Views presented here are quite simplistic views of what happens at a university. Most researchers are accustomed to releasing their results into the public domain for inspection, review, and improvement -- this goes for both articles and code. Remember the articles about "opening up" research journals? Most of those in favour were the researchers themselves. Furthermore, I don't understand the argument that
large companies can "dictate" what gets done with research at a university, somehow suggesting that this also extends to public funding. In general, industrial collaboration occurs under specifically worded contracts, which are usually open from the researcher's perspective -- why would a university researcher sign anything that prohibits him/her from publishing?
Besides, most often it's not some huge company locking up research, but the profs and grad students themselves who start companies based on their research (where do you think all these tech startups come from?).
I fear that the knee-jerk "open the source" response would discourage university researchers from doing commercially interesting work, because it would be more difficult for them to commercialize it on their own terms. The unintended effect would probably be to drive talented researchers into commercial labs (where the source is closed anyway), or to obfuscate their code so badly so as to make it worthless.
Furthermore, how do you define "public funding"? The professor's salary is usually subsidized by the government, so is everything the professor produces to be made public? And how do you "requrire" the open release of source? Would an auditor be sent to my lab to ensure that all my source is being released, and I'm not holding anything back?
Most research has little direct commercial value, and reserchers are normally happy to release their source. Why not leave the decision up to the researcher?
As I have said in the past though, most poor people don't give a fuck about 'software licencing'.
You're quite right. In fact most students I know (who certainly fit the definition of "poor people") demonstrate their contempt for licensing and end up spending $0 for Windows, using the CD-burner "discount". I wonder if you can justify piracy based on social justice or as a protest movement? --
people need internet access to be productive members of society, but they can't afford it legally, so they get Windows any way they can.
Linux helps, but it's not everything. Hardware is a much larger up-front cost than an OS, and there are always ongoing fees such as for internet access.
Of course he would arrest you, linux doesn't need rebooting. He might buy it if you said you were rebooting Windows... new meaning to "blue screen of death".
Are there providers in the US that supply ADSL at an unrestricted speed...
There are more serious technological challenges in ADSL. Residential phone lines are a very challenging medium for high-speed data transfer, since they were only designed to carry narrowband voice signals; transmission line effects cause major degradations in these channels. Transfer rates will depend on the quality and layout of the wire in your neighborhood, and distance from the central office. Huge speed increases are highly unlikely for the forseeable future, perhaps ever. By contrast, cable was designed to carry wideband television signals over long distances, and can therefore offer much higher potential rates. Consider that cable can carry around 100 TV channels at 5 MHz of bandwidth each; if you used a cable entirely for data, that's 500 MHz of bandwidth, which (depending on the signalling scheme) could be 1 GBit/s or more.
As a competitor to cable, what you should watch for are companies offering wireless systems that combine high speed internet, video, and local phone service... I've heard of some promising systems that may come to market soon.
Whatever you think of microsoft,.doc has become a de facto document standard, like.pdf. Pitting open source software against.doc risks marginalization. Maybe the effort should go into producing a good, free implementation of a document editor to produce.doc documents, thereby using.doc against microsoft?
ground troops
on
The Drone War
·
· Score: 5, Insightful
Before Afghanistan, conventional military wisdom held that a war
can't be won without substantial numbers of ground troops.
Firstly, that conventional wisdom was first broken in the Kosovo conflict, when Yugoslavia capitulated as a result of NATO air bombardment. Secondly, there are all kinds of ground troops on the ground in Afghanistan; not counting the small number of special forces, there are tens of thousands of Northern Alliance troops who actually captured the Taliban positions.
Unlike in the States, big compaines in Japan have a little bit of everything. Mitsubishi makes cars, trains, ships, aircraft, televisions, stereos, agricultural chemicals, food additives, synthetic rubber, molasses, canned foods, textiles, semiconductors... the list goes on. Any large project in Japan couldn't avoid being associated with a company that also makes cars.
Of course. The Americans shouldn't have developed nuclear weapons even though they had the technology to do so, and their rivals had active weapons programs. Then, once available, they shouldn't have used them, even though their use was not outside the norms of war at the time, and even though they brought the war to a prompt end. America should destroy its remaining weapons, and then there will be rainbows and bread and roses, and all of humanity can gather around the campfire to smoke pot and sing folk songs.
War is hell, period. But it's a fact of life. Get over it.
It doesn't help that AT&T gambled and lost hugely by jumping into cable broadband with both feet. As a result of that experience, most providers are probably wary of getting into the game, and most consumers probably think that broadband internet is slow and unreliable.
Of course patterns occur in random data. For example, if you toss a fair coin for a long time, you will get runs of three, four, or five heads which recur from time to time. The point is that in random, noncompressible data, the probability of occurrence for any given pattern is the same as the probability of any other pattern.
If you can get an array with a long enough baseline, and high precision in positioning, you can do interferometry with space telescopes. In principle that could give you enough resolution to spot small planets around nearby stars.
JPL is currently studying a space interferometry mission, to fly by 2009, which will fly a small interferometer to demonstrate the potential of this technique.
It used to be that saying "10 base X", where X was a number, implied that the medium was coax cable where the maximum length of the network was given by X. However, usually X was given in hundreds of feet, such as "10 base 2" or "10 base 5". This could yet be an error... haven't seen coax used in a real network for years.
To save costs, this thing was designed and built by students at Canadian universities. At the University of Toronto, a graduate student taking AER 1520H would have directly participated in the design of MOST. I have some friends who worked on this project...
I agree to some extent, but there's a little more intrinsic security in *nix... stuff like permission checking; anybody can do anything on a Windows box but only root can do the really nasty stuff on a *nix box.
You have to be a measure more clever to find a root exploit before applying your trojan payload... in fact maybe it's a good thing that Windows has low security; most crackers probably take the path of least resistance and leave *nix alone...
Let me suggest to you that useful but boring space research gets done because we also do exciting but expensive things. That is, it's hard to get the public interested in a fleet of Earth-orbiting atmospheric science satellites, but human spaceflight galvanizes the public interest enough that a few hundred million can sneak past for other, more scientifically interesting research.
I also think that the money spent on the ISS is worth it if the only thing it proves is that a massive international space project requiring detailed co-operation from former military adversaries is even possible. (PS: I'm all for letting the Chinese get on board too). The future of manned spaceflight depends on pan-national co-operation.
Why is WEP Broken?
The weakness in WEP stems back to a key derivation problem in the standard.... While the WEP standard had specified using
different keys for different data packets, the key derivation function (how to derive
a key from a common starting point) was flawed.
To all you undergrads doing math exams this week: yes, you really do have to know how to do this in the real world!
Slashdot Mirror
Technicaly it is a Brown Dwarf
Technically it is not a brown dwarf -- it would need about ten times as much mass to be officially classified as such.If the government isn't going to give me $50k to do research so I can start a company
The Canadian government gives R&D grants to private companies. I would assume the same is true of the U.S. government. So, what's the difference?
Views presented here are quite simplistic views of what happens at a university. Most researchers are accustomed to releasing their results into the public domain for inspection, review, and improvement -- this goes for both articles and code. Remember the articles about "opening up" research journals? Most of those in favour were the researchers themselves. Furthermore, I don't understand the argument that large companies can "dictate" what gets done with research at a university, somehow suggesting that this also extends to public funding. In general, industrial collaboration occurs under specifically worded contracts, which are usually open from the researcher's perspective -- why would a university researcher sign anything that prohibits him/her from publishing?
Besides, most often it's not some huge company locking up research, but the profs and grad students themselves who start companies based on their research (where do you think all these tech startups come from?). I fear that the knee-jerk "open the source" response would discourage university researchers from doing commercially interesting work, because it would be more difficult for them to commercialize it on their own terms. The unintended effect would probably be to drive talented researchers into commercial labs (where the source is closed anyway), or to obfuscate their code so badly so as to make it worthless.
Furthermore, how do you define "public funding"? The professor's salary is usually subsidized by the government, so is everything the professor produces to be made public? And how do you "requrire" the open release of source? Would an auditor be sent to my lab to ensure that all my source is being released, and I'm not holding anything back?
Most research has little direct commercial value, and reserchers are normally happy to release their source. Why not leave the decision up to the researcher?
On the subject of old but useful internet services ... whatever happened to archie? Does anybody still use it?
As I have said in the past though, most poor people don't give a fuck about 'software licencing'.
You're quite right. In fact most students I know (who certainly fit the definition of "poor people") demonstrate their contempt for licensing and end up spending $0 for Windows, using the CD-burner "discount". I wonder if you can justify piracy based on social justice or as a protest movement? -- people need internet access to be productive members of society, but they can't afford it legally, so they get Windows any way they can.
Linux helps, but it's not everything. Hardware is a much larger up-front cost than an OS, and there are always ongoing fees such as for internet access.
Of course he would arrest you, linux doesn't need rebooting. He might buy it if you said you were rebooting Windows ... new meaning to "blue screen of death".
Are there providers in the US that supply ADSL at an unrestricted speed ...
There are more serious technological challenges in ADSL. Residential phone lines are a very challenging medium for high-speed data transfer, since they were only designed to carry narrowband voice signals; transmission line effects cause major degradations in these channels. Transfer rates will depend on the quality and layout of the wire in your neighborhood, and distance from the central office. Huge speed increases are highly unlikely for the forseeable future, perhaps ever. By contrast, cable was designed to carry wideband television signals over long distances, and can therefore offer much higher potential rates. Consider that cable can carry around 100 TV channels at 5 MHz of bandwidth each; if you used a cable entirely for data, that's 500 MHz of bandwidth, which (depending on the signalling scheme) could be 1 GBit/s or more.
As a competitor to cable, what you should watch for are companies offering wireless systems that combine high speed internet, video, and local phone service ... I've heard of some promising systems that may come to market soon.
Whatever you think of microsoft, .doc has become a de facto document standard, like .pdf. Pitting open source software against .doc risks marginalization. Maybe the effort should go into producing a good, free implementation of a document editor to produce .doc documents, thereby using .doc against microsoft?
Before Afghanistan, conventional military wisdom held that a war can't be won without substantial numbers of ground troops.
Firstly, that conventional wisdom was first broken in the Kosovo conflict, when Yugoslavia capitulated as a result of NATO air bombardment. Secondly, there are all kinds of ground troops on the ground in Afghanistan; not counting the small number of special forces, there are tens of thousands of Northern Alliance troops who actually captured the Taliban positions.
Unlike in the States, big compaines in Japan have a little bit of everything. Mitsubishi makes cars, trains, ships, aircraft, televisions, stereos, agricultural chemicals, food additives, synthetic rubber, molasses, canned foods, textiles, semiconductors ... the list goes on. Any large project in Japan couldn't avoid being associated with a company that also makes cars.
Is the way to finally break the main dilemma of the hydrogen economy? (That it takes more energy to make the hydrogen than you gain in using it.)
No. In order to do that, you would have to repeal the laws of thermodynamics.
Of course. The Americans shouldn't have developed nuclear weapons even though they had the technology to do so, and their rivals had active weapons programs. Then, once available, they shouldn't have used them, even though their use was not outside the norms of war at the time, and even though they brought the war to a prompt end. America should destroy its remaining weapons, and then there will be rainbows and bread and roses, and all of humanity can gather around the campfire to smoke pot and sing folk songs.
War is hell, period. But it's a fact of life. Get over it.
Do you actually think government-administered internet would be any more efficient? Or any more permissive as far as content protection goes?
It doesn't help that AT&T gambled and lost hugely by jumping into cable broadband with both feet. As a result of that experience, most providers are probably wary of getting into the game, and most consumers probably think that broadband internet is slow and unreliable.
Of course patterns occur in random data. For example, if you toss a fair coin for a long time, you will get runs of three, four, or five heads which recur from time to time. The point is that in random, noncompressible data, the probability of occurrence for any given pattern is the same as the probability of any other pattern.
If you can get an array with a long enough baseline, and high precision in positioning, you can do interferometry with space telescopes. In principle that could give you enough resolution to spot small planets around nearby stars. JPL is currently studying a space interferometry mission, to fly by 2009, which will fly a small interferometer to demonstrate the potential of this technique.
Buffy, Angel, Twin Peaks, Iron Chef? What are you gonna do?
My choice is: CowboyNeal.
It used to be that saying "10 base X", where X was a number, implied that the medium was coax cable where the maximum length of the network was given by X. However, usually X was given in hundreds of feet, such as "10 base 2" or "10 base 5". This could yet be an error ... haven't seen coax used in a real network for years.
To save costs, this thing was designed and built by students at Canadian universities. At the University of Toronto, a graduate student taking AER 1520H would have directly participated in the design of MOST. I have some friends who worked on this project ...
It was a Python sketch about Oscar Wilde. I didn't know he actually said it.
To paraphrase an old Monty Python sketch: The only thing worse than receiving a million e-mail worms is receiving no e-mail worms.
I agree to some extent, but there's a little more intrinsic security in *nix ... stuff like permission checking; anybody can do anything on a Windows box but only root can do the really nasty stuff on a *nix box.
You have to be a measure more clever to find a root exploit before applying your trojan payload ... in fact maybe it's a good thing that Windows has low security; most crackers probably take the path of least resistance and leave *nix alone ...
Let me suggest to you that useful but boring space research gets done because we also do exciting but expensive things. That is, it's hard to get the public interested in a fleet of Earth-orbiting atmospheric science satellites, but human spaceflight galvanizes the public interest enough that a few hundred million can sneak past for other, more scientifically interesting research.
I also think that the money spent on the ISS is worth it if the only thing it proves is that a massive international space project requiring detailed co-operation from former military adversaries is even possible. (PS: I'm all for letting the Chinese get on board too). The future of manned spaceflight depends on pan-national co-operation.
From http://www.rsasecurity.com/rsalabs/index.html:
Why is WEP Broken? ... While the WEP standard had specified using
different keys for different data packets, the key derivation function (how to derive
a key from a common starting point) was flawed.
The weakness in WEP stems back to a key derivation problem in the standard.
To all you undergrads doing math exams this week: yes, you really do have to know how to do this in the real world!