We have something which can, so long as CamelotExchange wins, be easily and freely exchanged with dollars, and hence euros, yen, etc. In fact, multiple entities can compete for the sale, and so the price is set by the market. Sounds like currency trading to me, which makes these "gold pieces" currency. Not that there's anything inherently wrong with that. There are non-governmental currencies (such as eGold), and historically this was fairly common, however these days it's quite a headache.
Consider, for example:
If I earn 100 gold pieces do I have to pay income tax on them?
Since my gold pieces reside on Mythic's servers, are they a bank?
If I make a transaction over $5000 worth of gold pieces is Mythic required to report that transaction to the government?
How easily could one use this system to launder money?
I can see why Mythic would prefer that game-money stay in the game and acquire no real value outside of it, although a part of me likes the implications of it all.
Okay, okay. I'm as much for sticking it to the man as the next guy, but really, are there any uses for peer-to-peer other than for evading copyright?
Put another way, if we ignore the copyright issue for now, what situation wouldn't be better handled with some centralization?
For example, consider the various music trading protocols out there. All of the peer-to-peer systems suffer to some degree or another scalability issues. We could expect much better if we had a centralized search engine and dedicated servers, a la Google and the WWW. Naptster performed pretty well, although I think that one could argue successfully that it would have performed better if the songs themselves had been centralized.
The situation put forth by the submitter don't quite fit the same mold: we do not have world reachability. Instead we only talk to those that are near to us, which limits the possibilities considerably. Unlike the current situation in which we would like to find any matching resource if it is available, we can only hope to find them if they are available and close to us. This could be done now on the internet, and makes scalability much less of an issue. Nobody seems to want to do it, however.
I don't really see this as a reasonable situation: in 10 years we could expect to have devices that are always connected to the internet, and we would want that world-reachability. In this situation, dedicated servers are still possible and more desireable in many, if not most, cases.
Unless, of course, our motivation is simply to avoid intellectual property laws.
Give me a break. $3000 buck to listen to CDs? Watch a DVD? Monitor my (non-existant) baby?
In my kitchen?
People need to take a lesson from the success of the Palm. The Palm is a device that fills a niche well and does little else. The result is a well-tuned machine that doesn't cost too much. There are some things that I want to do in my kitchen. Optimize a device for them.
I'll consider wiring my kitchen when I can get a machine for a few hundred bucks that can serve me up recipies, access the Webtender, and check my email. If it does more, great, but form and price are key here.
Here are the specs that I'd like to see in a kitchen machine:
No more than 8"x6"x1" deep so that it will attatch under my cabinet and not get in the way. Counter space is a premium.
A small screen. Maybe 6" diagonal. I don't want the thing to get in the way too much even when I'm using it.
No keyboard. No mouse. Make it a touch screen. Counter space, remember?
Integrated 802.11. I don't want to have to recable anything.
Less than $500.
I don't see why this isn't doable now. Start with a palm, use a bigger screen, and integrate a wireless card. Voila!
Follow similar logic to create my wireless tablet, my bedside-table computer, my waterproof shower computer.
I agree that making computer companies more liable for their errors would reduce the bugginess of software and that it would increase the cost. I just doubt whether it's worth it in most cases, and I doubt whether most people would be willing for the cost.
Right now, if my word processor eats my document then I eat the cost. I protect myself by making backups. If the company who wrote my word processor was responsible, then we could expect my word processor to eat less documents and cost more. I'd rather save the money and do a backup, and I'd guess that most people would agree with me. If the software were running my car or my local neigborhood nuclear reactor I might feel differently.
More troubling is that my word processor of choice is TeX. If TeX decides to eat a document (which it hasn't done yet) then is Don Knuth responsible? Do you think that Don Knuth would have distributed this piece of software if he were?
I expect that the market will take care of the problems. In the case of the word processor, most people would be willing to live with a bug or two. In the case of online ordering software, the company deploying it might be willing to pay more to someone who will guarantee that the software works, since they could be held responsible if the software fails.
Perhaps the point is that software is quite diverse. If a program is the digital equivalent of an aeroplane, then perhaps more liability is a good thing, but the fact is that most software is more like the digital equivalent of a toaster. If the toaster malfunctions and burns a piece of toast is the manufacturer liable for the cost of the bread? If the toaster is labeled "unplug when not in use", and someone leaves it plugged in, is the company responsible for their house?
So I wonder, is my car's manufacturer liable if my car gets stolen? After all, with better security features the theif wouldn't be able to drive off with it.
Also, under the current law, there are special provisions about who is liable if my credit card gets stolen and used. They specifically limit my liability to $50. Without those special laws would the credit card company be liable, or would I? After all, with better security features, the crook wouldn't be able to use my card.
Even with automobiles we seem to require special laws to add safety features. If I have an accident and get injured, I don't think I can sue my car manufacturer as long as they've had the proper tests and include the neccessary safety equipment. Heck, not that long ago cars didn't have passenger-side airbags. It took special legislation to mandate those.
It doesn't seem to me that it's just the software industry who have limited liability when it comes to shortcomings in their products.
Look, I am not a lawyer, but if liability is responsibility for not meeting "the reasonable expectations of the consumer", then we need to know what reasonable expectations are. Right now, all of these software companies deny any responsibility for harm due to bugs in their license. If the consumer doesn't like that, they can shop around for different software, but they can't claim that they had a reasonable expectation to bug-free software.
It seems like the market is ripe for someone to step forward with some good, secure code that they're actually willing to stand behind. I'd bet that in some markets they'd get some defectors.
*** Please ignore the previous reply---it must have been a bug.;) ***
I completely agree that the market isn't there to give the providers a living, but having a large number of players generally benefits the consumer.
The second you decide that the author of the software is liable, you ensure that only large companies will be able to compete for that market. This does not include most open source, or small independent developers. If I post a small program I write to the internet, am I now liable if someone exploits a security hole in it? I find this quite troubling. I can imagine that most free software, for example, would dissapear as the authors would fear litigation.
The market should allow new competitors to enter the market, but new competitors are usually small, and wouldn't have the ability to ante enough to meet the bar. This seems to stifle competition itself, and most people agree that competition in the end yields benefits to consumers. (Heck, even Microsoft says that.)
Some consumers will want very secure software, and some will not. Perhaps if a company claims "100% Bug Free!" then they ought be liable to meet that claim. Consumers that want that will expect to pay for it, but those who don't won't. Unlike the other markets that the article discusses, we don't really have an interest in preventing folks who wan't cheap, untested software from getting it as long as they are informed about what they're getting.
Don't get me wrong---I think that Microsoft, or any other company for that matter, should be liable for the holes in its software, but I'm worried about having legal remedies for them. As Microsoft is finding out, economic remedies may be sufficient to solve the problem.
I completely agree that the market isn't there to give the providers a living, but having a large number of players generally benefits the consumer.
The second you decide that the author of the software is liable, you ensure that only large companies will be able to compete for that market, and this pushes out the little players. The market should allow new companies to enter the market, but new companies are usually small, and wouldn't have the ability to ante enough to compete. This seems to stifle competition itself, and most people agree that competition in the end yields benefits to consumers. (Heck, even Microsoft says that.)
Some consumers will want very secure software, and some will not. Perhaps if a company claims "100% bug free" then they ought be liable to meet that claim. Consumers that want that will expect to pay for it, but those who don't won't. Unlike the other markets that the article discusses, we don't really have an interest in preventing folks who wan't cheap, untested software from getting it.
A blue-ribbon panel of technology experts assembled by the National Academy of Sciences said lawmakers should consider ending Microsoft's and other software companies' special protection from product liability lawsuits, which have long forced makers of cars, medical devices and just about everything else to pay closer attention to the safety of their wares.
Of course, in the case of medical devices and automobiles, a failure can be fatal. In the case of Windows this seems unlikely. (At least until we get our WinCE pacemaker.) Instead of lives on the line we have corporate dollars.
This is the kind-of situation in which economics should be left to do its thing. If corporations view the holes as an economic hardship, they'll be more careful with the software they choose.
What makes more sense is that we hold companies more liable for the software they use---if someone hacks into Company X and steals my credit card number, they should be liable since, after all, they could choose better software to keep the data secure.
Holding the software companies liable would only ensure that only large companies could compete in the software market. This is the case with automobiles, and one could argue that many pieces of software on the market today are much more complex than a car. How then could any small company or individual compete?
In the end, this kind of legislation would likely have the effect of letting companies dodge the blame for security breaches, letting them point the finger at their software provider. Most providers, on the other hand, would be pushed out of the market due to the need for massive testing and legal departments.
How is this a win for the consumer? How is this a win for anyone but the Microsofts/IBMs/Suns of the world who can scale to provide such software?
From the article (emphasis mine): Most of the privacy rights - if there really are such things - vulnerable to a nationalized ID card have already been trampled under the wheels of increased security, more efficient law enforcement and better business long ago.
And there lies the problem.
It's too bad that the 28th amendment will probably ban flag burning instead of doing something useful.
So what if it runs linux? I mean this in a practical sense.
What impact does this have in terms of the user experience and/or the developer experience? Does having it run linux make some applications easier to write, etc.?
First, any drug company would buy the data if they had to when developing such a treatment. Celera is selling their data right now and the government data is freely available ().
Second, using your example, if they thought that some sequence controlled cancer, but in fact controlled how many balls you have, then this would come out in testing. (Pun inadvertant, but amusing.)
I don't disagree with your competition point, but your reason seems a bit strange.
Cancer drugs, and drugs in general, will still go through many rounds of testing to measure their safety and efficacy. If a drug makes it through the testing and proves its worth, how much do we care if it was developed from faulty data? In this case the ends really do justify the means. Hell, I can randomly stick atoms together, and if I come up with something that cures cancer, I'd call that a success.
Now, if I were a, say, cancer researcher, I'd want the most accurate sequence I could get, since it might make finding a useful drug much faster and easier. As a consumer, I want the researchers to have accurate sequences for just that reason, but I'm not too concerned about trusting what they come up with if it's not.
First, let's not jump to conclusions! It's still open whether factoring is NP-complete, if I recall correctly. (Although I believe that most experts think it's not. Then again, most experts think P!=NP, but that doesn't constitute proof.)
Second, you're right: Many NP-complete problems have easy cases. Take, for example, SAT. If we restrict ourselves to binary predicates, aka 2SAT, we can still things in polynomial time.
Finally, for crypto we don't really care if there are easy cases. All we need is an identifiable hard subset of the cases. RSA rests on the difficulty of factoring. If we can tell that all numbers of a certain form are very hard to factor (and there are sufficiently many of them) then we're fine---we just restrict ourselves to using numbers from the subset.
Remember, one way of thinking about the problems in NP is that we can solve them by (1) correctly guessing the solution and then (2) checking it using a polynomial-time algorithm.
Here's my NP algorithm for cypher X:
Guess the polynomial-sized key (correctly!)
Decrypt your message with the key and see if it looks right.
All ciphers I know of, and probably all useful ciphers could be broken like X. If
P=NP, then any cipher X can be broken
in polynomial time.
I think this isn't a bad move, really. We all know that a chip's speed has very little to do with its clock speed, yet the average consumer doesn't know any better. Can we blame them?
Consider what I look at when I shop for most consumer goods. When I buy a car I compare fuel miles per gallon and acceleration, when I buy an appliance I compare their efficiency ratings, etc. Most numbers advertized have a direct bearing on the performance of the product.
Compare the situation when I go to buy a computer. I see a 1.5 GHz Pentium III, a 1.1 GHz Athlon, and a 700 MHz Power PC. Which one is better? The Pentium, right? Wrong? Damned if I know. I do know which one the average consumer will think is the best.
By removing these misleading numbers, I think that AMD may be able to focus attention towards more important benchmarks. I think this is a good thing, and I hope that it doesn't blow up in their face.
Besides, the code is available for your perusal. If you think the uberspooks have put in a back door, get to work and find it!
Right. I'm no fan of the NSA, but my guess is that this is all on the level. If they were to put Evil Nasty Code into it, someone would find it, and that would be a major PR gaffe.
Who knows whether the arbitrators are biased or not? I certainly don't know what their motivations are. I think this is one of the big problems.
It seems to me that more and more we're seeing control handed over to organizations that have no direct public accountability. (Of course, I haven't been around that long.:) If an elected official does something we don't like they have to answer to the voters. (I don't doubt that they're often bought, but they still have some respect for the opinions of their constituents.) In the case of organizations like ICANN, although they accept public input, we have no direct control. This makes everyone suscpicious.
The WTO is another example. Sure they say that they have our best interests at heart, but we don't know that. We never voted for them. I think this is the biggest reason that there are so many upset about the WTO, and many more who are made very uneasy by it.
I am made nervous by unelected, powerful, rule-making agencies whose loyalties are obscure.
I'm not sure of all of the reasons in these applications, but it is true that encryption costs cycles, and most processors don't handle them very well.
RSA is very slow, for sure, which is due to the fact that they have to do a lot of arithmetic with very large numbers (1000s of bits each).
Even one-key cyphers, which don't usually make use of the large integer math, are slow on modern processors, however. The big reason for this is that encryption inherently "breaks" pipelinined and superscalar architectures, since in a good encryption scheme instructions do interact with those that directly preceeded them.
This is one reason why most commercial web sites are not encrypted, less a few "critical" pages, and why you can actually buy "SSL Accelerators" which take the SSL overhead off of the server.
I'd like to see someone start selling cheap masks of the folks on the FBI 10 most wanted list. We could wear them while walking/driving around cities that use this technology.
What's new?
on
HDTV Over IP
·
· Score: 3, Informative
What's new besides the mutlicast aspect? Hasn't this already been done.
What would really be nice it to have a PDA that could perform my public key cryptography and signing for me. Right now I have to trust whatever machine I'm using to keep my public key secure, which limits me to using machines I control. If I could carry the key in a PDA. then I could sign/encrypt anything anywhere:
Send document to PDA or create on PDA
Enter passphrase
PDA encrypts/signs
Export document
These steps could be laregely automated, and with bluetooth-like technologies, I might not even have to hook up the PDA to anything.
I think I remember reading a paper related to this, but I can't seem to dig it up. I think that the authors identified a couple problems with using the Palm for this:
The Palm was way too slow---30 seconds for a 512 bit signature, and something like 3 minutes for a 1024 bit signature, if I recall correctly.
PalmOS is unprotected---I could run my encryption, but I couldn't run any untrusted software on the Palm, since it could steal the private key.
One could also start using some sort-of digital cash system. (Probably something the government doesn't want you to have.:) Once you can do good crypto in the PDA, all sorts of fun things become possible.
I have no idea what the legal answer is, but it seems to me that morally it comes down to whether you believe it's the fault of the person who sent it, or the person who wrote Sircam.
If you blame the Sircam author, then it seems akin to publishing documents that have been stolen from a company.
If you blame the sender, then it comes down to publishig documents that they've already released, albeit accidently.
Personally, my reaction would be different in each case.
Slashdot Mirror
Who cares. There's a reason Divx died.
Consider, for example:
- If I earn 100 gold pieces do I have to pay income tax on them?
- Since my gold pieces reside on Mythic's servers, are they a bank?
- If I make a transaction over $5000 worth of gold pieces is Mythic required to report that transaction to the government?
- How easily could one use this system to launder money?
I can see why Mythic would prefer that game-money stay in the game and acquire no real value outside of it, although a part of me likes the implications of it all.Put another way, if we ignore the copyright issue for now, what situation wouldn't be better handled with some centralization?
For example, consider the various music trading protocols out there. All of the peer-to-peer systems suffer to some degree or another scalability issues. We could expect much better if we had a centralized search engine and dedicated servers, a la Google and the WWW. Naptster performed pretty well, although I think that one could argue successfully that it would have performed better if the songs themselves had been centralized.
The situation put forth by the submitter don't quite fit the same mold: we do not have world reachability. Instead we only talk to those that are near to us, which limits the possibilities considerably. Unlike the current situation in which we would like to find any matching resource if it is available, we can only hope to find them if they are available and close to us. This could be done now on the internet, and makes scalability much less of an issue. Nobody seems to want to do it, however.
I don't really see this as a reasonable situation: in 10 years we could expect to have devices that are always connected to the internet, and we would want that world-reachability. In this situation, dedicated servers are still possible and more desireable in many, if not most, cases.
Unless, of course, our motivation is simply to avoid intellectual property laws.
So, what are the advantages other than that?
In my kitchen?
People need to take a lesson from the success of the Palm. The Palm is a device that fills a niche well and does little else. The result is a well-tuned machine that doesn't cost too much. There are some things that I want to do in my kitchen. Optimize a device for them.
I'll consider wiring my kitchen when I can get a machine for a few hundred bucks that can serve me up recipies, access the Webtender, and check my email. If it does more, great, but form and price are key here.
Here are the specs that I'd like to see in a kitchen machine:
- No more than 8"x6"x1" deep so that it will attatch under my cabinet and not get in the way. Counter space is a premium.
- A small screen. Maybe 6" diagonal. I don't want the thing to get in the way too much even when I'm using it.
- No keyboard. No mouse. Make it a touch screen. Counter space, remember?
- Integrated 802.11. I don't want to have to recable anything.
- Less than $500.
I don't see why this isn't doable now. Start with a palm, use a bigger screen, and integrate a wireless card. Voila!Follow similar logic to create my wireless tablet, my bedside-table computer, my waterproof shower computer.
Lather. Rinse. Repeat.
Right now, if my word processor eats my document then I eat the cost. I protect myself by making backups. If the company who wrote my word processor was responsible, then we could expect my word processor to eat less documents and cost more. I'd rather save the money and do a backup, and I'd guess that most people would agree with me. If the software were running my car or my local neigborhood nuclear reactor I might feel differently.
More troubling is that my word processor of choice is TeX. If TeX decides to eat a document (which it hasn't done yet) then is Don Knuth responsible? Do you think that Don Knuth would have distributed this piece of software if he were?
I expect that the market will take care of the problems. In the case of the word processor, most people would be willing to live with a bug or two. In the case of online ordering software, the company deploying it might be willing to pay more to someone who will guarantee that the software works, since they could be held responsible if the software fails.
Perhaps the point is that software is quite diverse. If a program is the digital equivalent of an aeroplane, then perhaps more liability is a good thing, but the fact is that most software is more like the digital equivalent of a toaster. If the toaster malfunctions and burns a piece of toast is the manufacturer liable for the cost of the bread? If the toaster is labeled "unplug when not in use", and someone leaves it plugged in, is the company responsible for their house?
So I wonder, is my car's manufacturer liable if my car gets stolen? After all, with better security features the theif wouldn't be able to drive off with it.
Also, under the current law, there are special provisions about who is liable if my credit card gets stolen and used. They specifically limit my liability to $50. Without those special laws would the credit card company be liable, or would I? After all, with better security features, the crook wouldn't be able to use my card.
Even with automobiles we seem to require special laws to add safety features. If I have an accident and get injured, I don't think I can sue my car manufacturer as long as they've had the proper tests and include the neccessary safety equipment. Heck, not that long ago cars didn't have passenger-side airbags. It took special legislation to mandate those.
It doesn't seem to me that it's just the software industry who have limited liability when it comes to shortcomings in their products.
Look, I am not a lawyer, but if liability is responsibility for not meeting "the reasonable expectations of the consumer", then we need to know what reasonable expectations are. Right now, all of these software companies deny any responsibility for harm due to bugs in their license. If the consumer doesn't like that, they can shop around for different software, but they can't claim that they had a reasonable expectation to bug-free software.
It seems like the market is ripe for someone to step forward with some good, secure code that they're actually willing to stand behind. I'd bet that in some markets they'd get some defectors.
I completely agree that the market isn't there to give the providers a living, but having a large number of players generally benefits the consumer.
The second you decide that the author of the software is liable, you ensure that only large companies will be able to compete for that market. This does not include most open source, or small independent developers. If I post a small program I write to the internet, am I now liable if someone exploits a security hole in it? I find this quite troubling. I can imagine that most free software, for example, would dissapear as the authors would fear litigation.
The market should allow new competitors to enter the market, but new competitors are usually small, and wouldn't have the ability to ante enough to meet the bar. This seems to stifle competition itself, and most people agree that competition in the end yields benefits to consumers. (Heck, even Microsoft says that.)
Some consumers will want very secure software, and some will not. Perhaps if a company claims "100% Bug Free!" then they ought be liable to meet that claim. Consumers that want that will expect to pay for it, but those who don't won't. Unlike the other markets that the article discusses, we don't really have an interest in preventing folks who wan't cheap, untested software from getting it as long as they are informed about what they're getting.
Don't get me wrong---I think that Microsoft, or any other company for that matter, should be liable for the holes in its software, but I'm worried about having legal remedies for them. As Microsoft is finding out, economic remedies may be sufficient to solve the problem.
The second you decide that the author of the software is liable, you ensure that only large companies will be able to compete for that market, and this pushes out the little players. The market should allow new companies to enter the market, but new companies are usually small, and wouldn't have the ability to ante enough to compete. This seems to stifle competition itself, and most people agree that competition in the end yields benefits to consumers. (Heck, even Microsoft says that.)
Some consumers will want very secure software, and some will not. Perhaps if a company claims "100% bug free" then they ought be liable to meet that claim. Consumers that want that will expect to pay for it, but those who don't won't. Unlike the other markets that the article discusses, we don't really have an interest in preventing folks who wan't cheap, untested software from getting it.
Of course, in the case of medical devices and automobiles, a failure can be fatal. In the case of Windows this seems unlikely. (At least until we get our WinCE pacemaker.) Instead of lives on the line we have corporate dollars.
This is the kind-of situation in which economics should be left to do its thing. If corporations view the holes as an economic hardship, they'll be more careful with the software they choose. What makes more sense is that we hold companies more liable for the software they use---if someone hacks into Company X and steals my credit card number, they should be liable since, after all, they could choose better software to keep the data secure.
Holding the software companies liable would only ensure that only large companies could compete in the software market. This is the case with automobiles, and one could argue that many pieces of software on the market today are much more complex than a car. How then could any small company or individual compete?
In the end, this kind of legislation would likely have the effect of letting companies dodge the blame for security breaches, letting them point the finger at their software provider. Most providers, on the other hand, would be pushed out of the market due to the need for massive testing and legal departments.
How is this a win for the consumer? How is this a win for anyone but the Microsofts/IBMs/Suns of the world who can scale to provide such software?
Most of the privacy rights - if there really are such things - vulnerable to a nationalized ID card have already been trampled under the wheels of increased security, more efficient law enforcement and better business long ago.
And there lies the problem.
It's too bad that the 28th amendment will probably ban flag burning instead of doing something useful.
What impact does this have in terms of the user experience and/or the developer experience? Does having it run linux make some applications easier to write, etc.?
First, any drug company would buy the data if they had to when developing such a treatment. Celera is selling their data right now and the government data is freely available ( ).
Second, using your example, if they thought that some sequence controlled cancer, but in fact controlled how many balls you have, then this would come out in testing. (Pun inadvertant, but amusing.)
Cancer drugs, and drugs in general, will still go through many rounds of testing to measure their safety and efficacy. If a drug makes it through the testing and proves its worth, how much do we care if it was developed from faulty data? In this case the ends really do justify the means. Hell, I can randomly stick atoms together, and if I come up with something that cures cancer, I'd call that a success.
Now, if I were a, say, cancer researcher, I'd want the most accurate sequence I could get, since it might make finding a useful drug much faster and easier. As a consumer, I want the researchers to have accurate sequences for just that reason, but I'm not too concerned about trusting what they come up with if it's not.
Is it complete for the class?
Second, you're right: Many NP-complete problems have easy cases. Take, for example, SAT. If we restrict ourselves to binary predicates, aka 2SAT, we can still things in polynomial time.
Finally, for crypto we don't really care if there are easy cases. All we need is an identifiable hard subset of the cases. RSA rests on the difficulty of factoring. If we can tell that all numbers of a certain form are very hard to factor (and there are sufficiently many of them) then we're fine---we just restrict ourselves to using numbers from the subset.
Remember, one way of thinking about the problems in NP is that we can solve them by (1) correctly guessing the solution and then (2) checking it using a polynomial-time algorithm.
Here's my NP algorithm for cypher X:
All ciphers I know of, and probably all useful ciphers could be broken like X. If P=NP, then any cipher X can be broken in polynomial time.
Consider what I look at when I shop for most consumer goods. When I buy a car I compare fuel miles per gallon and acceleration, when I buy an appliance I compare their efficiency ratings, etc. Most numbers advertized have a direct bearing on the performance of the product.
Compare the situation when I go to buy a computer. I see a 1.5 GHz Pentium III, a 1.1 GHz Athlon, and a 700 MHz Power PC. Which one is better? The Pentium, right? Wrong? Damned if I know. I do know which one the average consumer will think is the best.
By removing these misleading numbers, I think that AMD may be able to focus attention towards more important benchmarks. I think this is a good thing, and I hope that it doesn't blow up in their face.
Right. I'm no fan of the NSA, but my guess is that this is all on the level. If they were to put Evil Nasty Code into it, someone would find it, and that would be a major PR gaffe.
It seems to me that more and more we're seeing control handed over to organizations that have no direct public accountability. (Of course, I haven't been around that long. :) If an elected official does something we don't like they have to answer to the voters. (I don't doubt that they're often bought, but they still have some respect for the opinions of their constituents.) In the case of organizations like ICANN, although they accept public input, we have no direct control. This makes everyone suscpicious.
The WTO is another example. Sure they say that they have our best interests at heart, but we don't know that. We never voted for them. I think this is the biggest reason that there are so many upset about the WTO, and many more who are made very uneasy by it.
I am made nervous by unelected, powerful, rule-making agencies whose loyalties are obscure.
RSA is very slow, for sure, which is due to the fact that they have to do a lot of arithmetic with very large numbers (1000s of bits each).
Even one-key cyphers, which don't usually make use of the large integer math, are slow on modern processors, however. The big reason for this is that encryption inherently "breaks" pipelinined and superscalar architectures, since in a good encryption scheme instructions do interact with those that directly preceeded them.
This is one reason why most commercial web sites are not encrypted, less a few "critical" pages, and why you can actually buy "SSL Accelerators" which take the SSL overhead off of the server.
I'd like to see someone start selling cheap masks of the folks on the FBI 10 most wanted list. We could wear them while walking/driving around cities that use this technology.
What's new besides the mutlicast aspect? Hasn't this already been done.
Hype begets hype.
These steps could be laregely automated, and with bluetooth-like technologies, I might not even have to hook up the PDA to anything.
I think I remember reading a paper related to this, but I can't seem to dig it up. I think that the authors identified a couple problems with using the Palm for this:
One could also start using some sort-of digital cash system. (Probably something the government doesn't want you to have. :) Once you can do good crypto in the PDA, all sorts of fun things become possible.
If you blame the Sircam author, then it seems akin to publishing documents that have been stolen from a company.
If you blame the sender, then it comes down to publishig documents that they've already released, albeit accidently.
Personally, my reaction would be different in each case.