Let me explain more clearly, because it seems that some of the moderators didn't understand my comment.
Think about a photon, which has a linear polarization: up-down, left-right, slantwise, or at whatever angle you want. You can in principle put in an arbitrary amount of information in setting the polarization angle of a photon. You could divide a circle into as many parts as you want, and set the polarization to an angle corresponding to the value you want to send. This is like how they pack 1024 bits into a 19 nuclei molecule.
Now, the problem is reading the data back out. If you have only one photon in a particular polarization state, you can't determine that state with any accuracy. You can in fact only get one bit of data out of that photon. You can pass it through a polarizer and either it makes it, or it does not. This gives you information about the polarization state but it destroys that state in the process. You can put lots of information into a single photon, but you can't read it back out.
Now let's imagine that we have lots of photons, in a laser beam for example. We can set them all to the same polarization state. Now we can read the polarization quite exactly, by using large numbers of photons and turning our polarizing detector until we get a peak in the output.
Even though all the photons are in the same state (like in the NMR molecule experiment), it is because there are large numbers of them that we can read the state back out accurately. We would NOT be able to read back the data from a single photon, and in the same way we would NOT be able to read back the data from a single molecule.
Hopefully that explains my comment above. A qubit, whether photon polarization or nuclear spin, holds only a limited amount of information, and you can't read more out than it holds. There's no way you can get 1024 bits into 19 nuclei, and no one should try to "spin" the results of this experiment that way.
Grover writes that we're limited to "reversible" gates, i.e. those for which one can infer the inputs from the outputs... but it seems that the pigeonhole principle would prevent any gate with more than one input from being reversible
You can turn any irreversible gate into a reversible one by adding outputs. For example, 2-input XOR becomes reversible if you add a 2nd output which is a copy of one of the inputs. There are many tricks like this which are studied under the topic of reversible computation.
AFAIK, there are a grand total of three sorts of problems that it is known can be solved more efficiently by quantum computers: integer factorization/discrete log (solved by the same algorithm, and both mainly of use in cracking public-key cryptography), brute-forcing symmetric key cryptography (easily defeated by doubling the size of the key), and quantum physics simulations. Outside that, they are completely useless.
It's application number 2 on your list that could be important. You're talking about Grover's algorithm, but it is good for much more than brute forcing crypto keys!
Grover's algorithm can be used for any kind of searching problem. Take AI for example. Most AI problems can be expressed as searching through some abstract space for a solution. Grover's algorithm will halve the depth of the search trees. Chess computers that can look 6 moves deep today will be able to look 12 moves deep on a quantum computer, and so on.
Many computationally limited computer problems can be expressed in the form of searching for a solution. Let's even look at a problem near and dear to/.ers, graphics rendering. In some forms this can come down to "is there a nearby object blocking the view of this distant one?" That means doing a search. Theoretically, quantum computers could speed up these kinds of searches, thereby improving graphics rendering and making the games of the future even better. Plus they'll have better AI.
So if you're like Goonie, bored by improvements in scientific modelling and our understanding of the universe, surely you'll get excited about having better computer games.
Second, it doesn't work, at least not the way they say it does. You can't store 1024 bits in the nuclear magnetic spins of a 19 atom molecule!
Or more precisely, you can't retrieve that many bits. The spin state of a nucleus can be described by a complex number, but when you do a measurement you only get one bit out. With 19 nuclei you can read out only about 19 bits.
So how do they make it work? They've got a huge number of molecules there. Each one is loaded with the same data value. Using the redundancy in those molecules, the researchers can read out the 1024 bits. But if they had only a single molecule holding the value in its nuclear spins, as the paper implies, there's no way they could read out 1024 bits. So the density is not as high as they make it sound.
His predictions may make sense technologically, but not socially. Your appliances might be on the Net, but they will only be accessible to you. Your refrigerator will notify you by email or equivalent when it needs service, not the manufacturer. People don't want their appliances talking to strangers. The owners will want to be in charge.
Likewise when checking out in the store, you will need to explicitly authorize the purchase, otherwise you could contest it later. You will be presented with an itemized receipt based on a scan of your items, and you're supposed to look it over quickly and make sure there's nothing on there that you didn't buy. Then you do something to initiate the payment. You can't let people take money out of your account without giving approval! There has to be some action you take to explicitly authorize a certain amount to be transfered.
With the "presence", again you will have much more control over it than he implies. You will be able to say who can find out how you are reachable. You can have filters that automatically email you when your voice mail comes in, etc., so that people with different levels of access don't necessarily know how much priority they're getting. That way you don't offend people.
As far as ubiquitous TiVo, it depends on the outcome (both legislative and technical) of the copyright wars. You may be able to record the shows only under the control of strict DRM software that won't FF through the commercials (like the way DVD players won't FF through the FBI warnings now).
One additional social/technical prediction I'd make is more use of webcams for business meetings, creating the virtual office. Assuming that terrorism scares keep happening, people will prefer not to travel so much, and employees will want to stay home and not come into the dangerously concentrated population areas downtown. We'll see a continued trend towards white collar workers using live video feeds to communicate with their co-workers both locally and around the world.
Re:SIMs as experiment
on
Virtual Simerica
·
· Score: 2, Interesting
That story doesn't really make sense. In the "original" timeline, before anyone came back from the future, everything was hunky-dory. So who came back to start the assassinations and mess everything up?
Anyway it kind of reminds me of Orson Scott Card's story about Columbus. They set up a time machine to convince him not to discover America because it set us on the path to environmental destruction. Only it turned out that Columbus had already been diverted to the Americas by an earlier time machine, because in the original timeline he'd conquered the Moslems in a Crusade and that had caused all kinds of problems on its own.
I met Bernstein briefly, and he seemed like a nice guy in person. He's relatively young, 30-ish, and soft spoken. But online he comes off as some kind of know-it-all curmudgeon.
Personally I liked the suggestion in the Usenet thread to return expired DNS cache data when the authoritative servers are unreachable, at least as an option. 99% of the time when you can't do a host lookup, the old cached data would still be right. All the DNS purists hated the idea of using expired data, like it's unclean or something. But if it's all you've got, isn't it better to use old information than to give up on letting the net work at all?
I mean think about it, where are the self assemblers going to get their energy and raw materials from?
In many proposals, assemblers are made largely of carbon in the form of solid crystals. You may have heard of it, it's called diamond. Believe it or not, diamond is actually a more stable form of carbon, which is why it forms spontaneously. Normally it's very slow to form, but there are lots of paper designs for how assemblers can build diamond shapes very quickly.
So where does the carbon come from, and the energy to form diamond? Both can come from biomatter. It's full of carbon. You're full of carbon. And because diamond is so low energy, the nanomachine actually gains energy by turning biomass into copies of itself.
In short, you are the ideal food for nanotech. It finds all the elements it needs, and all the energy, just by eating you.
We eat living matter for its energy and elements, and nanotech will, too. Only nanotech has teeth made of diamond crystals, it can be ultimately voracious, and our immune systems won't be able to touch it.
The problem of unconstrained nanotech replicators is real. Don't succumb to the head-in-the-sand belief that such things won't exist or won't be dangerous. Your body would have no more chance against a swarm of nanotech assemblers than a mouse in a food processor. You are tasty and tender, and don't forget it.
In his very first book, Engines of Creation, available online, Eric Drexler laid out the possible consequences of attempts to suppress nanotech research. See chapter 12 especially.
He describes an ambitious program which will allow nanotech to be developed safely, via active shields to protect the environment and sealed assembler labs to allow safe experimentation.
Of course Drexler was far, far ahead of his time, but his analysis should be a starting point for any consideration of the prospects for nanotech development.
When a user of an internet service provider posts copyright-infringing data, the ISP is obligated to take down the postings once it is notified of a claim by the copyright holder. This is part of the DMCA. Then there is a mechanism for the user to challenge the removal, saying that the removed material was not copyrighted. This is discussed under section 512(c) of the DMCA link above.
It's a different part of the DMCA than what we often discuss, like breaking copyright protection measures, which is section 1201.
Now, I agree that price information, especially if reformatted, cannot be copyrighted. Copyright covers a particular expression of the data, and not the raw data itself. So IMO the sites could safely ignore these warnings. But as someone who was recently sued for several billion dollars, I can certainly understand the reluctance to enter the legal system.
How does encrypting a user's files with a key, and then denying the key to the user improve security ?
The idea is that only the program which encrypted the file can decrypt it. The file is somehow "locked" to the hash of the program that was running when it was encrypted.
The reason this increases security is two-fold. First, if some malicious program is run, a trojan or a virus infected program, it can't access the data. Imagine that the data is your bank account PIN or credit card number. Right now an Outlook virus could look through your files and find the data. With Palladium this can't happen, because only the banking program can unlock those files.
Second, if the banking program itself gets infected with a virus, its hash will change, and it won't be able to unlock its own files any more! So even a virus which targets the financial software won't be able to steal its sensitive data. In this way, Palladium provides security against a much wider range of malware infections than is possible in computers today.
Another example of improving security would be a multi-player game or P2P system where keeping people from cheating is desirable. In that case it's to everyone's advantage that they can't run a rogue game client or change their data files. By giving up the ability to cheat in this way, the user ultimately gains security because he can participate in the system free from other people cheating as well.
Usually you have to give up something to get something. Giving up the ability to cheat and to break your promises can lead to real gains. If you are offered a download of some data only in exchange for promising not to share it with anyone else, Palladium allows you to make that promise in such a way that you can't go back on your word. This will eventually lead to more valuable data being made available than would be possible in world where cheating is easy and unpreventable.
1. How does Microsoft plan to allow non-proprietary Operating Systems access to Paladium media?
It's not clear what you mean by Palladium media. If you read the linked article, you see that Palladium has four components. None of them refer to Palladium media per se.
What they do have is attestation, which lets a remote server reliably determine that you are running WMP or some other DRM compliant software before you download. Then WMP can enforce whatever restrictions are specified in the data file. So you could call the media supplied by such a server "Palladium media", and chances are that no, the server won't give you the data if you're not running Palladium - but that's entirely up to the server operator. You can't force him to do what you want, and you can't fool him, thanks to Palladium.
2. Why would consumers want to purchase your product that removes rights they have over their own media?
Now, this doesn't make sense. It's not their own media! The data is on a server belonging to someone else. Palladium gives that server owner more information in deciding whether to let you download it. It allows the server to make sure you're running some software that will follow certain rules. If not, it won't give you the data.
So nobody is taking away rights over your own media. Anything you have today, you can continue to use. What Palladium does is let people decide whether to give you their media, and to do so only if you in effect agree to follow their rules.
In answer to your question about why consumers would want to purchase Palladium computers, the answer is obvious. Server operators won't give the data to people who don't have Palladium. So owning a Palladium computer will be the only way to get entertainment media in the future. Nobody's going to force you to buy one. But some (not all) content creators will refuse to give their content away unless you are running Palladium so that they can be confident that you won't steal their data.
You know what? They don't just look for e-mail addresses to send mail to. They also use the e-mail addresses as reply-to addresses. I found this out when I got an email from a guy who was puzzled by my auto-responder emailing him. It turns out that somebody sent a message to me and used his address as a reply-to address.
This might be due to the Klez virus or a variant. It forges the From address in email, using a random address from the victim's address book. So if someone has Alice and Bob in their address book and they get infected, they may send mail to Alice that claims to be from Bob. Here's a Wired article with more information.
There is nothing in the current motes that can not be miniaturized. In three years this demo will be done with a 6" aircraft, and millimeter-scale sensor nodes.
The majority of tracks on the CD are also often *worthless*. Just let me download the songs I like and pay a reasonable amount per song!
You think that's something new? That's always been true. It can't explain why CD sales are down. Records have had filler since the beginning. Even the old 45 RPM singles had an A side and a B side.
And the fact is, some music is more accessible than others. Some songs are instant hits, the first time you hear them you like them. Other songs take a few hearings before you appreciate them. I know I've bought albums for one or two hits, but after I've listened to the CD half a dozen times I like several of the other songs just as much.
But if they do start selling individual songs, most people will only buy the hits. Without being more or less forced to listen to the other ones because of the album format, they'll never get past that accessibility barrier. This means that a typical artist will only sell one or two songs where they now sell a whole album. And since we all know that CD prices are not based on manufacturing/distribution costs, this means that the sellers will have to charge almost as much per song as they do for a whole CD now.
In other words, for the record companies and everyone involved to continue to be as profitable as they are today, they'll have to charge probably five to ten dollars per song downloaded! That's just basic economics based on the number of hit songs per CD, and based on where the costs are for a record company (most of which won't be reduced by online distribution).
That's the reality. I hope you consider that a "reasonable amount" to pay for a song, because that's what it costs to create them.
Music sales are down for the simple reason that there is very little music these days that inspires folks to run out and get the album.
So what makes you an authoritative judge of good music? Why should we care about your tastes? Do you really think music was so great in other eras, and that's more than just nostalgia talking?
Remember those great bands from 1980? Like KC and the Sunshine Band? The Captain and Tenille? Boy, if only we had groups like that around today, people would just be rushing out to buy those CDs, wouldn't they? Andy Gibb, Christopher Cross, Air Supply, Olivia Newton-John, the great names just roll on and on.
Face it. Your taste in music is no more authoritative or absolute than anyone else's. I'm tired of people saying that modern music is no good. The same thing has been said every single year since at least 1960, by nostalgic old fogies who can't stand the junk that those darn kids listen to nowadays.
If you're in a hurry to reach your crotchety old age, fine. But don't pretend this is an excuse for why people are buying less music. There's been good and bad music throughout history. No generation had a monopoly on the good stuff. Popular music today is called "popular" because people like it! And that mostly means young people. The fact that it's not to your liking says more about you than about the quality of the music.
"Reduce the cost of new CDs; if discs cost only a few dollars each, buying them might be preferable to spending the time and effort to make copies or find them online."
The point being missed is that the cost of a CD has to cover much more than just the manufacturing and distribution of that disk. It has to cover the expenses in finding and developing talent, recording and touring, marketing and advertising. These expenses far exceed the costs of pressing a CD.
Most bands are commercial failures, too, so the few successful acts have to be priced high enough to cover the money lost on the others. It's like the oil business where you have to drill 20 dry holes for every one that hits the black gold.
I don't see anyone, not indy labels, not bands selling their own music, who sells CDs for a few bucks each. If the big, corrupt labels were engaging in profiteering, then everyone else would sell their disks for three bucks, right? Well, that's not happening. I think this shows that the people who think this would be a reasonable price for a CD are missing some basic economics.
Seriously, I often have vi running in the left window and emacs in the right hand one. It's a good mental exercise to switch back and forth between them frequently. I wish I could train myself to use my right hand for emacs and the left for vi, but I'm not there yet. Maybe I could do it with two chord keyboards?
More ambitious multiverses
on
One of Many
·
· Score: 3, Interesting
Max "Mad Max" Tegmark has a more ambitious multiverse theory.
It goes way beyond inflation, black holes, and even the Many Worlds Interpretation of quantum mechanics.
His idea is that all possible mathematical structures exist, and that we live in one of them! At some level, physics can be considered a branch of mathematics. Hence our universe can be considered as an enormously complicated mathematical structure. The question is, why this structure instead of some other?
His answer is that all mathematical structures exist, but that most of them are unsuitable for life. The paper linked above analyzes many different possibilities in terms of numbers of dimensions, numbers of time dimensions (yes, you could conceive of a universe with two-dimensional time), various other parameters, and he shows that structures that we would think of as living would have a hard time existing in universes much different from our own.
The Tegmark model can be thought of as the simplest possible physical theory. If physics is reducible to mathematics, then saying that all mathematical structures exist can be put more simply: Everything exists.
A similar model based on computation is proposed by Juergen Schmidhuber. Rather than Tegmark's mathematical structures, Schmidhuber proposes that all computations exist. Given that any mathematical model of a universe can be simulated by a computer program, these two formulations are roughly equivalent.
But Schmidhuber's approach has the advantage that it provides a natural way to say that some universes are more probable than others: namely, universes with short programs have more "measure" than universes with long programs. It follows that our universe probably has a relatively short program, which therefore explains why we observe that physical laws are mathematically simple.
It's pretty heavy stuff, but certainly exciting to see that researchers are (somewhat reluctantly) beginning to entertain multiverse models. The more ambitious "everything exists" theories are still too extreme for the mainstream, but I suspect that they, too, will get increasing attention over the next few decades.
It's ironic that Freenet, a tool for free speech and unfettered communication, is released under the GPL, a license which is designed to put restrictions on what you can do and say.
Under the GPL, if you distribute GPL'd software you are obligated to make the source available. Whether this is good or bad is not the point here. The point is that it is a limitation, a rule, that you must follow.
But Freenet is designed to free us from such limitations and rules. It allows us to publish information without regard to any restrictions that some third party wants to impose - even restrictions imposed by the GPL.
In short, using Freenet one could take GPL software, modify it and redistribute it without making the source code available. And you could get away with it. No one would be able to track you down and make you stop. Freenet makes it easy to bypass the GPL.
Granted, it would be hard to get paid for such software. Some people claim that the real point of the GPL is just that, to prevent people from getting paid for software (not that GPL advocates would ever admit it!). But it seems inconsistent at best for software designed to provide complete freedom of speech to be released under a restrictive license.
The first thing someone should do is to make a copy of the Freenet code with all the GPL licensing terms stripped out of the source, change it to a license that has no restrictions, and publish it on Freenet. That will truly demonstrate the nature of the beast.
Microsoft(tm)'s Palladium scheme will require signed drivers. There is simply no way to try to enforce that level of security while still allowing end users to insert arbitrary code into the kernel. Not with any standard definition of "driver" and "kernel".
That's not true. According to what Microsoft has said, Palladium runs "beside" Windows. It uses a new hardware mode (some people refer to it as ring -1) to get access to special "curtained" memory. Only a small Palladium kernel called the "nub" runs in this hardware mode.
The ordinary Windows kernel is hopelessly insecure and the Palladium guys knew they'd never get anywhere if they relied on MS kernel security. The Windows kernel does not have sufficient privileges to touch Palladium secure code. Therefore Windows device drivers are not a security issue for Palladium.
Microsoft has also said that Palladium does not involve having Microsoft certify code or verify signatures. Instead they provide a general mechanism by which application developers can create programs which authenticate themselves to servers, and store encrypted data that other programs can't decrypt. It's an open system and Microsoft doesn't want to be in the business of checking every application out there to see if it satisfies some kind of Palladium security requirements. Instead, it will be up to each application developer to decide which programs it will trust.
The paper goes on and on about how WiFi shows that open spectrum works, there's no problem with congestion, no need for ownership or frequency allocation. Then in recommendation 2 we read this:
Improving existing unlicensed bands isn't enough. Most are so narrow and congested that their utility for open spectrum is limited.
So here he's saying exactly the opposite, that congestion is a serious problem for open spectrum! Which is it?
It's also bogus to claim that WiFi proves that open spectrum works. The truth is that WiFi is so sparsely implemented that congestion hasn't yet been an issue. For all the hype, my town of 175,000 people has no wireless public access points. Even in the big cities they're not so close that congestion is a problem.
The article could just as well have used cordless phones and baby monitors (which use the same frequencies) as evidence that open spectrum works. The only difference is that they don't score as high on the hype meter. All these examples prove is that the technology works when the range and distribution of the transmitters is sufficiently limited, which everyone knew already.
Lucky has a nice idea, but I don't think the timing is really going to work. Here's the problem.
He wants to know if Microsoft is going to use Palladium for copy protection. We'd all like to know that. Well, of course, we're going to find out sooner or later, at least by the time they release Palladium, maybe around 2005. And chances are we'll find out sooner than that, because Microsoft will release specs and APIs to the developer community in order to have applications ready when the technology is released. So maybe we'll find out about 2004.
Lucky wants to speed up this process, so he files a patent hoping that Microsoft will either challenge it, or it will turn out that they have a patent of their own. But it's likely to take a couple of years for his patent to go through. So he's not going to find out until around 2004 anyway.
The timing doesn't really work. Waiting to see if Microsoft contests the patent won't give information for a couple of years. And by that time, chances are Microsoft will have revealed enough information about Palladium that we'll know the answer anyway.
The one thing that isn't going to happen, I guarantee, is that Microsoft will say "Oh no! Our secret plan to use Palladium for copy protection is ruined due to Lucky Green! Curses, foiled again!" If Microsoft does plan to use Palladium like this, they'll have the patent protection in place well in advance.
Gattaca is in part an allegory about racism. The genetic "invalids" were treated like blacks and other minorities have been in the past. By making the victims of discrimination be people like us, the film lets us see things from the perspective of the oppressed class.
Slashdot Mirror
Let me explain more clearly, because it seems that some of the moderators didn't understand my comment.
Think about a photon, which has a linear polarization: up-down, left-right, slantwise, or at whatever angle you want. You can in principle put in an arbitrary amount of information in setting the polarization angle of a photon. You could divide a circle into as many parts as you want, and set the polarization to an angle corresponding to the value you want to send. This is like how they pack 1024 bits into a 19 nuclei molecule.
Now, the problem is reading the data back out. If you have only one photon in a particular polarization state, you can't determine that state with any accuracy. You can in fact only get one bit of data out of that photon. You can pass it through a polarizer and either it makes it, or it does not. This gives you information about the polarization state but it destroys that state in the process. You can put lots of information into a single photon, but you can't read it back out.
Now let's imagine that we have lots of photons, in a laser beam for example. We can set them all to the same polarization state. Now we can read the polarization quite exactly, by using large numbers of photons and turning our polarizing detector until we get a peak in the output.
Even though all the photons are in the same state (like in the NMR molecule experiment), it is because there are large numbers of them that we can read the state back out accurately. We would NOT be able to read back the data from a single photon, and in the same way we would NOT be able to read back the data from a single molecule.
Hopefully that explains my comment above. A qubit, whether photon polarization or nuclear spin, holds only a limited amount of information, and you can't read more out than it holds. There's no way you can get 1024 bits into 19 nuclei, and no one should try to "spin" the results of this experiment that way.
Grover writes that we're limited to "reversible" gates, i.e. those for which one can infer the inputs from the outputs... but it seems that the pigeonhole principle would prevent any gate with more than one input from being reversible
You can turn any irreversible gate into a reversible one by adding outputs. For example, 2-input XOR becomes reversible if you add a 2nd output which is a copy of one of the inputs. There are many tricks like this which are studied under the topic of reversible computation.
AFAIK, there are a grand total of three sorts of problems that it is known can be solved more efficiently by quantum computers: integer factorization/discrete log (solved by the same algorithm, and both mainly of use in cracking public-key cryptography), brute-forcing symmetric key cryptography (easily defeated by doubling the size of the key), and quantum physics simulations. Outside that, they are completely useless.
/.ers, graphics rendering. In some forms this can come down to "is there a nearby object blocking the view of this distant one?" That means doing a search. Theoretically, quantum computers could speed up these kinds of searches, thereby improving graphics rendering and making the games of the future even better. Plus they'll have better AI.
It's application number 2 on your list that could be important. You're talking about Grover's algorithm, but it is good for much more than brute forcing crypto keys!
Grover's algorithm can be used for any kind of searching problem. Take AI for example. Most AI problems can be expressed as searching through some abstract space for a solution. Grover's algorithm will halve the depth of the search trees. Chess computers that can look 6 moves deep today will be able to look 12 moves deep on a quantum computer, and so on.
Many computationally limited computer problems can be expressed in the form of searching for a solution. Let's even look at a problem near and dear to
So if you're like Goonie, bored by improvements in scientific modelling and our understanding of the universe, surely you'll get excited about having better computer games.
First, here is the abstract for the article.
Second, it doesn't work, at least not the way they say it does. You can't store 1024 bits in the nuclear magnetic spins of a 19 atom molecule!
Or more precisely, you can't retrieve that many bits. The spin state of a nucleus can be described by a complex number, but when you do a measurement you only get one bit out. With 19 nuclei you can read out only about 19 bits.
So how do they make it work? They've got a huge number of molecules there. Each one is loaded with the same data value. Using the redundancy in those molecules, the researchers can read out the 1024 bits. But if they had only a single molecule holding the value in its nuclear spins, as the paper implies, there's no way they could read out 1024 bits. So the density is not as high as they make it sound.
His predictions may make sense technologically, but not socially. Your appliances might be on the Net, but they will only be accessible to you. Your refrigerator will notify you by email or equivalent when it needs service, not the manufacturer. People don't want their appliances talking to strangers. The owners will want to be in charge.
Likewise when checking out in the store, you will need to explicitly authorize the purchase, otherwise you could contest it later. You will be presented with an itemized receipt based on a scan of your items, and you're supposed to look it over quickly and make sure there's nothing on there that you didn't buy. Then you do something to initiate the payment. You can't let people take money out of your account without giving approval! There has to be some action you take to explicitly authorize a certain amount to be transfered.
With the "presence", again you will have much more control over it than he implies. You will be able to say who can find out how you are reachable. You can have filters that automatically email you when your voice mail comes in, etc., so that people with different levels of access don't necessarily know how much priority they're getting. That way you don't offend people.
As far as ubiquitous TiVo, it depends on the outcome (both legislative and technical) of the copyright wars. You may be able to record the shows only under the control of strict DRM software that won't FF through the commercials (like the way DVD players won't FF through the FBI warnings now).
One additional social/technical prediction I'd make is more use of webcams for business meetings, creating the virtual office. Assuming that terrorism scares keep happening, people will prefer not to travel so much, and employees will want to stay home and not come into the dangerously concentrated population areas downtown. We'll see a continued trend towards white collar workers using live video feeds to communicate with their co-workers both locally and around the world.
That story doesn't really make sense. In the "original" timeline, before anyone came back from the future, everything was hunky-dory. So who came back to start the assassinations and mess everything up?
Anyway it kind of reminds me of Orson Scott Card's story about Columbus. They set up a time machine to convince him not to discover America because it set us on the path to environmental destruction. Only it turned out that Columbus had already been diverted to the Americas by an earlier time machine, because in the original timeline he'd conquered the Moslems in a Crusade and that had caused all kinds of problems on its own.
I met Bernstein briefly, and he seemed like a nice guy in person. He's relatively young, 30-ish, and soft spoken. But online he comes off as some kind of know-it-all curmudgeon.
Personally I liked the suggestion in the Usenet thread to return expired DNS cache data when the authoritative servers are unreachable, at least as an option. 99% of the time when you can't do a host lookup, the old cached data would still be right. All the DNS purists hated the idea of using expired data, like it's unclean or something. But if it's all you've got, isn't it better to use old information than to give up on letting the net work at all?
I mean think about it, where are the self assemblers going to get their energy and raw materials from?
In many proposals, assemblers are made largely of carbon in the form of solid crystals. You may have heard of it, it's called diamond. Believe it or not, diamond is actually a more stable form of carbon, which is why it forms spontaneously. Normally it's very slow to form, but there are lots of paper designs for how assemblers can build diamond shapes very quickly.
So where does the carbon come from, and the energy to form diamond? Both can come from biomatter. It's full of carbon. You're full of carbon. And because diamond is so low energy, the nanomachine actually gains energy by turning biomass into copies of itself.
In short, you are the ideal food for nanotech. It finds all the elements it needs, and all the energy, just by eating you.
We eat living matter for its energy and elements, and nanotech will, too. Only nanotech has teeth made of diamond crystals, it can be ultimately voracious, and our immune systems won't be able to touch it.
The problem of unconstrained nanotech replicators is real. Don't succumb to the head-in-the-sand belief that such things won't exist or won't be dangerous. Your body would have no more chance against a swarm of nanotech assemblers than a mouse in a food processor. You are tasty and tender, and don't forget it.
In his very first book, Engines of Creation, available online, Eric Drexler laid out the possible consequences of attempts to suppress nanotech research. See chapter 12 especially.
He describes an ambitious program which will allow nanotech to be developed safely, via active shields to protect the environment and sealed assembler labs to allow safe experimentation.
Of course Drexler was far, far ahead of his time, but his analysis should be a starting point for any consideration of the prospects for nanotech development.
The DMCA relevance is as follows.
When a user of an internet service provider posts copyright-infringing data, the ISP is obligated to take down the postings once it is notified of a claim by the copyright holder. This is part of the DMCA. Then there is a mechanism for the user to challenge the removal, saying that the removed material was not copyrighted. This is discussed under section 512(c) of the DMCA link above.
It's a different part of the DMCA than what we often discuss, like breaking copyright protection measures, which is section 1201.
Now, I agree that price information, especially if reformatted, cannot be copyrighted. Copyright covers a particular expression of the data, and not the raw data itself. So IMO the sites could safely ignore these warnings. But as someone who was recently sued for several billion dollars, I can certainly understand the reluctance to enter the legal system.
How does encrypting a user's files with a key, and then denying the key to the user improve security ?
The idea is that only the program which encrypted the file can decrypt it. The file is somehow "locked" to the hash of the program that was running when it was encrypted.
The reason this increases security is two-fold. First, if some malicious program is run, a trojan or a virus infected program, it can't access the data. Imagine that the data is your bank account PIN or credit card number. Right now an Outlook virus could look through your files and find the data. With Palladium this can't happen, because only the banking program can unlock those files.
Second, if the banking program itself gets infected with a virus, its hash will change, and it won't be able to unlock its own files any more! So even a virus which targets the financial software won't be able to steal its sensitive data. In this way, Palladium provides security against a much wider range of malware infections than is possible in computers today.
Another example of improving security would be a multi-player game or P2P system where keeping people from cheating is desirable. In that case it's to everyone's advantage that they can't run a rogue game client or change their data files. By giving up the ability to cheat in this way, the user ultimately gains security because he can participate in the system free from other people cheating as well.
Usually you have to give up something to get something. Giving up the ability to cheat and to break your promises can lead to real gains. If you are offered a download of some data only in exchange for promising not to share it with anyone else, Palladium allows you to make that promise in such a way that you can't go back on your word. This will eventually lead to more valuable data being made available than would be possible in world where cheating is easy and unpreventable.
1. How does Microsoft plan to allow non-proprietary Operating Systems access to Paladium media?
It's not clear what you mean by Palladium media. If you read the linked article, you see that Palladium has four components. None of them refer to Palladium media per se.
What they do have is attestation, which lets a remote server reliably determine that you are running WMP or some other DRM compliant software before you download. Then WMP can enforce whatever restrictions are specified in the data file. So you could call the media supplied by such a server "Palladium media", and chances are that no, the server won't give you the data if you're not running Palladium - but that's entirely up to the server operator. You can't force him to do what you want, and you can't fool him, thanks to Palladium.
2. Why would consumers want to purchase your product that removes rights they have over their own media?
Now, this doesn't make sense. It's not their own media! The data is on a server belonging to someone else. Palladium gives that server owner more information in deciding whether to let you download it. It allows the server to make sure you're running some software that will follow certain rules. If not, it won't give you the data.
So nobody is taking away rights over your own media. Anything you have today, you can continue to use. What Palladium does is let people decide whether to give you their media, and to do so only if you in effect agree to follow their rules.
In answer to your question about why consumers would want to purchase Palladium computers, the answer is obvious. Server operators won't give the data to people who don't have Palladium. So owning a Palladium computer will be the only way to get entertainment media in the future. Nobody's going to force you to buy one. But some (not all) content creators will refuse to give their content away unless you are running Palladium so that they can be confident that you won't steal their data.
Ask him if Palladium will provide a way to support DRM for new media types like DVD Audio or some kind of future enhancement of DVD.
You know what? They don't just look for e-mail addresses to send mail to. They also use the e-mail addresses as reply-to addresses. I found this out when I got an email from a guy who was puzzled by my auto-responder emailing him. It turns out that somebody sent a message to me and used his address as a reply-to address.
This might be due to the Klez virus or a variant. It forges the From address in email, using a random address from the victim's address book. So if someone has Alice and Bob in their address book and they get infected, they may send mail to Alice that claims to be from Bob. Here's a Wired article with more information.
There is nothing in the current motes that can not be miniaturized. In three years this demo will be done with a 6" aircraft, and millimeter-scale sensor nodes.
This is all leading to "Smart Dust".
The majority of tracks on the CD are also often *worthless*. Just let me download the songs I like and pay a reasonable amount per song!
You think that's something new? That's always been true. It can't explain why CD sales are down. Records have had filler since the beginning. Even the old 45 RPM singles had an A side and a B side.
And the fact is, some music is more accessible than others. Some songs are instant hits, the first time you hear them you like them. Other songs take a few hearings before you appreciate them. I know I've bought albums for one or two hits, but after I've listened to the CD half a dozen times I like several of the other songs just as much.
But if they do start selling individual songs, most people will only buy the hits. Without being more or less forced to listen to the other ones because of the album format, they'll never get past that accessibility barrier. This means that a typical artist will only sell one or two songs where they now sell a whole album. And since we all know that CD prices are not based on manufacturing/distribution costs, this means that the sellers will have to charge almost as much per song as they do for a whole CD now.
In other words, for the record companies and everyone involved to continue to be as profitable as they are today, they'll have to charge probably five to ten dollars per song downloaded! That's just basic economics based on the number of hit songs per CD, and based on where the costs are for a record company (most of which won't be reduced by online distribution).
That's the reality. I hope you consider that a "reasonable amount" to pay for a song, because that's what it costs to create them.
Music sales are down for the simple reason that there is very little music these days that inspires folks to run out and get the album.
So what makes you an authoritative judge of good music? Why should we care about your tastes? Do you really think music was so great in other eras, and that's more than just nostalgia talking?
Remember those great bands from 1980? Like KC and the Sunshine Band? The Captain and Tenille? Boy, if only we had groups like that around today, people would just be rushing out to buy those CDs, wouldn't they? Andy Gibb, Christopher Cross, Air Supply, Olivia Newton-John, the great names just roll on and on.
Face it. Your taste in music is no more authoritative or absolute than anyone else's. I'm tired of people saying that modern music is no good. The same thing has been said every single year since at least 1960, by nostalgic old fogies who can't stand the junk that those darn kids listen to nowadays.
If you're in a hurry to reach your crotchety old age, fine. But don't pretend this is an excuse for why people are buying less music. There's been good and bad music throughout history. No generation had a monopoly on the good stuff. Popular music today is called "popular" because people like it! And that mostly means young people. The fact that it's not to your liking says more about you than about the quality of the music.
"Reduce the cost of new CDs; if discs cost only a few dollars each, buying them might be preferable to spending the time and effort to make copies or find them online."
The point being missed is that the cost of a CD has to cover much more than just the manufacturing and distribution of that disk. It has to cover the expenses in finding and developing talent, recording and touring, marketing and advertising. These expenses far exceed the costs of pressing a CD.
Most bands are commercial failures, too, so the few successful acts have to be priced high enough to cover the money lost on the others. It's like the oil business where you have to drill 20 dry holes for every one that hits the black gold.
I don't see anyone, not indy labels, not bands selling their own music, who sells CDs for a few bucks each. If the big, corrupt labels were engaging in profiteering, then everyone else would sell their disks for three bucks, right? Well, that's not happening. I think this shows that the people who think this would be a reasonable price for a CD are missing some basic economics.
Seriously, I often have vi running in the left window and emacs in the right hand one. It's a good mental exercise to switch back and forth between them frequently. I wish I could train myself to use my right hand for emacs and the left for vi, but I'm not there yet. Maybe I could do it with two chord keyboards?
His idea is that all possible mathematical structures exist, and that we live in one of them! At some level, physics can be considered a branch of mathematics. Hence our universe can be considered as an enormously complicated mathematical structure. The question is, why this structure instead of some other?
His answer is that all mathematical structures exist, but that most of them are unsuitable for life. The paper linked above analyzes many different possibilities in terms of numbers of dimensions, numbers of time dimensions (yes, you could conceive of a universe with two-dimensional time), various other parameters, and he shows that structures that we would think of as living would have a hard time existing in universes much different from our own.
The Tegmark model can be thought of as the simplest possible physical theory. If physics is reducible to mathematics, then saying that all mathematical structures exist can be put more simply: Everything exists.
A similar model based on computation is proposed by Juergen Schmidhuber. Rather than Tegmark's mathematical structures, Schmidhuber proposes that all computations exist. Given that any mathematical model of a universe can be simulated by a computer program, these two formulations are roughly equivalent.
But Schmidhuber's approach has the advantage that it provides a natural way to say that some universes are more probable than others: namely, universes with short programs have more "measure" than universes with long programs. It follows that our universe probably has a relatively short program, which therefore explains why we observe that physical laws are mathematically simple.
It's pretty heavy stuff, but certainly exciting to see that researchers are (somewhat reluctantly) beginning to entertain multiverse models. The more ambitious "everything exists" theories are still too extreme for the mainstream, but I suspect that they, too, will get increasing attention over the next few decades.
It's ironic that Freenet, a tool for free speech and unfettered communication, is released under the GPL, a license which is designed to put restrictions on what you can do and say.
Under the GPL, if you distribute GPL'd software you are obligated to make the source available. Whether this is good or bad is not the point here. The point is that it is a limitation, a rule, that you must follow.
But Freenet is designed to free us from such limitations and rules. It allows us to publish information without regard to any restrictions that some third party wants to impose - even restrictions imposed by the GPL.
In short, using Freenet one could take GPL software, modify it and redistribute it without making the source code available. And you could get away with it. No one would be able to track you down and make you stop. Freenet makes it easy to bypass the GPL.
Granted, it would be hard to get paid for such software. Some people claim that the real point of the GPL is just that, to prevent people from getting paid for software (not that GPL advocates would ever admit it!). But it seems inconsistent at best for software designed to provide complete freedom of speech to be released under a restrictive license.
The first thing someone should do is to make a copy of the Freenet code with all the GPL licensing terms stripped out of the source, change it to a license that has no restrictions, and publish it on Freenet. That will truly demonstrate the nature of the beast.
Microsoft(tm)'s Palladium scheme will require signed drivers. There is simply no way to try to enforce that level of security while still allowing end users to insert arbitrary code into the kernel. Not with any standard definition of "driver" and "kernel".
That's not true. According to what Microsoft has said, Palladium runs "beside" Windows. It uses a new hardware mode (some people refer to it as ring -1) to get access to special "curtained" memory. Only a small Palladium kernel called the "nub" runs in this hardware mode.
The ordinary Windows kernel is hopelessly insecure and the Palladium guys knew they'd never get anywhere if they relied on MS kernel security. The Windows kernel does not have sufficient privileges to touch Palladium secure code. Therefore Windows device drivers are not a security issue for Palladium.
Microsoft has also said that Palladium does not involve having Microsoft certify code or verify signatures. Instead they provide a general mechanism by which application developers can create programs which authenticate themselves to servers, and store encrypted data that other programs can't decrypt. It's an open system and Microsoft doesn't want to be in the business of checking every application out there to see if it satisfies some kind of Palladium security requirements. Instead, it will be up to each application developer to decide which programs it will trust.
The paper goes on and on about how WiFi shows that open spectrum works, there's no problem with congestion, no need for ownership or frequency allocation. Then in recommendation 2 we read this:
Improving existing unlicensed bands isn't enough. Most are so narrow and congested that their utility for open spectrum is limited.
So here he's saying exactly the opposite, that congestion is a serious problem for open spectrum! Which is it?
It's also bogus to claim that WiFi proves that open spectrum works. The truth is that WiFi is so sparsely implemented that congestion hasn't yet been an issue. For all the hype, my town of 175,000 people has no wireless public access points. Even in the big cities they're not so close that congestion is a problem.
The article could just as well have used cordless phones and baby monitors (which use the same frequencies) as evidence that open spectrum works. The only difference is that they don't score as high on the hype meter. All these examples prove is that the technology works when the range and distribution of the transmitters is sufficiently limited, which everyone knew already.
Lucky has a nice idea, but I don't think the timing is really going to work. Here's the problem.
He wants to know if Microsoft is going to use Palladium for copy protection. We'd all like to know that. Well, of course, we're going to find out sooner or later, at least by the time they release Palladium, maybe around 2005. And chances are we'll find out sooner than that, because Microsoft will release specs and APIs to the developer community in order to have applications ready when the technology is released. So maybe we'll find out about 2004.
Lucky wants to speed up this process, so he files a patent hoping that Microsoft will either challenge it, or it will turn out that they have a patent of their own. But it's likely to take a couple of years for his patent to go through. So he's not going to find out until around 2004 anyway.
The timing doesn't really work. Waiting to see if Microsoft contests the patent won't give information for a couple of years. And by that time, chances are Microsoft will have revealed enough information about Palladium that we'll know the answer anyway.
The one thing that isn't going to happen, I guarantee, is that Microsoft will say "Oh no! Our secret plan to use Palladium for copy protection is ruined due to Lucky Green! Curses, foiled again!" If Microsoft does plan to use Palladium like this, they'll have the patent protection in place well in advance.
Gattaca is in part an allegory about racism. The genetic "invalids" were treated like blacks and other minorities have been in the past. By making the victims of discrimination be people like us, the film lets us see things from the perspective of the oppressed class.