Hmm - does this give anybody for a new processor rating system now that we're trying to get out of the MHz race.
NS - Not Slow VVSS1-VVSS2 Very Very Slightly Slow VSS1-VSS2 Very Slightly Slow SS1-SS2 Slightly Slow S1-S3 Slow
Re:You are Totally off there
on
Revamping Freenet
·
· Score: 2, Interesting
I can personally vouch that these are significant issues.
Speed is a design issue, and there are numerous pros/cons anonymity-wise with changing it. GNUnet is a different approach - it has its own pros/cons.
I do take issue with the java design. It doesn't even run on amd64 (since there is no stable VM on this arch) unless you run it 32-bit (using an x86 VM). It typically uses around 150MB of RAM when it is running - that is quite a bit for a single application. It also needs a several GB datastore, but that is adjustable and less of an issue (plus, all that data has to be stored somewhere).
I've run Freenet for numerous versions now (going on years at this point). For the last few months though I've tended to not keep it running. I was just getting tired of constant swapping anytime I wanted to do anything - to me a solution to a computer problem is not to keep throwing RAM chips at it, and if I did want to investing in an extra 512MB I could think of better things to do with it than using it for a single application.
Honestly, I don't think there is much excuse for an application that needs more than 5-10MB of RAM in general - especially not one that runs in the background all the time. If it needs a few dozen MB I might be able to entertain the idea.
On the other hand, when an app needs a 128MB DIMM just for its own use, we're really starting to push things...
Not to mention when you find a bug in software and instead of issuing a patch the vendor tells you to simply upgrade to the latest version. Suppose this is customized software on an enterprise scale - instead of sending you a 10K DLL file in which two lines of code changed, they're talking about potentially reintegrating your entire project...
And that is just counting stuff like the OS/Office/etc.
You have thousands of little apps, and a number of enterprise-scale apps that have all kinds of quirky licenses.
At a recent teleconference at a business I am familiar with, half the time was spent on trying to work out if the company needed more licenses as it shifted its server deployment model to a more central one. No doubt dozens of phone calls will be made, and now that the business is locked into the vendor's proprietary application, it would not surprise me if the deals are not nearly as generous as they were during the original vendor selection a few years prior.
That is a big problem with vendor lock-in. In the beginning, then need you, and deals are good. In a five years, you need them, and both deals and support tend to go downhill as a result...
Well, if it was visible at the end of class it wasn't AgNO3. That takes a few hours to show up - which is why it works so well - you don't realize you need to wash until it has already soaked in below the skin. It cannot be removed - you need to wait for skin growth to remove it...
If the FSF actually does what you suggest the most likely outcome is a gain in market share for C# at Java's expense and a fragmentation in OpenOffice that will slow its adoption.
The FSF may actually have more to lose than Sun.
Uh, how is any of this a loss for the FSF? If OO gets forked they'll be happy, and if the fork delays things they'll be able to say "I told you so." If C# kills Java they will probably care less - neither is free software.
The FSF doesn't exist to oppose Microsoft. They exist to promote free software, and oppose all proprietary solutions. Which isn't to say they oppose all commercial solutions - just proprietary commercial solutions.
Now, if all of this slows down the adoption of Gnome, KDE, Linux, GNU OS, or whatever else, then that might be a partial loss for the FSF. However, the FSF does not promote Java, and their interest in OO is to steer it away from Java.
Are you perhaps thinking of AgNO3? While I haven't washed my hands with HNO3 I hadn't heard that it causes them to turn brown. AgNO3 will certainly do so - although it is actually closer to black if you get enough on you. The delayed photoreaction is of course the key - you don't realize you need to wash it off until it is too late. That is, unless you're smart enough to wash off any chemical spill...
Actually, a more valid issue might be the lack of 3D transformations on those displays.
Making a 9 megapixel display board is probably fairly cheap if you don't need a GPU. ATC displays are generally 2D from what I've seen (which isn't a lot, I admit). Granted, the planes are flying in 3D, but the displays are generally top-down with altitude info displayed.
That being the case, you just need an old-fashioned dumb video board with lots of RAM on it (well, without textures to worry about you really only need 32MB), and fast RAM access from the CPU. Toss in BitBLT filling capability for good measure.
What makes a GPU powerful is all the 3D capability. If you can drop the 3D requirements the technology is conceptually far simpler...
I know somebody who works for a defence department contractor. He passed around the joke that there are only 10 types of people in this world - those who understand binary and those who don't. One person in the room didn't get it. Suddenly everybody understood the QA problems they were having...
Well, the cross-users filters have a lot of power that per-user filters won't have - a huge sample volume, and the fact that once 50 people say that a piece of mail is junk there is really no reason for 5000 other people to have to look at it.
However, I think that the spam/ham feedback buttons should have weights assigned. Give new users almost no weight. Increase their weight as they remain paying customers who do normal web-browsing and emailing (so that spammers can't use throwaway accounts and bots - the stats of their online habits would give them away). Then, anytime somebody gives a false report drop their rating tremendously, and do the reverse for good reports.
Now you have a system that has a good feedback system designed to work out which users it should listen to. Users need to be educated that the spam button shouldn't be used as a delete button, but if a user is too dumb to figure that out the system will quickly ignore them.
Much of this can be automated as well. When you start spamfiltering a list due to some dumb user feedback, the flood of good users that dive into their junk folders to rescue the mail can be used to discover the error.
A big ISP should be able to benefit from this sort of thing if they are smart. On the other hand, the AOL target demographic could cause them problems...
Shouldn't we just apply the same logic to getting drunk without somebody willing to assume responsibility for you? Driving drunk is a foregone conclusion as soon as you get drunk and don't have anybody present to stop you from driving. It isn't like you're going to get behind the wheel and start reasoning that you shouldn't actually turn the key...
Of course, that would lead to it being illegal to serve alcohol to anybody who isn't with a sober person, and bars wouldn't stand for that...
I'm all for stopping drunk driving. However, like most problems we're not attacking it at the root. People who are drunk don't control their own behavior. So, punishing them is rather arbitrary when you think about it. The only thing you can really justify punishing for is what the did when they were sober - and that was deciding to get drunk...
I heard a story of a student getting his hand smashed and pinned to the side of an NMR dewar by a gas cylinder. Big metallic objects and high magnetic fields are a big no-no. Even objects with small metal content can have quite noticable force exerted on them from a few feet away. Something like a gas cylnder could probably become a bullet at a range of a few feet. Oscilliscopes are occassional casulties of NMR labs for the same reason, and the cost to get the magnet back to normal after such a collision can be quite high (tens of thousands of dollars).
Hmm - back in grad school I don't think that the PI would have been happy to see students sprinting past the 600 MHz NMR. Those things are somewhat top-heavy, and I wouldn't want to be the guy who toppled the $1.5 million magnet. Nor would I want to be in the vicinity when the hundreds of liters of liquid He and N2 start to boil. Nor would I want to be the the office downstairs when it starts to rain LN2 from the ceiling...:)
I had never heard of this effect before. I guess it shouldn't be surprising. In theory any kind of a loop of conductive material could generate currents when moving through a multi-tesla field.
He made his decision as a nod to the conservative wing of his party.
What ethical reason would there be for denying individuals the medical advances that come from stem cell research?
Uh, why exactly do you think that the conservative wing of his party opposes stem cell research? For ethical reasons!
Sure, many people may disagree with the ethical judgement being made, but the decision is purely ethical. What other motivation would they have? Do you think that they're doing it just so that they can watch people with various diseases die?
In this case the ethical dilema is whether it is OK to destroy embryos to harvest their stem cells. What makes it a dilema is that it is strongly debated whether embryos are fully entitled to human rights. In fact, that is not all that different from the debate about putting human brains in sheep - is that enough to make a being with human rights? (Whoa, and suddenly we're back on-topic...)
Just because you don't happen to agree with the ethics of the situation doesn't mean that it isn't an ethical decision.
A decision to ban all animal research would also be an ethical decision, and one that many people would disagree with, but which many would also agree with.
Unfortunately, ethical problems will only be straightfoward when everybody else gets with it and just agrees that I'm the only one who really knows what is right and wrong...:)
at what point did we convince even ourselves that we could put a pricetag on life
That's silly. Everybody puts a pricetag on life. It is just wrong to actually admit you do it and publicly debate what the price should be set at.
Take environmental regulations. Your drinking water has some some limit on various heavy metals - say Cd is 5ppb or something like that (just making numbers up). Suppose 1 life per decade could be saved by lowering that to 1ppb - should we do it? Maybe that lower threshold would cost an extra $10 million per year in water bills - so that one life is being valued at $100 million. Extend it further - suppose 10ppt would save another life at the cost of $100 million per year. Now we're valuing life at $1 billion. At some point, you can't keep taxing the entire country just to save individual lives.
Roads would be safer if we limited speeds everywhere to 5mph, or if we just outright banned cars and trucks and switched to trains and bicycles. However, this would increase the costs of goods, lower the availability of labor, increase costs of living (since to work at a high-paying job you need to live within biking distance, where the houses are likely to now be more expensive), and generally discourage travel. Either that or people have to start living in apartment complexes.
Of course, that $100 million/year going into saving a life per decade with cleaner water is that much less we can spend on education, so now people are living ever so slightly longer with a much lower quality of life.
The point is that if you want to spend all your savings to save your grandmother in the hospital, that is fine. The problem comes when you expect everybody else to be compelled to do so, since life is "priceless".
In any case, society is free to develop medicines in the public domain. Why should medicine be any different than any other commercial practice? If I tell you that I'll save your dying grandmother, but only if you pay me $x, is that really wrong? Am I to be compelled to give away my time to "save" everybody at the edge of death.
It is unfortunate that people die. The pharma industry isn't selling immortality - it is selling time. For $x you get an extra y days to live. A merket for this kind of stuff is inevitable, but pushing it underground just vastly lowers it efficiency. In this case, banning drug patents will essentially eliminate all private investment in drug research, and will unemploy most of the pharma industry, and ensure that very few people study in college to become biochemists. The drugs out there now will of course become free, but the supply of new ones will be limited to whatever governments decide to research on their own. It is unlikely that the government will save all that much money in the long run doing it themselves...
If you really want to see public domain pills, perhaps a better model would be starting some UN-sponsored drug lab and allowing it to compete with the industry. That would exert cost pressure, and would generate true public-domain pills which would be funded by all nations (since everybody benefits from them). You can then directly compare the output of the public and private sectors. However, the public organization should be required to meet the same safety standards (obviously costs are lower if you just cut quality - most of the cost of drug development is clinical trials), and should not be given special liability protection (another major cost in the drug industry, which also leads to more clinical trials, and reluctance to release drugs whose risk/benefit ratio is borderline).
I think that such an experiement would be a positive one. I'm not convinced the public sector would do all that much better though. However, it would be better than just having people whine that drugs should both be free and developed using some corporation's money...
Well, besides poorly written software which requires admin to run, you also have the fact that it is often painful to quickly run an administrative task in windows with temporary privilege escalation.
Somebody mentioned writing batch files for specific programs. That is certainly a pain - what about an SUID bit? Granted, software should rarely need to be SUID, but if the programmer was dumb at least we can have one program running as admin rather than all of them.
Also - suppose I need to quickly change a setting in control panel? How can I do that without logging out of windows and back in? If you right-click on control panel you don't get a run-as option...
Run-as is a step in the right direction, but it really isn't up to the level of su/sudo or the KDE control panel, which has a quick-access admin mode button on it.
On linux I rarely log in as admin. On windows I'm basically admin all the time. It is just too much of a pain to have to log in and out all the time. Of course, I don't care if my windows PC gets hosed since I have all my important stuff on my linux server, which I back up regularly.
MS just needs to get on the ball with this and take steps to bring application writers into line. Perhaps if they made the default one where users couldn't belong to the admin group, but instead had to enter a non-trivial admin password each and every time they needed admin access companies would realize that they could no longer get away with requiring admin access. Who wants to type in a password every time their video game needs to save data to the hard drive?
Imagine P2P/bittorent with forged headers for outbound data. Boy that would slow the RIAA down.
Uh, how would a BT client know to contact your computer to download the file? It would get your IP from a tracker. The same tracker that the RIAA gets your IP from.
Spoofing packets really only helps when you're mounting attacks. When you are communicating legitimately both parties need to have some way of finding each other. Now, there are techniques like Freenet or Onion routing which would stop the RIAA, but those don't require spoofing to be effective, and in fact using spoofing wouldn't add any more security since again the side of the connection closer to the RIAA knows your IP already.
Of course, what would be the point of spending billions of US tax dollars to buy some system that anybody else in the world can also buy?
If you're an average nation that wants an average air force, then you can save money and buy F-16s, Mirages, etc. If you want your aircraft to be able to go head-to-head with anybody else on the planet and always win, then you pay to develop F-22s and forbid them from being sold elsewhere, and don't let those who work on them work for any of your competitors.
While I don't believe in government-industry subsidides, I do think that military is one area where staying local is just sensible. Now, why the US allows those contractors to use parts made overseas I have no idea - in a war they could become hard to obtain...
Dont forget, the FED creates tonnes of new money out of thin air, they could easily say, "well lets print 1000 billion in Tbills , sell em at 5% and give the cash to NASA".
Uh, the FED doesn't set the interest rates on Tbills. The Tbill-buyer does that.
Forgive me for not knowning the exact details, but the process is something like:
I have here bills that plege to pay you $10,000 in 10 years. Anybody willing to pay $10,000 for them? No? Ok, how about $9,000? And so on.
The bills are sold for the highest price available on the market. The interest rate is essentially what you get if you put a principal value of the selling price and a P+I value of $10,000 into the compound-interest formula and solving for the rate.
If the Fed just wanted to issue a trillion dollars in tbills they'd have to practically give them away since nobody has that much money to buy them.
When the government issues $1 million in savings bonds they don't get $1 million. They get somewhat less. The more they issue, the less they get compared to what they have to give back.
Slashdot Mirror
Hmm - does this give anybody for a new processor rating system now that we're trying to get out of the MHz race.
NS - Not Slow
VVSS1-VVSS2 Very Very Slightly Slow
VSS1-VSS2 Very Slightly Slow
SS1-SS2 Slightly Slow
S1-S3 Slow
I can personally vouch that these are significant issues.
Speed is a design issue, and there are numerous pros/cons anonymity-wise with changing it. GNUnet is a different approach - it has its own pros/cons.
I do take issue with the java design. It doesn't even run on amd64 (since there is no stable VM on this arch) unless you run it 32-bit (using an x86 VM). It typically uses around 150MB of RAM when it is running - that is quite a bit for a single application. It also needs a several GB datastore, but that is adjustable and less of an issue (plus, all that data has to be stored somewhere).
I've run Freenet for numerous versions now (going on years at this point). For the last few months though I've tended to not keep it running. I was just getting tired of constant swapping anytime I wanted to do anything - to me a solution to a computer problem is not to keep throwing RAM chips at it, and if I did want to investing in an extra 512MB I could think of better things to do with it than using it for a single application.
Honestly, I don't think there is much excuse for an application that needs more than 5-10MB of RAM in general - especially not one that runs in the background all the time. If it needs a few dozen MB I might be able to entertain the idea.
On the other hand, when an app needs a 128MB DIMM just for its own use, we're really starting to push things...
Not to mention when you find a bug in software and instead of issuing a patch the vendor tells you to simply upgrade to the latest version. Suppose this is customized software on an enterprise scale - instead of sending you a 10K DLL file in which two lines of code changed, they're talking about potentially reintegrating your entire project...
And that is just counting stuff like the OS/Office/etc.
You have thousands of little apps, and a number of enterprise-scale apps that have all kinds of quirky licenses.
At a recent teleconference at a business I am familiar with, half the time was spent on trying to work out if the company needed more licenses as it shifted its server deployment model to a more central one. No doubt dozens of phone calls will be made, and now that the business is locked into the vendor's proprietary application, it would not surprise me if the deals are not nearly as generous as they were during the original vendor selection a few years prior.
That is a big problem with vendor lock-in. In the beginning, then need you, and deals are good. In a five years, you need them, and both deals and support tend to go downhill as a result...
Well, if it was visible at the end of class it wasn't AgNO3. That takes a few hours to show up - which is why it works so well - you don't realize you need to wash until it has already soaked in below the skin. It cannot be removed - you need to wait for skin growth to remove it...
If the FSF actually does what you suggest the most likely outcome is a gain in market share for C# at Java's expense and a fragmentation in OpenOffice that will slow its adoption.
The FSF may actually have more to lose than Sun.
Uh, how is any of this a loss for the FSF? If OO gets forked they'll be happy, and if the fork delays things they'll be able to say "I told you so." If C# kills Java they will probably care less - neither is free software.
The FSF doesn't exist to oppose Microsoft. They exist to promote free software, and oppose all proprietary solutions. Which isn't to say they oppose all commercial solutions - just proprietary commercial solutions.
Now, if all of this slows down the adoption of Gnome, KDE, Linux, GNU OS, or whatever else, then that might be a partial loss for the FSF. However, the FSF does not promote Java, and their interest in OO is to steer it away from Java.
Better yet, try 64-bit AMD64 code.
Freenet swallows 100+MB of RAM to run in a JVM, and doesn't run at all in 64-bit, since just about every JVM is buggy as anything on that platform.
Are you perhaps thinking of AgNO3? While I haven't washed my hands with HNO3 I hadn't heard that it causes them to turn brown. AgNO3 will certainly do so - although it is actually closer to black if you get enough on you. The delayed photoreaction is of course the key - you don't realize you need to wash it off until it is too late. That is, unless you're smart enough to wash off any chemical spill...
16 +5 comments on this article, and this is the only one which was remotely serious.
Jokes are nice and all, but can we at least have a little content, mods?
Actually, a more valid issue might be the lack of 3D transformations on those displays.
Making a 9 megapixel display board is probably fairly cheap if you don't need a GPU. ATC displays are generally 2D from what I've seen (which isn't a lot, I admit). Granted, the planes are flying in 3D, but the displays are generally top-down with altitude info displayed.
That being the case, you just need an old-fashioned dumb video board with lots of RAM on it (well, without textures to worry about you really only need 32MB), and fast RAM access from the CPU. Toss in BitBLT filling capability for good measure.
What makes a GPU powerful is all the 3D capability. If you can drop the 3D requirements the technology is conceptually far simpler...
Potassium is probably nicer than cesium, but it is defintely not fun to get into an accident with.
There is a reason why Na is used in demos with water, but K is not. The energy liberated is quite high!
Uh, that was intended as sarcasm...
The mods are asleep at the wheel. -1 redundant.
Half the posts on this site look like this...
I know somebody who works for a defence department contractor. He passed around the joke that there are only 10 types of people in this world - those who understand binary and those who don't. One person in the room didn't get it. Suddenly everybody understood the QA problems they were having...
That's why /dev/hda is chmod 777 on my system. Who needs all this filesystem stuff anyway?
Well, the cross-users filters have a lot of power that per-user filters won't have - a huge sample volume, and the fact that once 50 people say that a piece of mail is junk there is really no reason for 5000 other people to have to look at it.
However, I think that the spam/ham feedback buttons should have weights assigned. Give new users almost no weight. Increase their weight as they remain paying customers who do normal web-browsing and emailing (so that spammers can't use throwaway accounts and bots - the stats of their online habits would give them away). Then, anytime somebody gives a false report drop their rating tremendously, and do the reverse for good reports.
Now you have a system that has a good feedback system designed to work out which users it should listen to. Users need to be educated that the spam button shouldn't be used as a delete button, but if a user is too dumb to figure that out the system will quickly ignore them.
Much of this can be automated as well. When you start spamfiltering a list due to some dumb user feedback, the flood of good users that dive into their junk folders to rescue the mail can be used to discover the error.
A big ISP should be able to benefit from this sort of thing if they are smart. On the other hand, the AOL target demographic could cause them problems...
I never understood the logic of this.
Shouldn't we just apply the same logic to getting drunk without somebody willing to assume responsibility for you? Driving drunk is a foregone conclusion as soon as you get drunk and don't have anybody present to stop you from driving. It isn't like you're going to get behind the wheel and start reasoning that you shouldn't actually turn the key...
Of course, that would lead to it being illegal to serve alcohol to anybody who isn't with a sober person, and bars wouldn't stand for that...
I'm all for stopping drunk driving. However, like most problems we're not attacking it at the root. People who are drunk don't control their own behavior. So, punishing them is rather arbitrary when you think about it. The only thing you can really justify punishing for is what the did when they were sober - and that was deciding to get drunk...
I heard a story of a student getting his hand smashed and pinned to the side of an NMR dewar by a gas cylinder. Big metallic objects and high magnetic fields are a big no-no. Even objects with small metal content can have quite noticable force exerted on them from a few feet away. Something like a gas cylnder could probably become a bullet at a range of a few feet. Oscilliscopes are occassional casulties of NMR labs for the same reason, and the cost to get the magnet back to normal after such a collision can be quite high (tens of thousands of dollars).
Hmm - back in grad school I don't think that the PI would have been happy to see students sprinting past the 600 MHz NMR. Those things are somewhat top-heavy, and I wouldn't want to be the guy who toppled the $1.5 million magnet. Nor would I want to be in the vicinity when the hundreds of liters of liquid He and N2 start to boil. Nor would I want to be the the office downstairs when it starts to rain LN2 from the ceiling... :)
I had never heard of this effect before. I guess it shouldn't be surprising. In theory any kind of a loop of conductive material could generate currents when moving through a multi-tesla field.
He made his decision as a nod to the conservative wing of his party.
:)
What ethical reason would there be for denying individuals the medical advances that come from stem cell research?
Uh, why exactly do you think that the conservative wing of his party opposes stem cell research? For ethical reasons!
Sure, many people may disagree with the ethical judgement being made, but the decision is purely ethical. What other motivation would they have? Do you think that they're doing it just so that they can watch people with various diseases die?
In this case the ethical dilema is whether it is OK to destroy embryos to harvest their stem cells. What makes it a dilema is that it is strongly debated whether embryos are fully entitled to human rights. In fact, that is not all that different from the debate about putting human brains in sheep - is that enough to make a being with human rights? (Whoa, and suddenly we're back on-topic...)
Just because you don't happen to agree with the ethics of the situation doesn't mean that it isn't an ethical decision.
A decision to ban all animal research would also be an ethical decision, and one that many people would disagree with, but which many would also agree with.
Unfortunately, ethical problems will only be straightfoward when everybody else gets with it and just agrees that I'm the only one who really knows what is right and wrong...
at what point did we convince even ourselves that we could put a pricetag on life
That's silly. Everybody puts a pricetag on life. It is just wrong to actually admit you do it and publicly debate what the price should be set at.
Take environmental regulations. Your drinking water has some some limit on various heavy metals - say Cd is 5ppb or something like that (just making numbers up). Suppose 1 life per decade could be saved by lowering that to 1ppb - should we do it? Maybe that lower threshold would cost an extra $10 million per year in water bills - so that one life is being valued at $100 million. Extend it further - suppose 10ppt would save another life at the cost of $100 million per year. Now we're valuing life at $1 billion. At some point, you can't keep taxing the entire country just to save individual lives.
Roads would be safer if we limited speeds everywhere to 5mph, or if we just outright banned cars and trucks and switched to trains and bicycles. However, this would increase the costs of goods, lower the availability of labor, increase costs of living (since to work at a high-paying job you need to live within biking distance, where the houses are likely to now be more expensive), and generally discourage travel. Either that or people have to start living in apartment complexes.
Of course, that $100 million/year going into saving a life per decade with cleaner water is that much less we can spend on education, so now people are living ever so slightly longer with a much lower quality of life.
The point is that if you want to spend all your savings to save your grandmother in the hospital, that is fine. The problem comes when you expect everybody else to be compelled to do so, since life is "priceless".
In any case, society is free to develop medicines in the public domain. Why should medicine be any different than any other commercial practice? If I tell you that I'll save your dying grandmother, but only if you pay me $x, is that really wrong? Am I to be compelled to give away my time to "save" everybody at the edge of death.
It is unfortunate that people die. The pharma industry isn't selling immortality - it is selling time. For $x you get an extra y days to live. A merket for this kind of stuff is inevitable, but pushing it underground just vastly lowers it efficiency. In this case, banning drug patents will essentially eliminate all private investment in drug research, and will unemploy most of the pharma industry, and ensure that very few people study in college to become biochemists. The drugs out there now will of course become free, but the supply of new ones will be limited to whatever governments decide to research on their own. It is unlikely that the government will save all that much money in the long run doing it themselves...
If you really want to see public domain pills, perhaps a better model would be starting some UN-sponsored drug lab and allowing it to compete with the industry. That would exert cost pressure, and would generate true public-domain pills which would be funded by all nations (since everybody benefits from them). You can then directly compare the output of the public and private sectors. However, the public organization should be required to meet the same safety standards (obviously costs are lower if you just cut quality - most of the cost of drug development is clinical trials), and should not be given special liability protection (another major cost in the drug industry, which also leads to more clinical trials, and reluctance to release drugs whose risk/benefit ratio is borderline).
I think that such an experiement would be a positive one. I'm not convinced the public sector would do all that much better though. However, it would be better than just having people whine that drugs should both be free and developed using some corporation's money...
Well, besides poorly written software which requires admin to run, you also have the fact that it is often painful to quickly run an administrative task in windows with temporary privilege escalation.
Somebody mentioned writing batch files for specific programs. That is certainly a pain - what about an SUID bit? Granted, software should rarely need to be SUID, but if the programmer was dumb at least we can have one program running as admin rather than all of them.
Also - suppose I need to quickly change a setting in control panel? How can I do that without logging out of windows and back in? If you right-click on control panel you don't get a run-as option...
Run-as is a step in the right direction, but it really isn't up to the level of su/sudo or the KDE control panel, which has a quick-access admin mode button on it.
On linux I rarely log in as admin. On windows I'm basically admin all the time. It is just too much of a pain to have to log in and out all the time. Of course, I don't care if my windows PC gets hosed since I have all my important stuff on my linux server, which I back up regularly.
MS just needs to get on the ball with this and take steps to bring application writers into line. Perhaps if they made the default one where users couldn't belong to the admin group, but instead had to enter a non-trivial admin password each and every time they needed admin access companies would realize that they could no longer get away with requiring admin access. Who wants to type in a password every time their video game needs to save data to the hard drive?
Imagine P2P/bittorent with forged headers for outbound data. Boy that would slow the RIAA down.
Uh, how would a BT client know to contact your computer to download the file? It would get your IP from a tracker. The same tracker that the RIAA gets your IP from.
Spoofing packets really only helps when you're mounting attacks. When you are communicating legitimately both parties need to have some way of finding each other. Now, there are techniques like Freenet or Onion routing which would stop the RIAA, but those don't require spoofing to be effective, and in fact using spoofing wouldn't add any more security since again the side of the connection closer to the RIAA knows your IP already.
Of course, what would be the point of spending billions of US tax dollars to buy some system that anybody else in the world can also buy?
If you're an average nation that wants an average air force, then you can save money and buy F-16s, Mirages, etc. If you want your aircraft to be able to go head-to-head with anybody else on the planet and always win, then you pay to develop F-22s and forbid them from being sold elsewhere, and don't let those who work on them work for any of your competitors.
While I don't believe in government-industry subsidides, I do think that military is one area where staying local is just sensible. Now, why the US allows those contractors to use parts made overseas I have no idea - in a war they could become hard to obtain...
Dont forget, the FED creates tonnes of new money out of thin air, they could easily say, "well lets print 1000 billion in Tbills , sell em at 5% and give the cash to NASA".
Uh, the FED doesn't set the interest rates on Tbills. The Tbill-buyer does that.
Forgive me for not knowning the exact details, but the process is something like:
I have here bills that plege to pay you $10,000 in 10 years. Anybody willing to pay $10,000 for them? No? Ok, how about $9,000? And so on.
The bills are sold for the highest price available on the market. The interest rate is essentially what you get if you put a principal value of the selling price and a P+I value of $10,000 into the compound-interest formula and solving for the rate.
If the Fed just wanted to issue a trillion dollars in tbills they'd have to practically give them away since nobody has that much money to buy them.
When the government issues $1 million in savings bonds they don't get $1 million. They get somewhat less. The more they issue, the less they get compared to what they have to give back.
Supply and demand...