation in a system with more than 400 stations and a hundred or so trains in operation at any given time -- and communication in the entire system was shut down while they reactivated the old system. Imagine, in a few years, if a fire occurs in the control center -- the entire system would be out of order until the older control rooms were unlocked and reactivated. As a case in point, there was a fire in a control room in the 1980s, much more serious than a relay room like the one at Chambers St., and a service along one of the subway lines was shut down until control could be transferred to other locations (this was a complex move at that time). No other lines were affected by the fire, because the other control rooms operated completely independently, from an electronic perspective (they are coordinated by phone).
People just assume that technology from the 1920-30s, which runs the majority of the system, must in inadequate and that upgrading it will make the transit system better. Experience shows that this is simply not true. The 1930s vacuum tube relay equipment, which is controlled with electromechanical lever machines, is remarkably reliable, and gets the job done just fine. The only real deficiency is that there is no electronic method for tracking multiple lines on a single segment of track, but this is made up for through a system of buttons installed at major station stops, which allow train operators to indicate their route to a control room when it is necessary to do so (as it is at points where lines are separated and sent down different tracks). In some cases, even that is overkill, because the control room sits at the exact point where trains stop. Computerization offers little advantage beyond more accurate accounting and schedule measurements.
I am not against the idea of more modern equipment. It is certainly the case that a computer could calculate the placement of trains in maintenance yards more efficiently than a human can, or even the placement of trains on relay tracks at locations where several lines are terminated. The current plan, however, is deeply flawed in that it seeks to centralize everything and leave the old system locked up for emergencies. Central command systems might work well on smaller systems like Chicago or Boston, but given the enormous size of the NYC transit system, it seems severely misguided; no surprise, though, since the MTA has not made many good decisions over the past few years.
9.6kbps is acceptable for email access, assuming POP3 (though your blackberry will already do this, so what would be the point?). It is also decent for IM, although Jabber is too bandwidth heavy so you'd want to stick with something like IRC.
Also, this IS dialup, it just isn't v.92 56k dialup.
Cell providers also routinely filter data traffic, because they want to charge extra for mobile Internet and fax plans. Verizon definitely does this, and T-mobile kills fax but not dialup users. How this is not a violation of the Sherman act is a mystery to me.
My first experience was using a normal, run of the mill cell phone that had an IR port as a modem -- back in high school. That was beyond annoying, because the alignment had to be withing about 5 degrees, but it worked, and back then 9.6kbps was not unbearable even for the web. These days I do it with Bluetooth when I am traveling, and I can really only check my email via POP3.
It was conceived of as a way to retain the wilderness of Manhattan. Back then, it was considered OK to create a park that had the appearance of wilderness as part of achieving that goal.
Beyond there, the park was originally north of the developed part of the city. All the stuff that surrounds central part right now was build decades later. The "poor who already lived there" were living in shanty towns, not really anything that would be considered part of a city, any more than the people they found living in an Amtrak tunnel a few years ago were. The park was originally conceived of as a way to retain some of the wilderness of Manhattan, which is a bit ironic considering that they brought in a team of landscapers.
New York City's mayors have a bad track record, especially over the past 30 years, but Giuliani did a decent job of reducing crime. When I was just a baby, Times Square was actually a dangerous place to be, 5th Avenue was full of seedy shops and illegal fronts; those are now considered to be safe places to be, even in the middle of the night. The subways are also very safe, especially since the institution of "off hours" waiting areas that are in plain view of the police and MTA personnel.
"I don't deny that nature is important, and don't doubt that experiencing natural settings regularly is a contributor to mental health."
Funny you should mention this, this TFA is way off with regards to central park. Central park was not originally in the middle of Manhattan, it was north of the developed part of Manhattan, and was put in place to preserve some part of the natural landscape of the region. Of course, it did not really preserve the natural landscape, since they started the construction of the park by razing the forest that was there and then creating meadows in its place, but it turned out to be a good idea. Ironically, there was some pressure against the park originally, by people who felt it robbed the city of prime real estate; ironic, because a park view vastly increases the value of apartments and condos, probably because people so desperately want to see green instead of gray.
"It's also noisy. Maybe they should do some maintenance, and switch over to a rubber-tired system."
That would be difficult, since the NYC subways use the rails for grounding. The main reason the subways in NYC are so noisy is the speed that the trains operate at, which is typically 10-20MPH faster than other subway systems in America (which is why similarly old systems, like the Boston subways, are so much quieter). There is an effort with the newest subway trains to reduce noise, but that is mainly aimed at the passengers riding the train, not those standing on the platforms.
One of the things that the NYC subway system has going for it, that other systems do not really have, is the ability to operate 24x7x365 with few disruptions in service. There are several reasons for this, but the primary two are the distributed nature of the control system (which is unfortunately due to be centralized as part of a plan to install computers to replace the ancient equipment they use) and the large number of lines and tracks which make reroutes possible. It is possible to perform maintenance on the NYC subways, and in fact, this is done on nights and weekends, which is why there are route changes every weekend, with the exception of holidays.
"Must be all those pollutants in the river. Maybe they've permanently altered your taste buds."
NYC's water supply does not come from the rivers that surrounding Manhattan. The water in NYC comes from a large reservoir in the mountains in the middle of New York State, and is carried to the city using three enormous pipes. The tap water is actually among the cleanest in the US, and NYC is one of the few places where the majority of contaminants in tap water come from old pipes in the final stages of delivery, rather than the supply itself.
"Seriously, the air absolutely stinks and the streets are filthy."
This is not unique to NYC, it is the case in any large city. Large cities always have been and always will be more polluted than small cities and towns. When you have millions of people living in such a small area, it is difficult to keep the ground and air pristine.
"Clearly you live in some parallel dimension filled with revisionist history."
The Tet offensive was a significant military victory for the US forces in Vietnam, as it had resulted in a vast weakening of Northern Vietnamese forces. Following that, the US was in a position to take the rest of that country, if we had been committed enough to increase the number of troops in the region. The US public was not committed, due to the number of casualties on the US side of the war and a belief that previous defeats indicated that such action would also end in defeat.
I never said it was an overall good thing. Politically, it would have been a disaster to send more troops there and raise the death toll. It would also have entailed expanding the draft to the middle classes, which would have marked the end of any politicians involved. My point was that, from a military perspective, victory was possible.
"The are plenty of plausible scenarios where the NSA might suggest a deliberately weak cryptosystem. For example, they know it will be weak only with keys not tested against criteria X. They approve the system and secretly test keys against this criteria while simultaneously exploiting our enemy's adoption of awesome "US military grade cryptography"."
Which would also open the possibility of a foreign power, particularly that large on in Asia or its northern neighbor that has some territory in Europe, discovering criterion X and exploiting US use of the system. Various foreign nations have electronic and signals surveillance operations in place (granted, so does the USA) that target US corporations and government agencies, and the discovery of a back door would be a boon to those operations. The NSA knows this; in fact, this is a principle that is central to modern cryptography, usually expressed through the concept of "the strength of a cipher should depend solely on the secret key." Deviating from this principle would sabotage national security, regardless of how you want to state it. It is not a straw man, because this principle has been central to the NSA's cipher designs since the beginnings of the NSA, and so violating it would be a deliberate, intentional effort to weaken national security.
"We did give Saddam Hussein the key to the city of Detroit."
He was once an ally, but that is irrelevant because it was not done by the NSA.
"How'd that Vietnam war ever turn out?"
From a military perspective, we were winning prior to the pull-out. We left because of eroded support for the war among the American public.
"How are things in Iran these days?"
You are 1 for 3, things are bad in Iran. But, as with the key to Detroit, this was not an NSA action.
"No the US would never shortsightedly adopt a policy against its own interests, especially with regard to cryptography."
The laws surround cryptography are not passed by the NSA, they are passed by congressmen with little to no understanding of the field or how it works. Export restrictions on cryptography have nothing to do with the NSA, in fact, the NSA operates under the assumption that regardless of export law, publicly available cryptography systems will escape US borders. The idea that a cipher itself must be kept secret is beyond outdated; in fact, it is an idea that was dropped centuries ago, when the Kama Sutra cipher was published. While the NSA has, in the past, kept the nature of the ciphers used for SECRET and TOP SECRET level documents classified, this is no longer the case; AES represents a departure from that position.
AES is a mandatory standard for SECRET and TOP SECRET communications. This goes beyond the NSA, to every branch of the government. If the NSA had deliberately inserted a back door into AES, it would open the possibility of a foreign power deciphering high security communication within the US government. If you do not trust the NSA -- which hires expert cryptographers and security researchers -- to make good decisions about the security of the USA, then you might as well leave now for your own protection.
Of course, you do trust the NSA, and I notice that you never questioned my assertion about the DES S-boxes or anything relevant to actual cryptography. Another example would be the revision of SHA-0 to SHA-1 by the NSA; SHA-1 is more resistant to collision attacks than SHA-0. You do not seem to be interested in questioning whether or not the NSA introduced a weakness of some kind into SHA-1 or SHA-2. I agree that congress has a habit of passing stupid laws when it comes to cryptography, but to claim that this implies that the NSA has been trying to sabotage national security just screams of tin foil.
They are probably not allowed to do it, by law. Until they can prove that a computer is being used for illegal purposes, hacking their way into it and messing with the data stored on it is more likely to get the criminals off "on a technicality" than get them locked away for life.
It is very unlikely that the US government would deliberately sabotage the encryption standard for the entire country. It is asking for trouble to do so, since foreign powers are known to be engaged in hacking campaigns against US businesses and agencies, and back doors could be discovered by those powers. I thought we learned this lesson with DES, when the government demanded different S-boxes without telling anyone why, and the S-boxes they chose turned out to make the algorithm more resilient to differential attacks?
The probably just brute forced the key. It probably required a significant amount of time -- the article does not actually give timescales here, and Aragon's trial could have taken nearly two years, considering the high level operation that we are talking about here. With that much time, and the priority of the case, I would not doubt that the government could have devoted enough CPU time to brute force the password.
There are other ways that they could have gotten the password. For example, they could have analyzed the wear on each key from his keyboard, to figure out which letters were more likely to have been used in a password. It would not have cracked the password instantly, but it would remove many months of work from the process.
Slashdot Mirror
"No, what I'm saying is that, at the moment, the government(s) largely prop up one organisation's development: microsoft's."
Actually, the US government contracts with Microsoft, Red Hat, Novell, Sun, IBM, and various other companies. It is actually considered a problem, and one of the things Obama seeks to fix, because different branches and government agencies use different software and different file formats, which impedes communication between other agencies and the people.
"Redhat could be contracted to provide government with OS, but Ubuntu might be preferred for a future upgrade. Currently, we don't have this freedom."
What makes you think that switching from Red Hat to Canonical would be so easy? We are not talking about a single machine here, we are talking about thousands (millions?) of computers at a time, with various regulations on data retention. If the government standardized on RHEL, it is unlikely that a migration to another Linux, let alone a different kinda of OS altogether, would be a simple or easy move.
Check the license agreement, if you can find it. I could not find it, but other licenses from SAS indicate that they provide no guarantees of the software's fitness for any purpose. You can be that a simulation error in SAS would not matter one bit if it caused an airplane crash, the company would be in the clear.
I love how she uses the term, "freeware," rather than "open source" or "free software."
Matlab is a programming language. Let me guess, you also think that Rexx and AWK are not programming languages, because they have weird syntax and are specialized for certain tasks?
You could compile an R program if you wanted to, and if you were willing to write a compiler. Same with Matlab. Your argument is like saying that AWK is not a programming language, because it is interpreted.
"Our taxes should not prop up individual corporations -- especially when that monopolizes their power and cripples other parts of the economy."
I agreed with everything else in your post, but here is where I have to disagree. Suppose the US government standardizes on free software -- do you really want tax dollars spent creating a massive government IT force, or would you rather the government contracted with Red Hat or Novell? It is not inherently bad for the government to contract with corporations -- in fact, this is why corporations were created in the first place. We should focus on the real issue, which is that software used by the government should be free-libre, except in certain cases involving military systems (missile control, RADAR, etc.).
No, but software machines do exist. They run inside of a hardware machine. That's computing...
Actually, those two lines were only shut down in a single area, and very briefly while new parts were installed. The rest of the system was operational, and there were no stations where service was shut down -- the stations in the area where those lines were shut down were being serviced by other subway lines. The system was robust enough to allow for a very quick reroute of trains from West 4 street (three stations north of the fire) to Jay street (two stations south of the fire). Not only that, but the relay room that was destroyed by the fire controlled only two of the four tracks to that station (Chambers St./World Trade Center), and the other two tracks, which were used for relaying the third line that services that station (and terminates there) remained operational because they were controlled by another relay room.
Around the same period, the MTA attempted to bring the first phase of the new, computerized central control system online; this was the radio coordination system, which is currently running equipment that was installed in the 1960s. The computerized system crashed when three simultaneous emergency calls were made -- a typical situation in a system with more than 400 stations and a hundred or so trains in operation at any given time -- and communication in the entire system was shut down while they reactivated the old system. Imagine, in a few years, if a fire occurs in the control center -- the entire system would be out of order until the older control rooms were unlocked and reactivated. As a case in point, there was a fire in a control room in the 1980s, much more serious than a relay room like the one at Chambers St., and a service along one of the subway lines was shut down until control could be transferred to other locations (this was a complex move at that time). No other lines were affected by the fire, because the other control rooms operated completely independently, from an electronic perspective (they are coordinated by phone).
People just assume that technology from the 1920-30s, which runs the majority of the system, must in inadequate and that upgrading it will make the transit system better. Experience shows that this is simply not true. The 1930s vacuum tube relay equipment, which is controlled with electromechanical lever machines, is remarkably reliable, and gets the job done just fine. The only real deficiency is that there is no electronic method for tracking multiple lines on a single segment of track, but this is made up for through a system of buttons installed at major station stops, which allow train operators to indicate their route to a control room when it is necessary to do so (as it is at points where lines are separated and sent down different tracks). In some cases, even that is overkill, because the control room sits at the exact point where trains stop. Computerization offers little advantage beyond more accurate accounting and schedule measurements.
I am not against the idea of more modern equipment. It is certainly the case that a computer could calculate the placement of trains in maintenance yards more efficiently than a human can, or even the placement of trains on relay tracks at locations where several lines are terminated. The current plan, however, is deeply flawed in that it seeks to centralize everything and leave the old system locked up for emergencies. Central command systems might work well on smaller systems like Chicago or Boston, but given the enormous size of the NYC transit system, it seems severely misguided; no surprise, though, since the MTA has not made many good decisions over the past few years.
Oh really? What about SED scripts, which are abstract state machines but are (usually) not implemented in hardware?
9.6kbps is acceptable for email access, assuming POP3 (though your blackberry will already do this, so what would be the point?). It is also decent for IM, although Jabber is too bandwidth heavy so you'd want to stick with something like IRC.
Also, this IS dialup, it just isn't v.92 56k dialup.
Cell providers also routinely filter data traffic, because they want to charge extra for mobile Internet and fax plans. Verizon definitely does this, and T-mobile kills fax but not dialup users. How this is not a violation of the Sherman act is a mystery to me.
My first experience was using a normal, run of the mill cell phone that had an IR port as a modem -- back in high school. That was beyond annoying, because the alignment had to be withing about 5 degrees, but it worked, and back then 9.6kbps was not unbearable even for the web. These days I do it with Bluetooth when I am traveling, and I can really only check my email via POP3.
Abstract state machines?
It was conceived of as a way to retain the wilderness of Manhattan. Back then, it was considered OK to create a park that had the appearance of wilderness as part of achieving that goal.
Beyond there, the park was originally north of the developed part of the city. All the stuff that surrounds central part right now was build decades later. The "poor who already lived there" were living in shanty towns, not really anything that would be considered part of a city, any more than the people they found living in an Amtrak tunnel a few years ago were. The park was originally conceived of as a way to retain some of the wilderness of Manhattan, which is a bit ironic considering that they brought in a team of landscapers.
New York City's mayors have a bad track record, especially over the past 30 years, but Giuliani did a decent job of reducing crime. When I was just a baby, Times Square was actually a dangerous place to be, 5th Avenue was full of seedy shops and illegal fronts; those are now considered to be safe places to be, even in the middle of the night. The subways are also very safe, especially since the institution of "off hours" waiting areas that are in plain view of the police and MTA personnel.
"I don't deny that nature is important, and don't doubt that experiencing natural settings regularly is a contributor to mental health."
Funny you should mention this, this TFA is way off with regards to central park. Central park was not originally in the middle of Manhattan, it was north of the developed part of Manhattan, and was put in place to preserve some part of the natural landscape of the region. Of course, it did not really preserve the natural landscape, since they started the construction of the park by razing the forest that was there and then creating meadows in its place, but it turned out to be a good idea. Ironically, there was some pressure against the park originally, by people who felt it robbed the city of prime real estate; ironic, because a park view vastly increases the value of apartments and condos, probably because people so desperately want to see green instead of gray.
"It's also noisy. Maybe they should do some maintenance, and switch over to a rubber-tired system."
That would be difficult, since the NYC subways use the rails for grounding. The main reason the subways in NYC are so noisy is the speed that the trains operate at, which is typically 10-20MPH faster than other subway systems in America (which is why similarly old systems, like the Boston subways, are so much quieter). There is an effort with the newest subway trains to reduce noise, but that is mainly aimed at the passengers riding the train, not those standing on the platforms.
One of the things that the NYC subway system has going for it, that other systems do not really have, is the ability to operate 24x7x365 with few disruptions in service. There are several reasons for this, but the primary two are the distributed nature of the control system (which is unfortunately due to be centralized as part of a plan to install computers to replace the ancient equipment they use) and the large number of lines and tracks which make reroutes possible. It is possible to perform maintenance on the NYC subways, and in fact, this is done on nights and weekends, which is why there are route changes every weekend, with the exception of holidays.
"Must be all those pollutants in the river. Maybe they've permanently altered your taste buds."
NYC's water supply does not come from the rivers that surrounding Manhattan. The water in NYC comes from a large reservoir in the mountains in the middle of New York State, and is carried to the city using three enormous pipes. The tap water is actually among the cleanest in the US, and NYC is one of the few places where the majority of contaminants in tap water come from old pipes in the final stages of delivery, rather than the supply itself.
"Seriously, the air absolutely stinks and the streets are filthy."
This is not unique to NYC, it is the case in any large city. Large cities always have been and always will be more polluted than small cities and towns. When you have millions of people living in such a small area, it is difficult to keep the ground and air pristine.
"Clearly you live in some parallel dimension filled with revisionist history."
The Tet offensive was a significant military victory for the US forces in Vietnam, as it had resulted in a vast weakening of Northern Vietnamese forces. Following that, the US was in a position to take the rest of that country, if we had been committed enough to increase the number of troops in the region. The US public was not committed, due to the number of casualties on the US side of the war and a belief that previous defeats indicated that such action would also end in defeat.
I never said it was an overall good thing. Politically, it would have been a disaster to send more troops there and raise the death toll. It would also have entailed expanding the draft to the middle classes, which would have marked the end of any politicians involved. My point was that, from a military perspective, victory was possible.
"The are plenty of plausible scenarios where the NSA might suggest a deliberately weak cryptosystem. For example, they know it will be weak only with keys not tested against criteria X. They approve the system and secretly test keys against this criteria while simultaneously exploiting our enemy's adoption of awesome "US military grade cryptography"."
Which would also open the possibility of a foreign power, particularly that large on in Asia or its northern neighbor that has some territory in Europe, discovering criterion X and exploiting US use of the system. Various foreign nations have electronic and signals surveillance operations in place (granted, so does the USA) that target US corporations and government agencies, and the discovery of a back door would be a boon to those operations. The NSA knows this; in fact, this is a principle that is central to modern cryptography, usually expressed through the concept of "the strength of a cipher should depend solely on the secret key." Deviating from this principle would sabotage national security, regardless of how you want to state it. It is not a straw man, because this principle has been central to the NSA's cipher designs since the beginnings of the NSA, and so violating it would be a deliberate, intentional effort to weaken national security.
I have a better solution: stop using the Internet. Get everyone to agree to start a new, private, NZ only network using modems and UUCP.
"We did give Saddam Hussein the key to the city of Detroit."
He was once an ally, but that is irrelevant because it was not done by the NSA.
"How'd that Vietnam war ever turn out?"
From a military perspective, we were winning prior to the pull-out. We left because of eroded support for the war among the American public.
"How are things in Iran these days?"
You are 1 for 3, things are bad in Iran. But, as with the key to Detroit, this was not an NSA action.
"No the US would never shortsightedly adopt a policy against its own interests, especially with regard to cryptography." The laws surround cryptography are not passed by the NSA, they are passed by congressmen with little to no understanding of the field or how it works. Export restrictions on cryptography have nothing to do with the NSA, in fact, the NSA operates under the assumption that regardless of export law, publicly available cryptography systems will escape US borders. The idea that a cipher itself must be kept secret is beyond outdated; in fact, it is an idea that was dropped centuries ago, when the Kama Sutra cipher was published. While the NSA has, in the past, kept the nature of the ciphers used for SECRET and TOP SECRET level documents classified, this is no longer the case; AES represents a departure from that position.
AES is a mandatory standard for SECRET and TOP SECRET communications. This goes beyond the NSA, to every branch of the government. If the NSA had deliberately inserted a back door into AES, it would open the possibility of a foreign power deciphering high security communication within the US government. If you do not trust the NSA -- which hires expert cryptographers and security researchers -- to make good decisions about the security of the USA, then you might as well leave now for your own protection.
Of course, you do trust the NSA, and I notice that you never questioned my assertion about the DES S-boxes or anything relevant to actual cryptography. Another example would be the revision of SHA-0 to SHA-1 by the NSA; SHA-1 is more resistant to collision attacks than SHA-0. You do not seem to be interested in questioning whether or not the NSA introduced a weakness of some kind into SHA-1 or SHA-2. I agree that congress has a habit of passing stupid laws when it comes to cryptography, but to claim that this implies that the NSA has been trying to sabotage national security just screams of tin foil.
They are probably not allowed to do it, by law. Until they can prove that a computer is being used for illegal purposes, hacking their way into it and messing with the data stored on it is more likely to get the criminals off "on a technicality" than get them locked away for life.
It is very unlikely that the US government would deliberately sabotage the encryption standard for the entire country. It is asking for trouble to do so, since foreign powers are known to be engaged in hacking campaigns against US businesses and agencies, and back doors could be discovered by those powers. I thought we learned this lesson with DES, when the government demanded different S-boxes without telling anyone why, and the S-boxes they chose turned out to make the algorithm more resilient to differential attacks?
The probably just brute forced the key. It probably required a significant amount of time -- the article does not actually give timescales here, and Aragon's trial could have taken nearly two years, considering the high level operation that we are talking about here. With that much time, and the priority of the case, I would not doubt that the government could have devoted enough CPU time to brute force the password.
There are other ways that they could have gotten the password. For example, they could have analyzed the wear on each key from his keyboard, to figure out which letters were more likely to have been used in a password. It would not have cracked the password instantly, but it would remove many months of work from the process.
Apple's latest developer product: the iLL! Your linked lists will never be more effective!
I thought that half the Slashdot crowd made a resolution to "bulk up and stop exercising only the fingers."