Nah, they just need to distribute the code in archive form, compressed with wavelet technology. Then if you can uncompress the archive you obviously have a license to use wavelet technology, or live in a country where it's unpatentable.;-)
That's such a great idea I might have to steal it.:-) You could do it with partial hash collisions: each vote consists of the hash of the file, the rating (say 0-5 stars for the content and 0-5 stars for the encoding, to stop people from voting down well-encoded songs because they don't like the song), a random string to distinguish between votes, and an authentication string. The hash of the authentication string must match the hash of the rest of the vote in a certain number of bits.
Because each vote contains the hash of the file, votes can be distributed independently of files: if you find ten peers with the same file, you can download and combine the votes from all of them even if you only download the file from one. You can share votes that you agree with to make them more easily available, but equally you can't stop other people from sharing votes that disagree with yours.
You can avoid the Sybil attack by only connecting to people you know in real life, but obviously you lose the main advantage of a peer-to-peer network that way: the ability to find strangers and their files. However, with careful design I believe you can still communicate (and share files) with strangers across a trust network - that's what I'm attempting to do in my PhD project.
Some packets have to travel several hops over the trust network, so you have two new problems: sharing the bandwidth and finding short routes.
The first problem is solved by requiring every participant to contribute as many resources to the network as they use. You do this by charging your neighbours for forwarding their packets, and paying them to forward your packets. You're free to set the price as high as you want, and they're free to send the packets by a cheaper route, so you're in competition with their other neighbours to carry their traffic. The payment happens hop-by-hop so you don't need a digital currency, you just keep score with each of your neighbours.
The second problem (finding short routes) is solved by flooding, because that's a good way of finding the lowest-latency route in a dynamic network. But you don't want to flood the entire network because that kills scalability. Instead, anyone who wants to receive connections sends out periodic advertisement broadcasts. Each node that receives the advertisement adds an entry to its route cache. Anyone who wants to establish a connection broadcasts a search for a node with a route to the destination. If the search reaches a node that has seen the advertisement, it proceeds along the cached route to the destination. Since the broadcasts only need to overlap at one node, the diameter of each broadcast is on average half the diameter of the network, so the traffic scales according to the square root of the number of nodes, which is better than unlimited flooding but still not great.
You don't necessarily trust people more than one hop away, so you need end-to-end proof of delivery using digital signatures. When a packet is acknowledged, each node along the route updates its route cache (even the nodes that weren't in range of the original advertisement), so finding a route to a well-known destination just requires finding someone who's communicated with it recently.
A port scan is completely parallelizable, so the speed is only limited by bandwidth. Check out Paketto Keiretsu's port scanner. It uses raw sockets - no need to open a socket for each port.
Does packet radio offer an alternative to licensing frequencies? Here's my half-baked non-engineer's suggestion:
Imagine a portion of the spectrum is set aside for packet radio. Anyone can broadcast, at any power level, provided they obey the proper collision-detection rules: if you detect a collision (ie if what you hear isn't what you sent), wait for a random period before retransmitting. The backoff period increases exponentially with each subsequent collision, like a LAN.
Problem: if you have a less powerful transmitter than the person you're colliding with, it's possible that you can hear the collision but they can't, so you'll keep backing off and you won't get any air time. Solution: move to a less crowded frequency.
Instead of being identified by their frequencies, radio stations are identified by their public keys. If you come across a station that you like, hit a button and your digital radio saves the station's public key. When you want to find that station again, start at the channel where you last heard it and hop through the channels in an order derived from the station's key, until you start picking up packets that match the key.
The FCC can police the spectrum by deliberately causing collisions and checking that transmitters back off and change channels like they're supposed to. Anyone with a misbehaving transmitter is tracked down and punished, but everyone else can use the spectrum without a license.
Dear flatmates,
PLEASE remember to SHUT THE BACK DOOR before you go out, the table escaped AGAIN today and was halfway home before I caught it.
Thanx,
Vikki
Re:What about the physical characteristic changes?
on
Robocones
·
· Score: 1
Comes from Middle English, possibly from Old Norse "bolr" meaning tree trunk. Just off the top of my head, of course.;-)
Re:Drunk in charge of a bollard?
on
Robocones
·
· Score: 1
Whereas now you could just program the cones to go home on their own... and you'd be snoring contentedly by the time the police realised you'd actually programmed them to reduce the width of the M25 by one centimetre a week.
Thermoacoustic refrigeration has been a focus of research for more than a decade at Purdue and elsewhere, and has reportedly flown on the Space Shuttle...
Apparently thermoacoustic refrigeration works better in orbit because in space, no-one can hear ice cream.
The referendum is not about whether the UK stays in the EU, but some people will try to paint it as such in order to scare people into voting "yes". The actual issue is whether the UK should sign the new constitutional treaty, which will change the structure of the EU (voting arrangements etc), replacing the structure created by earlier treaties which the UK signed without a referendum.
As for anti-EU feeling in Britain, I think it's partly ignorant xenophobia, partly the fact that our loyalties are divided between the US and Europe, partly resentment of yet another layer of expensive, unaccountable, unresponsive government, and partly skepticism about whether closer economic ties with (eg) Portugal and Slovenia are really in our interest.
I agree that it's a good idea for immigrants to learn English, but that's exactly why I disagree with Trevor Phillips - multiculturalism is an important concept. To me, it means embracing aspects of another culture without abandoning your own. In other words, it's not society that's multicultural so much as those individuals within it who understand more than one language, one set of manners, one way of dressing etc. People can be multicultural just like they can be multilingual.
It's unrealistic to expect immigrants to abandon their culture when moving to a new country, and it's unrealistic to expect their children to grow up without a sense of "dual identity". Integration happens when people gain a British identity, not when they lose their foreign identity. So I don't have a problem with immigrants choosing to mix with other immigrants from the same country, to keep their language and traditions alive, etc, but I think we should encourage the second and third generation to see themselves as British as well as immigrants - which, to a large extent, I think they already do.
The scale of benefit fraud is insignificant compared to the scale of corporate tax fraud. If 80% of Britons are in favour of an ID card because they think it's going to stop benefit fraud, they are being misled. The government has chosen the benefit fraud angle because it appeals to the tabloids, who love to dig up "asylum seekers" (ie immigrants, ie non-whites) who are defrauding the taxpayer. This draws attention away from the large-scale thieves (including many newspaper owners, MPs and their friends) who earn millions in the UK but move their money offshore to avoid paying taxes.
Please explain how a national identity card for all citizens (not just benefit claimants) is going to reduce benefit fraud, when the criteria for getting a card will not be any stricter than the current criteria for getting a passport. Also, please explain how the much-vaunted biometric data will help to prevent fraud (are there going to be DNA testing stations in every dole office?). Wouldn't it be just as effective to require people to present a passport when claiming benefits, after giving those who don't have a passport time to apply for one? Why is a new nationwide biometric database necessary? Certainly not to prevent benefit fraud.
I'm constantly astounded at how badly some 2nd and 3rd generation immigrants speak English.
The Romans never learned to speak ancient British.
The Saxons never learned to speak Latin or ancient British.
The Normans never learned to speak Anglo-Saxon.
Nevertheless the languages merged, adapted and evolved into what we now call English - remnants of all the languages that came before, plus a thousand years of borrowing from French, Italian, Hebrew, Arabic, Hindi...
Don't worry about the English language, it will survive a lot longer than you.
And how would an ordinary person, or even an ordinary police officer, verify biometric information? When you present your ID card to open a bank account, are they really going to take a DNA sample, send it off to the lab and compare it with the information encoded on your card?
Biometric information will not be a practical deterrent to identity card theft and fraud. The only pieces of information that will deter theft are the pieces that can be checked by everyone - the photograph and the signature. Thus the cards will not be any more secure than passports or driving licenses. In fact the identity card will be an identity thief's dream - a single card that's accepted everywhere, from banks to benefit offices to police checkpoints, and which (unlike a passport) will routinely be carried in the owner's wallet.
If I am very angry at someone and want to say or do something harmful to them, but decide not to at the last moment, I have censored myself. That's called self-control.
That's true for an individual, but don't you think journalists have a responsibility to report the whole truth, rather than the subset of the truth that's palatable, convenient and politically neutral?
Of course you can't report everything - somebody has to decide what's worth reading about, and the choice of agenda is arguably a more powerful (because more subtle) editorial tool than simply telling a one-sided story. But shouldn't the decision be based on how significant and how verifiable a story is, rather than on how damaging it could be? What is a free press for, if not to rock the boat?
I am having a hard time trying to figure out what Google would gain from it
In standard fractional reserve banking, only a small fraction of the currency's value needs to be backed by gold. If people bought Google currency using, say, dollars, Google would make a profit equal to the face value of the currency minus the much smaller value of the reserve. A license to print money! The only difficulty would be persuading people to buy the currency, but if they wait for the next tech boom I'm sure they'll have plenty of takers.
However, if the currency is purely digital I'd expect its value to be very unstable thanks to all the Perl hackers writing scripts to speculate on the exchange rate.;-)
Self-knowledge seems like a paradox, but consider that it takes some amount of time, no matter how small, to experience a sensation - even the sensation of knowing something about yourself. So the "self" that you know something about is a version of yourself from the recent past... and the "self" that he or she knows something about is a version from the slightly more distant past. Instead of an apparently paradoxical circle you have a spiral in time. You can follow it back as far as you like, with each version's self-knowledge getting less and less acute until the ego disappears altogether in infancy.
So you can see "the I that sees me" - you just have to wait a second.;-)
Yes, you can do it, but it takes a few steps, and a little bit of knowledge. People are intrensically lazy, so aren't just going to do it the majority of the time. (Also, do you have any Idea how many people out there *can't* figure out how to write a cd?)
But the point about digital copying is that only one person has to crack it. After that it's just a question of distribution. Look at the warez scene - the big problem is finding distribution channels and staying one step ahead of the law (am I quoting the Dukes of Hazzard there?). Cracking the software in the first place is a relatively small piece of the puzzle. That's why the real battleground is not DRM, but P2P.
Just because it's free at the point of use doesn't mean it's free. The universities have to get their money somewhere - either from tuition fees or from taxes - and that money ends up in the pockets of publishers like Elsevier, who in many cases don't pay a penny to the authors or editors of the journals! Instead, the authors and editors have their salaries paid by universities, who once again get their money from tuition fees and taxes.
This is starting to change in computer science, although other fields are a long way behind. I'm studying for a PhD at the moment, and most of the papers I need are available online, either on the authors' websites or on Citeseer. Even in CS, older papers are less likely to be available, but most of the work in my area was published in the last four years or is still awaiting publication. That's the other advantage of publishing online - the process of getting a paper reviewed and published can take years, so in fast-moving fields the journals are really an archive of significant work rather than a news medium. To keep up with recent work you have to look online.
Of course, the problem with self-publication is lack of peer review. However, Citeseer does a pretty good job of finding significant papers based on the number of citations (think Pagerank), and the database of citations also helps you to find papers that might contradict or reinforce the conclusions of the paper you've just read. This makes it less important to have editors filtering out biased or unreproducible results.
I hope that authors in other fields will start to embrace online self-publication. Unfortunately, many institutions see publication count as a good measure of an academic's standing, partly because the peer review process tends to ensure that a frequently-published author is well respected in his or her field. If insitutions started to pay attention to citation count instead, self-publication would become a viable alternative to journal publication, saving students and taxpayers an awful lot of money.
Slashdot Mirror
...which suggests that the sky wouldn't fall in if heroin was available cheaply in supermarkets.
Nah, they just need to distribute the code in archive form, compressed with wavelet technology. Then if you can uncompress the archive you obviously have a license to use wavelet technology, or live in a country where it's unpatentable. ;-)
Because each vote contains the hash of the file, votes can be distributed independently of files: if you find ten peers with the same file, you can download and combine the votes from all of them even if you only download the file from one. You can share votes that you agree with to make them more easily available, but equally you can't stop other people from sharing votes that disagree with yours.
Some checksum algorithms aren't designed to be collision-resistant, but it's a key property of cryptographic hashes like MD5 and SHA-1.
Some packets have to travel several hops over the trust network, so you have two new problems: sharing the bandwidth and finding short routes.
The first problem is solved by requiring every participant to contribute as many resources to the network as they use. You do this by charging your neighbours for forwarding their packets, and paying them to forward your packets. You're free to set the price as high as you want, and they're free to send the packets by a cheaper route, so you're in competition with their other neighbours to carry their traffic. The payment happens hop-by-hop so you don't need a digital currency, you just keep score with each of your neighbours.
The second problem (finding short routes) is solved by flooding, because that's a good way of finding the lowest-latency route in a dynamic network. But you don't want to flood the entire network because that kills scalability. Instead, anyone who wants to receive connections sends out periodic advertisement broadcasts. Each node that receives the advertisement adds an entry to its route cache. Anyone who wants to establish a connection broadcasts a search for a node with a route to the destination. If the search reaches a node that has seen the advertisement, it proceeds along the cached route to the destination. Since the broadcasts only need to overlap at one node, the diameter of each broadcast is on average half the diameter of the network, so the traffic scales according to the square root of the number of nodes, which is better than unlimited flooding but still not great.
You don't necessarily trust people more than one hop away, so you need end-to-end proof of delivery using digital signatures. When a packet is acknowledged, each node along the route updates its route cache (even the nodes that weren't in range of the original advertisement), so finding a route to a well-known destination just requires finding someone who's communicated with it recently.
Here's a paper from Columbia University on using GPUs to accelerate cryptographic calculations.
A port scan is completely parallelizable, so the speed is only limited by bandwidth. Check out Paketto Keiretsu's port scanner. It uses raw sockets - no need to open a socket for each port.
Imagine a portion of the spectrum is set aside for packet radio. Anyone can broadcast, at any power level, provided they obey the proper collision-detection rules: if you detect a collision (ie if what you hear isn't what you sent), wait for a random period before retransmitting. The backoff period increases exponentially with each subsequent collision, like a LAN.
Problem: if you have a less powerful transmitter than the person you're colliding with, it's possible that you can hear the collision but they can't, so you'll keep backing off and you won't get any air time. Solution: move to a less crowded frequency.
Instead of being identified by their frequencies, radio stations are identified by their public keys. If you come across a station that you like, hit a button and your digital radio saves the station's public key. When you want to find that station again, start at the channel where you last heard it and hop through the channels in an order derived from the station's key, until you start picking up packets that match the key.
The FCC can police the spectrum by deliberately causing collisions and checking that transmitters back off and change channels like they're supposed to. Anyone with a misbehaving transmitter is tracked down and punished, but everyone else can use the spectrum without a license.
Your megahertz may vary?
Dear flatmates,
PLEASE remember to SHUT THE BACK DOOR before you go out, the table escaped AGAIN today and was halfway home before I caught it.
Thanx,
Vikki
Comes from Middle English, possibly from Old Norse "bolr" meaning tree trunk. Just off the top of my head, of course. ;-)
Whereas now you could just program the cones to go home on their own... and you'd be snoring contentedly by the time the police realised you'd actually programmed them to reduce the width of the M25 by one centimetre a week.
Apparently thermoacoustic refrigeration works better in orbit because in space, no-one can hear ice cream.
As for anti-EU feeling in Britain, I think it's partly ignorant xenophobia, partly the fact that our loyalties are divided between the US and Europe, partly resentment of yet another layer of expensive, unaccountable, unresponsive government, and partly skepticism about whether closer economic ties with (eg) Portugal and Slovenia are really in our interest.
It's unrealistic to expect immigrants to abandon their culture when moving to a new country, and it's unrealistic to expect their children to grow up without a sense of "dual identity". Integration happens when people gain a British identity, not when they lose their foreign identity. So I don't have a problem with immigrants choosing to mix with other immigrants from the same country, to keep their language and traditions alive, etc, but I think we should encourage the second and third generation to see themselves as British as well as immigrants - which, to a large extent, I think they already do.
Please explain how a national identity card for all citizens (not just benefit claimants) is going to reduce benefit fraud, when the criteria for getting a card will not be any stricter than the current criteria for getting a passport. Also, please explain how the much-vaunted biometric data will help to prevent fraud (are there going to be DNA testing stations in every dole office?). Wouldn't it be just as effective to require people to present a passport when claiming benefits, after giving those who don't have a passport time to apply for one? Why is a new nationwide biometric database necessary? Certainly not to prevent benefit fraud.
The Romans never learned to speak ancient British.
The Saxons never learned to speak Latin or ancient British.
The Normans never learned to speak Anglo-Saxon.
Nevertheless the languages merged, adapted and evolved into what we now call English - remnants of all the languages that came before, plus a thousand years of borrowing from French, Italian, Hebrew, Arabic, Hindi...
Don't worry about the English language, it will survive a lot longer than you.
Biometric information will not be a practical deterrent to identity card theft and fraud. The only pieces of information that will deter theft are the pieces that can be checked by everyone - the photograph and the signature. Thus the cards will not be any more secure than passports or driving licenses. In fact the identity card will be an identity thief's dream - a single card that's accepted everywhere, from banks to benefit offices to police checkpoints, and which (unlike a passport) will routinely be carried in the owner's wallet.
That's true for an individual, but don't you think journalists have a responsibility to report the whole truth, rather than the subset of the truth that's palatable, convenient and politically neutral?
Of course you can't report everything - somebody has to decide what's worth reading about, and the choice of agenda is arguably a more powerful (because more subtle) editorial tool than simply telling a one-sided story. But shouldn't the decision be based on how significant and how verifiable a story is, rather than on how damaging it could be? What is a free press for, if not to rock the boat?
In standard fractional reserve banking, only a small fraction of the currency's value needs to be backed by gold. If people bought Google currency using, say, dollars, Google would make a profit equal to the face value of the currency minus the much smaller value of the reserve. A license to print money! The only difficulty would be persuading people to buy the currency, but if they wait for the next tech boom I'm sure they'll have plenty of takers.
However, if the currency is purely digital I'd expect its value to be very unstable thanks to all the Perl hackers writing scripts to speculate on the exchange rate. ;-)
Personally I can't help believing in free will, but if you want to be a fatalist that's your choice. ;-)
So you can see "the I that sees me" - you just have to wait a second. ;-)
So are there formal versions of number theory in which Godel's theorem can't be proved?
But the point about digital copying is that only one person has to crack it. After that it's just a question of distribution. Look at the warez scene - the big problem is finding distribution channels and staying one step ahead of the law (am I quoting the Dukes of Hazzard there?). Cracking the software in the first place is a relatively small piece of the puzzle. That's why the real battleground is not DRM, but P2P.
This is starting to change in computer science, although other fields are a long way behind. I'm studying for a PhD at the moment, and most of the papers I need are available online, either on the authors' websites or on Citeseer. Even in CS, older papers are less likely to be available, but most of the work in my area was published in the last four years or is still awaiting publication. That's the other advantage of publishing online - the process of getting a paper reviewed and published can take years, so in fast-moving fields the journals are really an archive of significant work rather than a news medium. To keep up with recent work you have to look online.
Of course, the problem with self-publication is lack of peer review. However, Citeseer does a pretty good job of finding significant papers based on the number of citations (think Pagerank), and the database of citations also helps you to find papers that might contradict or reinforce the conclusions of the paper you've just read. This makes it less important to have editors filtering out biased or unreproducible results.
I hope that authors in other fields will start to embrace online self-publication. Unfortunately, many institutions see publication count as a good measure of an academic's standing, partly because the peer review process tends to ensure that a frequently-published author is well respected in his or her field. If insitutions started to pay attention to citation count instead, self-publication would become a viable alternative to journal publication, saving students and taxpayers an awful lot of money.