The New York Times is down for the count at the moment so I couldn't actually read the article. I've taken a number of online courses over the last couple of years at the graduate electrical engineering level so maybe I can submit a relavent experience.
Online courses have worked out really well for me. I work full time but am still interested in learning as much as possible about my chosen field. In the geography where I work there is no local engineering college and there may never be. About 100 miles from me is an excellent engineering college. My company is more than willing to fund any education which I can make use of (even if its sometimes rather tangential) for my job.
Travelling 200 miles 3 days a week for classes would be a big problem though. Not insurmountable but it is probably enough that it would make me think twice about taking courses. One of the perks that has been set up is distance learning through the UNITE program at this particular university. I can either attend classes in real time in a video conference or watch streaming video over my computer. Assignments can be faxed and fedexed in, midterms and finals are done in the videoconference room. So far I haven't encountered any problems that in class participants do.
No they got it 100% right this time. They are hackers, they're using their hacking skills for malicious purposes therefore they're also crackers. The term hacker itself is grey, there are good hackers and there are bad hackers. The problem is that in the media hackers has been used to refer to the population of hackers who operate contrary to the law as opposed to the entire population of hackers.
Even leaving it at reports of 'malicious hackers' would've been correct. They're hackers and they're malicious. It isn't implying that all hackers are malicious anymore than saying 'corrupt police officer' would imply that all police officers are corrupt.
You're assuming that source code is distributed to your friends. Chances are its binary or a driver or whatever. "Here, does this work?" as opposed to "Compile this and tell me how it works." I understand the issue, I agree that something was probably violated, but I say that this is selective enforcement. In general nobody enforces this clause because a) nobody knows when its violated b) nobody cares c) if every time a beta was released full source had to be made available certain forms of development would grind to a screeching halt (such as reverse engineered drivers for obscure closed source hardware). It happens that this publicized event involves a commercial company and so everybody is screaming about something that already happens frequently.
Fine, then Corel should give the beta testers a shiny new penny as payment. This whole argument is a bit disingenous at best. Everybody who has ever developed or worked on Open Source software has probably at some point had some code which they've tested via friends or aquaintances without making it available to the general public until obvious bugs are out. It might be as simple as "run this and tell me if it immediately crashes your computer" but the concept is the same. To a lone developer how do you define "non-internal"? Basically since there is no corporation anybody other than that developer is non-internal. This would mean that any lone developer would have to immediately do a full source code release if any other party ever was to do any testing.
It's unworkable.
The GPL is a good thing, but that doesn't mean that its not wrong some of the time. Rather than griping at Corel take a look at the problem and fix or be ready for yet another Open Source license.
The computer manufactures are trying to capture the 75% of the market that has been resistant to purchasing computers beyond a game machine like a PlayStation in the past. To do this they needed systems that weren't too much more than the typical game system or preferably (with a little deceit and trickery) the same price.
Hand me down systems would actually be fine for this market. Every system I've stopped using to upgrade to something bigger and better has been given to a family member or friend (other than my C-64 and TI/99-4A) and they're all being well used. They aren't the type of people who read mainstream computer magazines so this is just fine. In fact my most recent donation is being used to start up a graphic design business and is working wonderfully. Fast machines aren't that important for most people, most of the time at the computer is spent thinking about what to do next.
This breaks down because most people read the typical trade publication which insist that you need the fastest machine possible, the latest release of Windows and really fast 3D acceleration for the latest 3D games. People buy into it. There's truth in that statement for people who do like 3D games etc, but the target audience for these low cost systems are the people who were resistant to buying one despite the internet hype. They'll visit a few web sites, play a few games, maybe balance a cheque book and write to Aunt Sally. A few of them will get into computers in a big way and will end up purchasing a quicker computer, most of them will only use it an hour or two a week.
The problem for high end users is that these machines have either a miniscule profit margin or even a negative one. The high end machines fund the profit shortcomings of the low end ones, though most companies do absolutely nil R&D anyway.
That's pretty much what they are doing and the manufacturers of the machines are selling them at firesale prices. They're counting on a cut from the provider to make up for their losses. The only problem is that it isn't working. The manufacturers who sell these low cost systems are getting hammered (check EE times or EDN etc). I wouldn't be at all suprised to see the number of computer manufacturers thinned out a lot over the next few years.
Sure, they're losing money on each computer, but they're making up for it in volume!
Conspiracy to commit murder, attempted murder, assault. If there's no body there's no murder but there still may be a crime. If I shoot at you and miss its not murder but it is attempted murder. If I hire somebody to kill you and that person is an undercover cop no murder happens but there is a conspiracy to commit murder.
There is a similar gamut of crimes to cover near-offences of child molestation.
Just pretending to be a 13 year old girl doesn't make it entrapment. An undercover cop impersonating a prostitute in an area known to be frequented by prostitutes isn't entrapment for instance. If the undercover enticed somebody who wasn't trolling for it then it would be entrapment (I was on my way to work, I was at a stop light and this person offered to show me a good time)
It all depends on who brings up the sexual context first in this case. The fact that kiddy porn was found in his posession probably indicates that there is some sort of history that he does go out of his way for it.
Becoming a certified professional engineer sometimes is required and never hurts. In general most people who work in an engineering capacity do not require certification. They work as designers or consultants within a company or more usually a small group within a company. If you're working in the capacity of a consultant than most likely being certified as a professional engineer will be required before anybody will even talk to you. The certification is an insurance policy that you are legally responsible for your own actions and are aware of the laws governing your profession and have agreed to adhere to the ethical mandates which are part of being a licensed engineer.
What does all this mean? You are legally responsible for your own actions, much more so than the general public is (to the governing body the general public is synonymous with the unwashed masses). Any advice you give, even for free, at a bar, on a cocktail napkin must be technically sound or you can be sued. The certification body will operate against you in order to protect the profession. Even if the advice was technically sound but the person receiving the advice wasn't technically proficient to make use of it and as a result suffered financial or physical injury you may be in trouble.
If you've ever wondered why those prickish lawyers and doctors refuse to give free advice even to close personal friends then this is why. The cards are very much stacked against the licensed practicioner, to operate requires very expensive insurance policies. If you're a consultant within a consulting agency the company should provide insurance for you. Make sure to check that you're still protected after you've left the company for past work.
Sadly there's a good chance that that isn't the status quo for commercial products. With any sort of review process imposed by the government you can bet that the goal of the review is to have some way of recovering encrypted data. It may be as simple as a back door or it may be as subtle as reducing entropy during the encryption process. The measuring stick for passing the review process won't be: Is this software package protecting the interests of the consumer? It'll be: Can we recover encrypted data in an amount of time less than 'X', where 'X' is some duration which the various law enforcement agencies agree is acceptable.
The review process boils down to the ability of the government to hold a companies software for ransom until they deliver a product insecure enough to please the government but secure enough not to raise too many eyebrows among users.
Consumers will feel secure because they'll see "128 bit encryption" on the box and think "128 bits, thats pretty strong stuff" not realizing that it has somehow been compromised. It's a bit like allowing PGP for export without key length restrictions so long as any digits in the key beyond 128 are 0.
The other problem is that it will probably result in the weakening of security we already have. Right now for online banking you can use 128 bit encryption because browsers with that level of encryption are not allowed to be exported. Under the new legislation browsers will be allowed to be exported after the review process. If part of the review process is sufficiently weakening the encryption so that it isn't 'too difficult' to break then there will be a browser with the strength of 56 bits of encryption masquerading as 128 bits. Having two identical products both supporting 128 bit encryption wouldn't work (one with the real deal, one with the watered down version) since interoperability is required and something that crude could be easily discerned. Either the companies won't release '128 bit' encryption to foreign countries and lose potential revenue or they'll weaken the encryption and release the weakened version in both the US and overseas. Loyalty to stock holders implies that the ultimate decision won't be favorable for privacy.
For the NSA to get involved in certain actions would be beyond the scope of their charter, it'd be illegal. The purpose of the NSA (pilfered from the NSA web page) is
The National Security Agency is the Nation's cryptologic organization. It coordinates, directs, and performs highly specialized activities to protect U.S. information systems and produce foreign intelligence information. A high technology organization, NSA is on the frontiers of communications and data processing. It is also one of the most important centers of foreign language analysis and research within the Government.
The FBI's charter however is:
The Mission of the FBI is to uphold the law through the investigation of violations of federal criminal law; to protect the United States from foreign intelligence and terrorist activities; to provide leadership and law enforcement assistance to federal, state, local, and international agencies; and to perform these responsibilities in a manner that is responsive to the needs of the public and is faithful to the Constitution of the United States.
The NSA's goal is to provide signal intelligence from foreign sources while the FBI's goal is uphold federal law and protect the US against foreign threats. They can be a consumer of information from the NSA if it relates to protecting us from foreign threats but not for residents breaking federal law.
If I'm forced to have an orginization trying to spy on my signals I'd rather have the FBI do it, they won't have near the resources of the NSA (the worlds leading employer of mathematicians). To reduce the chances of me being spyed on I avoid breaking any federal laws.
Every dollar you spend on non-Microsoft products is a vote against the corporation but its not that easy to actually do it. I haven't gone out of my way to avoid Microsoft products, I've just found that the solutions I prefer happen to not come from Microsoft. I prefer the MacOS or Linux to Windows, I prefer Frame Maker to Word, Netscape to Internet Explorer and there's nothing that I could do with Exel that I can't do easier with MatLab.
Still I've been forced to go Microsoft on two occasions. The most recent was for a piece of hardware that proudly proclaimed it was Windows, Linux and MacOS compatible right in the advertising copy. This was true as long as you were using the hardware, upgrading it required Windows. So I was forced to run out and buy a copy of Windows 98. Returning the hardware wasn't really an option due to my particular circumstances. I don't believe in situational ethics so I didn't pirate it. I am doing my best to resolve the situation with the engineers of the product in question and they agree in principle and with management approval firmware upgrades will be coming under Linux or MacOS in the near future.
The other is that for some unknown reason my company has decided that Microsoft Office is the standard application for document exchange. Us engineering folks almost always ignore it in favour of FrameMaker, but when we get things from Human Resources its in Word format. If it was just babbling I'd delete it and not worry about it, but its stuff like forms for performance reviews etc. Since FrameMaker doesn't like reading in Word files I'm forced to use the Microsoft product. In this case even if I used ApplixWare its a financial win for Microsoft since management will still license the same number of copies of Microsoft products.
Possibly, but there might be a loophole. Right now crypto software is illegal to distribute as source code. But there's nothing to stop a US citizen from downloading it from a US site or another brand of citizen from downloading it from another countries site. That still remains which makes intercontinental development efforts difficult.
For end users though some company could go through the effort of passing the encryption technology through the government review process. Say, Red Hat for instance, and make those binaries available to anybody.
I don't think this loophole will work personally because I think the review results will be: "Uhh, this encryption doesn't suck, make it suck or we'll tie you up in the review process for eternity". Of course, maybe we'd be pleasantly suprised.
This only seems to be an improvement at first gloss. It'll make it easier for companies to distribute products with strong encryption once they've passed a technical review. However consider the technical review itself for a minute. Will the government be using experts to ensure that the underlying cryptological techniques are sound and that the implementation is sound? A more likely scenario is that the technical review will center more around required backdoors or weaknesses so that the government can get in if they deem it necessary. The end result isn't easy export of strong encryption, the end result is easy export of "strong" encryption.
270 Megabits is 180 times the bandwidth of a T1. This can't be done over the internet now by a longshot except under very highly constrained circumstances. Perhaps if two hosts were directly connected to a wide pipe they would be able to do this but beyond that the chances of success are vanishingly small.
Re:To tap, or not to tap
on
CALEA update
·
· Score: 1
Capt Dan wrote: And isn't there some law about not being able to use information against you that was discovered while looking for something else?
The law exists but if the crime is worthy enough in the eyes of the law it'd be pretty easy to work around. They're surveilling you for suspicion of cocaine trafficing and find out that you're also a nefarious member of an child pornography ring. This really really annoys the feds, many of them have children. They probably can't do much right off the bat. Soon after a tip is generated that you're involved with sexually exploiting the underage and the next court order authorizes a wire/net tap that covers it, a later tap reveals the same information.
The end result is that you're in jail and you deserved it, but the law was bent or mangled to put you there.
Thanks Jan, I had actually poked around the XILINX web site before posting that but I couldn't find any information on low cost tools other than web based CPLD stuff.
I still don't feel that trying to make an Open Source microprocessor will give the best investment for the community. There are very cheap 32 bit processors on the market already that would exceed the performance of an FPGA based design, they're just not Intel so most people don't think of them.
What I'm more inclined to try doing is build hardware acceleration for various CODECS. Because these don't have to be constrained to a Von Neumann architecture I think there's a lot more potential here. (I'm assuming that the goal of an Open Sourced 32 bit microprocessor would be something to potentially run GNU software on which implies a Von Neumann architecture)
Regardless of the lack of a real Open Source license almost nobody other than rather deep pocketed corporations could make use of it in the same sense as you can make use of compiling Open Source source code. It's not even that the average person couldn't afford the fabrication costs. The average person couldn't even afford the software required to validate the design. The prices for leasing commercial Ecad software are insane, much of it's in the six digit range per user. Open Source microprocessor design at this level just won't really work. The companies that can actually afford to do anything with the technology could also afford some level of licensing fee. It's not even realistic to pretend that improvements could be circulated back to Sun in the standard Open Source fashion. The costs of the efforts to verify it would be too high without a specific product on a specific timeline.
It's still not nice to pretend its Open Source of course.
It is educationally useful, not so much for a university to build a tweaked SPARC microprocessor but to reverse-inference what architectural features were implied by various performance goals. I could see making a very excellent case study course around the technology. There's probably a lot of interesting information in the circuit design and architectural details of things like the ALU circuitry. Basically you could learn from industry experts that most universities couldn't possibly afford as instructors.
I think the coolest thing that could possibly happen as far as Open Source hardware goes is actually free (probably not as in speech) software, such as FPGA design tools. If somebody could talk XILINX or other FPGA vendor to let Joe Public make use of their tools for free (or very small nominal cost) real Open Source hardware could happen. Not CPU's, but other technologies would be possible.
Well, since I work for the company in question, I know that in the short term this will cost the company money, just as it has the numerous other times its happened. As another employee stated the news report was factually incorrect anyway. There are approximately 1500 layoffs. With the divesture of Cray Research and the Visual Workstation business unit the number goes up to that stated.
I'm not arguing that this is good news, because its not, but you're basing your opinion on an incorrect press release and random judgements pulled from thin air (to be generous)
The quality of moderation is steadily going down hill. The goal of moderation was supposed to be to promote quality, not promote agendas. Sadly more agenda promotion than content promotion has been happening. Over the past short while I've been watching what gets moderated up and what gets moderated down. Well written posts that are go along with popular opinion are being moderated up. Anything that doesn't go along with the popular opinion is either getting no net-moderation or is being moderated downards.
Moderators are letting their personal opinions get in the way of the actual goal of moderation. Meta-moderation excaberates this since the few upward moderations that unpopular views get are being pinged. When the number of people that use moderation as a bludgeon dwarf the number of people that actually moderate quality the threads become very one sided.
Also, how does Karma work (or where can I find that info)? I noticed that my Karma was reduced very abruptly (by 20 some points) and would like to understand why.
You'll be holding your breath for a long time considering we don't even have pictures of pluto's surface yet unless you count this. I have a lot more faith in a reputable scientists explanation for gravitational anomaly and such than what most people think they've seen with their eyes. To an astronomer this is seeing, though it may seem unconventional to some people. The human eye is just a sensor that happens to operate in the range of wavelengths we call visible light. A film used in an X-ray is a sensor that detects energy not visible to the human eye. Most people don't doubt doctors interpretations of X-rays. We even except more estoric sensing systems such as magnetic resonance imaging where things are mapped non-linearily into a range of colours. Given that the average doctor is much less rigorous than the average astronomer it shouldn't be much of a leap of faith to at least accept that there is a good chance that the object is a non-gas-giant planet orbiting a star.
Well, humble opinions aside, not every application in the world can be shoe horned into 2 or 4 processors and not every application will work well over a Beowulf or distributed computing model. This is true regardless of the processor used. The Alpha processor is wonderful and all but its not going to be able to make up for gross architectural deficiencies.
If a problem scales well to SMP but requires a lot of memory bandwidth between non-local processors then distributed computing will not work efficiently. Moving from an IA32 processor to a Alpha processor will only result in more cycles being spent idle waiting for data to come in across the network interface. In real high performance systems processor speed is only a small part of the equation for overall system performance if you're actually in a scenario where that high speed system is appropriate. Look up Amdahl's law in Hennesy & Patterson's Computer Architecture: A Quantitative Approach second edition and work through some of the problems.
These systems also have redundancy, fault tolerance and fault isolation built in to them. When you accidently dump your coffee into them they'll be recovering before your manager sends you on the way to your next job operating a cash register at the local McDonald's.
Linux SMP is said not to scale well largely because its all based on a single bus at the moment. It'll scale fine up to 2 or 4 processors or so but not very much farther. A number of reasonably fast processors sharing a common bus isn't a good thing when it comes to feeding those processors data. It also depends on the type of application, for something that stresses the memory interface Linux may scale less well. This isn't slamming Linux or those results, I'm just pointing out that using this as ammunition that everybodies wrong and Linux currently scales well is plain wrong. When Linux gets ported to big iron and the underlying hardware architectures involved you'll start having data that supports Linux scaling well.
Remember that compared to what a present day SGI or IBM machine can do 4 processors represents the very lowest SMP ability. That Linux has picked up the best performance in this class is great, that represents a substantial amount of potential money.
Slashdot Mirror
The New York Times is down for the count at the moment so I couldn't actually read the article. I've taken a number of online courses over the last couple of years at the graduate electrical engineering level so maybe I can submit a relavent experience.
Online courses have worked out really well for me. I work full time but am still interested in learning as much as possible about my chosen field. In the geography where I work there is no local engineering college and there may never be. About 100 miles from me is an excellent engineering college. My company is more than willing to fund any education which I can make use of (even if its sometimes rather tangential) for my job.
Travelling 200 miles 3 days a week for classes would be a big problem though. Not insurmountable but it is probably enough that it would make me think twice about taking courses. One of the perks that has been set up is distance learning through the UNITE program at this particular university. I can either attend classes in real time in a video conference or watch streaming video over my computer. Assignments can be faxed and fedexed in, midterms and finals are done in the videoconference room. So far I haven't encountered any problems that in class participants do.
I
No they got it 100% right this time. They are hackers, they're using their hacking skills for malicious purposes therefore they're also crackers. The term hacker itself is grey, there are good hackers and there are bad hackers. The problem is that in the media hackers has been used to refer to the population of hackers who operate contrary to the law as opposed to the entire population of hackers.
Even leaving it at reports of 'malicious hackers' would've been correct. They're hackers and they're malicious. It isn't implying that all hackers are malicious anymore than saying 'corrupt police officer' would imply that all police officers are corrupt.
You're assuming that source code is distributed to your friends. Chances are its binary or a driver or whatever. "Here, does this work?" as opposed to "Compile this and tell me how it works." I understand the issue, I agree that something was probably violated, but I say that this is selective enforcement. In general nobody enforces this clause because a) nobody knows when its violated b) nobody cares c) if every time a beta was released full source had to be made available certain forms of development would grind to a screeching halt (such as reverse engineered drivers for obscure closed source hardware). It happens that this publicized event involves a commercial company and so everybody is screaming about something that already happens frequently.
Fine, then Corel should give the beta testers a shiny new penny as payment. This whole argument is a bit disingenous at best. Everybody who has ever developed or worked on Open Source software has probably at some point had some code which they've tested via friends or aquaintances without making it available to the general public until obvious bugs are out. It might be as simple as "run this and tell me if it immediately crashes your computer" but the concept is the same. To a lone developer how do you define "non-internal"? Basically since there is no corporation anybody other than that developer is non-internal. This would mean that any lone developer would have to immediately do a full source code release if any other party ever was to do any testing.
It's unworkable.
The GPL is a good thing, but that doesn't mean that its not wrong some of the time. Rather than griping at Corel take a look at the problem and fix or be ready for yet another Open Source license.
The computer manufactures are trying to capture the 75% of the market that has been resistant to purchasing computers beyond a game machine like a PlayStation in the past. To do this they needed systems that weren't too much more than the typical game system or preferably (with a little deceit and trickery) the same price.
Hand me down systems would actually be fine for this market. Every system I've stopped using to upgrade to something bigger and better has been given to a family member or friend (other than my C-64 and TI/99-4A) and they're all being well used. They aren't the type of people who read mainstream computer magazines so this is just fine. In fact my most recent donation is being used to start up a graphic design business and is working wonderfully. Fast machines aren't that important for most people, most of the time at the computer is spent thinking about what to do next.
This breaks down because most people read the typical trade publication which insist that you need the fastest machine possible, the latest release of Windows and really fast 3D acceleration for the latest 3D games. People buy into it. There's truth in that statement for people who do like 3D games etc, but the target audience for these low cost systems are the people who were resistant to buying one despite the internet hype. They'll visit a few web sites, play a few games, maybe balance a cheque book and write to Aunt Sally. A few of them will get into computers in a big way and will end up purchasing a quicker computer, most of them will only use it an hour or two a week.
The problem for high end users is that these machines have either a miniscule profit margin or even a negative one. The high end machines fund the profit shortcomings of the low end ones, though most companies do absolutely nil R&D anyway.
That's pretty much what they are doing and the manufacturers of the machines are selling them at firesale prices. They're counting on a cut from the provider to make up for their losses. The only problem is that it isn't working. The manufacturers who sell these low cost systems are getting hammered (check EE times or EDN etc). I wouldn't be at all suprised to see the number of computer manufacturers thinned out a lot over the next few years.
Sure, they're losing money on each computer, but they're making up for it in volume!
Conspiracy to commit murder, attempted murder, assault. If there's no body there's no murder but there still may be a crime. If I shoot at you and miss its not murder but it is attempted murder. If I hire somebody to kill you and that person is an undercover cop no murder happens but there is a conspiracy to commit murder.
There is a similar gamut of crimes to cover near-offences of child molestation.
Just pretending to be a 13 year old girl doesn't make it entrapment. An undercover cop impersonating a prostitute in an area known to be frequented by prostitutes isn't entrapment for instance. If the undercover enticed somebody who wasn't trolling for it then it would be entrapment (I was on my way to work, I was at a stop light and this person offered to show me a good time)
It all depends on who brings up the sexual context first in this case. The fact that kiddy porn was found in his posession probably indicates that there is some sort of history that he does go out of his way for it.
Becoming a certified professional engineer sometimes is required and never hurts. In general most people who work in an engineering capacity do not require certification. They work as designers or consultants within a company or more usually a small group within a company. If you're working in the capacity of a consultant than most likely being certified as a professional engineer will be required before anybody will even talk to you. The certification is an insurance policy that you are legally responsible for your own actions and are aware of the laws governing your profession and have agreed to adhere to the ethical mandates which are part of being a licensed engineer.
What does all this mean? You are legally responsible for your own actions, much more so than the general public is (to the governing body the general public is synonymous with the unwashed masses). Any advice you give, even for free, at a bar, on a cocktail napkin must be technically sound or you can be sued. The certification body will operate against you in order to protect the profession. Even if the advice was technically sound but the person receiving the advice wasn't technically proficient to make use of it and as a result suffered financial or physical injury you may be in trouble.
If you've ever wondered why those prickish lawyers and doctors refuse to give free advice even to close personal friends then this is why. The cards are very much stacked against the licensed practicioner, to operate requires very expensive insurance policies. If you're a consultant within a consulting agency the company should provide insurance for you. Make sure to check that you're still protected after you've left the company for past work.
Sadly there's a good chance that that isn't the status quo for commercial products. With any sort of review process imposed by the government you can bet that the goal of the review is to have some way of recovering encrypted data. It may be as simple as a back door or it may be as subtle as reducing entropy during the encryption process. The measuring stick for passing the review process won't be: Is this software package protecting the interests of the consumer? It'll be: Can we recover encrypted data in an amount of time less than 'X', where 'X' is some duration which the various law enforcement agencies agree is acceptable.
The review process boils down to the ability of the government to hold a companies software for ransom until they deliver a product insecure enough to please the government but secure enough not to raise too many eyebrows among users.
Consumers will feel secure because they'll see "128 bit encryption" on the box and think "128 bits, thats pretty strong stuff" not realizing that it has somehow been compromised. It's a bit like allowing PGP for export without key length restrictions so long as any digits in the key beyond 128 are 0.
The other problem is that it will probably result in the weakening of security we already have. Right now for online banking you can use 128 bit encryption because browsers with that level of encryption are not allowed to be exported. Under the new legislation browsers will be allowed to be exported after the review process. If part of the review process is sufficiently weakening the encryption so that it isn't 'too difficult' to break then there will be a browser with the strength of 56 bits of encryption masquerading as 128 bits. Having two identical products both supporting 128 bit encryption wouldn't work (one with the real deal, one with the watered down version) since interoperability is required and something that crude could be easily discerned. Either the companies won't release '128 bit' encryption to foreign countries and lose potential revenue or they'll weaken the encryption and release the weakened version in both the US and overseas. Loyalty to stock holders implies that the ultimate decision won't be favorable for privacy.
The FBI's charter however is:
The NSA's goal is to provide signal intelligence from foreign sources while the FBI's goal is uphold federal law and protect the US against foreign threats. They can be a consumer of information from the NSA if it relates to protecting us from foreign threats but not for residents breaking federal law.
If I'm forced to have an orginization trying to spy on my signals I'd rather have the FBI do it, they won't have near the resources of the NSA (the worlds leading employer of mathematicians). To reduce the chances of me being spyed on I avoid breaking any federal laws.
Every dollar you spend on non-Microsoft products is a vote against the corporation but its not that easy to actually do it. I haven't gone out of my way to avoid Microsoft products, I've just found that the solutions I prefer happen to not come from Microsoft. I prefer the MacOS or Linux to Windows, I prefer Frame Maker to Word, Netscape to Internet Explorer and there's nothing that I could do with Exel that I can't do easier with MatLab.
Still I've been forced to go Microsoft on two occasions. The most recent was for a piece of hardware that proudly proclaimed it was Windows, Linux and MacOS compatible right in the advertising copy. This was true as long as you were using the hardware, upgrading it required Windows. So I was forced to run out and buy a copy of Windows 98. Returning the hardware wasn't really an option due to my particular circumstances. I don't believe in situational ethics so I didn't pirate it. I am doing my best to resolve the situation with the engineers of the product in question and they agree in principle and with management approval firmware upgrades will be coming under Linux or MacOS in the near future.
The other is that for some unknown reason my company has decided that Microsoft Office is the standard application for document exchange. Us engineering folks almost always ignore it in favour of FrameMaker, but when we get things from Human Resources its in Word format. If it was just babbling I'd delete it and not worry about it, but its stuff like forms for performance reviews etc. Since FrameMaker doesn't like reading in Word files I'm forced to use the Microsoft product. In this case even if I used ApplixWare its a financial win for Microsoft since management will still license the same number of copies of Microsoft products.
Possibly, but there might be a loophole. Right now crypto software is illegal to distribute as source code. But there's nothing to stop a US citizen from downloading it from a US site or another brand of citizen from downloading it from another countries site. That still remains which makes intercontinental development efforts difficult.
For end users though some company could go through the effort of passing the encryption technology through the government review process. Say, Red Hat for instance, and make those binaries available to anybody.
I don't think this loophole will work personally because I think the review results will be: "Uhh, this encryption doesn't suck, make it suck or we'll tie you up in the review process for eternity". Of course, maybe we'd be pleasantly suprised.
It's pretty much what the Anonymous Source said to the San Jose Mercury News. I'm still uncomfortable with the review process though.
This only seems to be an improvement at first gloss. It'll make it easier for companies to distribute products with strong encryption once they've passed a technical review. However consider the technical review itself for a minute. Will the government be using experts to ensure that the underlying cryptological techniques are sound and that the implementation is sound? A more likely scenario is that the technical review will center more around required backdoors or weaknesses so that the government can get in if they deem it necessary. The end result isn't easy export of strong encryption, the end result is easy export of "strong" encryption.
270 Megabits is 180 times the bandwidth of a T1. This can't be done over the internet now by a longshot except under very highly constrained circumstances. Perhaps if two hosts were directly connected to a wide pipe they would be able to do this but beyond that the chances of success are vanishingly small.
Capt Dan wrote:
And isn't there some law about not being able to use information against you that was discovered while looking for something else?
The law exists but if the crime is worthy enough in the eyes of the law it'd be pretty easy to work around. They're surveilling you for suspicion of cocaine trafficing and find out that you're also a nefarious member of an child pornography ring. This really really annoys the feds, many of them have children. They probably can't do much right off the bat. Soon after a tip is generated that you're involved with sexually exploiting the underage and the next court order authorizes a wire/net tap that covers it, a later tap reveals the same information.
The end result is that you're in jail and you deserved it, but the law was bent or mangled to put you there.
Thanks Jan, I had actually poked around the XILINX web site before posting that but I couldn't find any information on low cost tools other than web based CPLD stuff.
I still don't feel that trying to make an Open Source microprocessor will give the best investment for the community. There are very cheap 32 bit processors on the market already that would exceed the performance of an FPGA based design, they're just not Intel so most people don't think of them.
What I'm more inclined to try doing is build hardware acceleration for various CODECS. Because these don't have to be constrained to a Von Neumann architecture I think there's a lot more potential here. (I'm assuming that the goal of an Open Sourced 32 bit microprocessor would be something to potentially run GNU software on which implies a Von Neumann architecture)
Regardless of the lack of a real Open Source license almost nobody other than rather deep pocketed corporations could make use of it in the same sense as you can make use of compiling Open Source source code. It's not even that the average person couldn't afford the fabrication costs. The average person couldn't even afford the software required to validate the design. The prices for leasing commercial Ecad software are insane, much of it's in the six digit range per user. Open Source microprocessor design at this level just won't really work. The companies that can actually afford to do anything with the technology could also afford some level of licensing fee. It's not even realistic to pretend that improvements could be circulated back to Sun in the standard Open Source fashion. The costs of the efforts to verify it would be too high without a specific product on a specific timeline.
It's still not nice to pretend its Open Source of course.
It is educationally useful, not so much for a university to build a tweaked SPARC microprocessor but to reverse-inference what architectural features were implied by various performance goals. I could see making a very excellent case study course around the technology. There's probably a lot of interesting information in the circuit design and architectural details of things like the ALU circuitry. Basically you could learn from industry experts that most universities couldn't possibly afford as instructors.
I think the coolest thing that could possibly happen as far as Open Source hardware goes is actually free (probably not as in speech) software, such as FPGA design tools. If somebody could talk XILINX or other FPGA vendor to let Joe Public make use of their tools for free (or very small nominal cost) real Open Source hardware could happen. Not CPU's, but other technologies would be possible.
Well, since I work for the company in question, I know that in the short term this will cost the company money, just as it has the numerous other times its happened. As another employee stated the news report was factually incorrect anyway. There are approximately 1500 layoffs. With the divesture of Cray Research and the Visual Workstation business unit the number goes up to that stated.
I'm not arguing that this is good news, because its not, but you're basing your opinion on an incorrect press release and random judgements pulled from thin air (to be generous)
Actually layoffs reduce profits in the near term. Severance packages are expensive.
The quality of moderation is steadily going down hill. The goal of moderation was supposed to be to promote quality, not promote agendas. Sadly more agenda promotion than content promotion has been happening. Over the past short while I've been watching what gets moderated up and what gets moderated down. Well written posts that are go along with popular opinion are being moderated up. Anything that doesn't go along with the popular opinion is either getting no net-moderation or is being moderated downards.
Moderators are letting their personal opinions get in the way of the actual goal of moderation. Meta-moderation excaberates this since the few upward moderations that unpopular views get are being pinged. When the number of people that use moderation as a bludgeon dwarf the number of people that actually moderate quality the threads become very one sided.
Also, how does Karma work (or where can I find that info)? I noticed that my Karma was reduced very abruptly (by 20 some points) and would like to understand why.
You'll be holding your breath for a long time considering we don't even have pictures of pluto's surface yet unless you count this. I have a lot more faith in a reputable scientists explanation for gravitational anomaly and such than what most people think they've seen with their eyes. To an astronomer this is seeing, though it may seem unconventional to some people. The human eye is just a sensor that happens to operate in the range of wavelengths we call visible light. A film used in an X-ray is a sensor that detects energy not visible to the human eye. Most people don't doubt doctors interpretations of X-rays. We even except more estoric sensing systems such as magnetic resonance imaging where things are mapped non-linearily into a range of colours. Given that the average doctor is much less rigorous than the average astronomer it shouldn't be much of a leap of faith to at least accept that there is a good chance that the object is a non-gas-giant planet orbiting a star.
Well, humble opinions aside, not every application in the world can be shoe horned into 2 or 4 processors and not every application will work well over a Beowulf or distributed computing model. This is true regardless of the processor used. The Alpha processor is wonderful and all but its not going to be able to make up for gross architectural deficiencies.
If a problem scales well to SMP but requires a lot of memory bandwidth between non-local processors then distributed computing will not work efficiently. Moving from an IA32 processor to a Alpha processor will only result in more cycles being spent idle waiting for data to come in across the network interface. In real high performance systems processor speed is only a small part of the equation for overall system performance if you're actually in a scenario where that high speed system is appropriate. Look up Amdahl's law in Hennesy & Patterson's Computer Architecture: A Quantitative Approach second edition and work through some of the problems.
These systems also have redundancy, fault tolerance and fault isolation built in to them. When you accidently dump your coffee into them they'll be recovering before your manager sends you on the way to your next job operating a cash register at the local McDonald's.
Linux SMP is said not to scale well largely because its all based on a single bus at the moment. It'll scale fine up to 2 or 4 processors or so but not very much farther. A number of reasonably fast processors sharing a common bus isn't a good thing when it comes to feeding those processors data. It also depends on the type of application, for something that stresses the memory interface Linux may scale less well. This isn't slamming Linux or those results, I'm just pointing out that using this as ammunition that everybodies wrong and Linux currently scales well is plain wrong. When Linux gets ported to big iron and the underlying hardware architectures involved you'll start having data that supports Linux scaling well.
Remember that compared to what a present day SGI or IBM machine can do 4 processors represents the very lowest SMP ability. That Linux has picked up the best performance in this class is great, that represents a substantial amount of potential money.