This continual failure of mouse makers to address mousing on mirrors annoys me. Hey, mouse makers, here's how you can address this.
Originally, your mice were based on a rolling ball. The mouse felt the ball rolling, and figured out the movement. In human sense terms, your mice were based on touch.
The current mice are based on sight. They look at the surface under the mouse and see the relative movement.
The sense you should be looking at is smell.
In the center of the bottom of the mouse, there should be an emitter that leaves a chemical trail on the mousing surface. The rest of the bottom of the mouse should be covered with odor sensors that can sense when they are near the chemical. By laying a pattern of odors, and sensing them as they move under the mouse's "noses", the mouse can determine position.
Think ants and the chemical trails they leave to find their way around.
Not only would this address the mirror problem, you could also use this technology to address a serious workplace health issue. You could make it so the tracking chemical has antibacterial properties. This would help keep shared computers sanitary during cold season, reducing sick days and boosting overall productivity.
I didn't say you could take out everything you don't need for your specific application. On Windows, you can take out nearly all of the bundled user applications (Outlook Express, games, etc), including Internet Explorer. (You can't take out all the DLLs associated with IE, but you can get rid of the application).
When you do this, you end up with about the same level of security in an embedded Windows system as you do with an embedded Linux system. Of course I can't prove that, is it is a statement of experimental fact, not a statement of a proposed theorem.
My point, though, is that a voting machine should not be running either Linux of Windows. Even stripped down, you are left with a known insecure kernel.
A voting machine should run a custom kernel, designed just for that, with a custom voting application on top of that, and both the kernel and the application should be designed and implemented using state of the art formal methods, with formal proofs of correctness for both.
Linux in embedded applications is not necessarily any more secure than Windows. On both, if you take out things you don't need, so just run the kernel, minimal support applications, plus the custom application for your embedded device, you end up with about the same level of security.
And that level is NOT good enough for voting machines. The right way to do a voting machine is to design a system (hardware and software) specifically for this one task. This system should be subjected to state of the art formal methods, form the specification through the implementation (with all steps open for public review by experts). This would be hard, and might take a few years, but it would be worth it. Voting machine can have a very long service life, on the order of decades or even a century or more, so taking a few years to get it right up front is justified.
I don't think so. However, there is good news. On the forums on the OO site, the developers finally acknowledged that it's "navigator" mode (or whatever it's called--the thing they kept telling people to use who said they wanted something like Word's outline mode) is not an adequate substitute, and said that a proper outline mode is a high priority.
Unfortunately, they also said that implementing this will be a lot of work, so will take a while to get in.
Lawrence Lessig is a board member for the Electronic Frontier Foundation.
According to TFA:
She pressed that question through a number of channels until it found its way to the Electronic Frontier Foundation (on whose board I sat until the beginning of 2008)
For a long time, my plan has been to build something like this into my house. When I want to adjust my thermostat, for example, I want the house to give me a quick little test to see if I'm mentally alert. Regardless of the outcome, the house would then let me adjust the thermostat.
However, it would remember the results. And if it determines over a long period (say, a couple months), that I'm suffering significant mental degradation, to the point where I'm likely to not be able to take care of myself, the house will wait until I'm sound asleep one night (which it can determine by monitoring my temperature with infrared sensors, and listening to my respiration, for example), and then do something to kill me in my sleep (gas, probably).
When the house is sure that I'm dead (no breathing for a long time, and body temperature down to ambient room temperature), it will then call the coroner's office to report my death.
He definitely has played Everquest. He was the leader of the guild Legacy of Steel on The Nameless. They were one of the top three uberguilds across all EQ servers (the other two were Fires of Heaven and Afterlife).
Uberguild web sites are usually full of somewhat childish bragging about their accomplishments. LoS's site was notable for having much less of that, and having interesting articles written by Tigole about the problems with the EQ high end game, and how to fix them. They were very good suggestions and would have vastly improved the high end game, but of course Sony ignored them. I bet they now wish they had listened.
But Windows XP is dead. Microsoft wants you using Vista, and Vista is one of the hardest OSes to learn after using XP
How so? I recently built a new gaming rig, and it has Vista. I last seriously used XP on my previous gaming rig, but in the last two years, I haven't done much PC gaming, so haven't used that system much.
I've had no trouble figuring out Vista. When I have some problem, or want to make configuration change, or something like that, I've found that if I ask myself "what would I do in XP?", and then try to apply that answer in Vista, I end up at the right place. And when I get there, I almost always have found that the way Vista actually handles accomplishing what I want to do is much better than XP's way.
Also, how are they going to take control of changes you already made? You've already licensed them under the BSD license; someone else could just republish them. You can't revoke things like that.
That's a bit of an overstatement. If an open source license is treated in court as a contract, it may or may not be revokable.
Furthermore, it is important how third parties receive their license. Suppose a license, is non-revokable. If the owner of the software, O, licenses the software to a licensee, L, and L gives the software to a third party, T, where does T get his license? If the license is written so that it gives L the right to grant licenses, and T's license comes from L, no problem. However, if the way the license works is that T receives his license from O when L gives T the software, then O could refuse to grant T a license.
The effect of this later situation would be that, from the time O stops granting new licenses, people who already have licenses could redistribute, as they have licenses to do so. Their recipients would not have a license, but would be owners of their particular copies, and so able to use them, but not redistribute copies themselves.
But wait. The license to L might include an implicit promise that O will continue to grant licenses. That's part of the value of an open source license--you can redistribute the (possibly modified) software, and you know that your recipients can modify and redistribute. L could be relying on that, O could know that, and this could be part of the consideration for the contract.
For licenses that are not interpreted as contracts (not many, as it doesn't take much to be seen as a contract--you pretty much have to go out of your way to avoid it), but rather are seen as bare copyright licenses, they can be revoked at any time. Every act a recipient does that requires copyright permission is a new occasion on which permission must be obtained from the copyright owner, and the owner can stop granting that permission at any time. This is one of the big reasons that Larry Rosen, in his book on open source licensing, recommends that authors of new open source licenses be explicit about making them contracts, not bare licenses.
Anyway, the bottom line here is that this area is unclear, and is going to remain so until we have some serious attempts at revocation of open source licenses, and these result in litigation. That will at least provide some clarity (although it might only provide clarity for the specific licenses involved--we may not get any good general principles, but that could be a good thing, as it would encourage people to use one of the licenses where this is settled, rather than writing new licenses)
In particular, how much is applicable to the Mac versions? 2008 dropped VBA support, I believe, which sounds like it could wipe out much of the applicability of this book.
Companies cannot be inventors, as far as the patent system is concerned. If the company wants a patent, it has to have the inventors apply, and have the inventors assign the patent to the company.
Note that most of his research seems to be about properties of fluids and modifying them electromagnetically, and he has quite a few publications in peer-reviewed journals in this area.
This seems quite likely to be legitimate. You are making the elementary logic error of assuming that because many crackpots have claimed to accomplish a certain result, anyone who claims to accomplish that result must be a crackpot.
Hell, if I had it to do over again, I wouldn't have bought my DVD collection in the first place. Most of my hundreds of DVDs, just like most of my hundred or so Laserdiscs before them, have not been off the shelf since their first viewing.
As for code? Well I've used the GPL for all my research related code, and I've had no problems
Why GPL? I would expect that when one is publishing academic research, one would want to license any accompanying code under a license that is as compatible as possible with whatever code other researchers are using. That would be something like BSD.
EA does do Mac games. Spore and several of the sports games, for example. Furthermore, while they have declined to comment on whether or not they are going to do a Mac version of Warhammer, Mark Jacob's has strongly hinted. E.g., he's said no comment, and then added that the other computer next to him is not a PC.
They got people talking. The last two TWiT podcasts, for example, included long discussions of the ads. The discussion was critical of them--but the ads got Leo Laporte and his guests to spend something like 30 minutes of some of the most valuable podcast time on the...er......uhm..,pod(?) talking about them.
One of the guests, not quite seriously, did a detailed symbolic analysis of the second ad. He said the old lady represented Steve Jobs--she had been living with the family for the same amount of time since Steve came back to Apple.
How about making a capacitor that uses the platters of a hard disk as the plates? The idea would be to combine the power source and hard disk on a laptop, to save space and weight.
Sure, it sounds like a good idea to have compulsory license fees except for the fact that terrestrial radio pays no royalties for anything it plays?
They do pay. Copyright licensing for most radio broadcast music is handled by BMI and ASCAP. The radio stations report what they play, and BMI and ASCAP collect the royalties for the copyright owners.
I think many people are a little unclear on how compulsory licenses work. Once that is understood, a lot becomes clearer.
For most things you might want to do that require permission of the copyright owner, how much that costs you, if anything is entirely up to whatever agreement you and the copyright own make. If you are unable to come to an agreement, you can't go ahead with your proposed use of the copyrighted material.
For some particular uses of some kinds of work, copyright law provides a different default. If you and the copyright holder can't agree to terms, you can pay an amount determined by the Copyright Royalty Board, and go ahead with your proposed use, even if the copyright owner does not want to give you a license.
The Board generally sets these rates higher than what the market rate would be. In most cases, the party that wants to use the work and the copyright owner comes to terms, and agree on a rate well below the compulsory rate.
It is quite sensible that the compulsory rate is set high, as its a rate of last resort, for the case where the copyright owner is going to be forced against their will to license the work. Also, it would probably not be possible for the Board to set a rate that is close to market rate, even if they wanted to, as market rate would vary quite a bit from work to work. At best, they could approximate the average market rate for the type of work in question, but then that would greatly mess up the market, as for about half the works, the compulsory rate would be lower than the market rate!
In the Pandora case, we have that rare situation where the copyright owners and the party that wants a license are not able to agree to terms, leaving Pandora with the choice of stopping, or taking the expensive compulsory license.
Slashdot Mirror
This continual failure of mouse makers to address mousing on mirrors annoys me. Hey, mouse makers, here's how you can address this.
Originally, your mice were based on a rolling ball. The mouse felt the ball rolling, and figured out the movement. In human sense terms, your mice were based on touch.
The current mice are based on sight. They look at the surface under the mouse and see the relative movement.
The sense you should be looking at is smell.
In the center of the bottom of the mouse, there should be an emitter that leaves a chemical trail on the mousing surface. The rest of the bottom of the mouse should be covered with odor sensors that can sense when they are near the chemical. By laying a pattern of odors, and sensing them as they move under the mouse's "noses", the mouse can determine position.
Think ants and the chemical trails they leave to find their way around.
Not only would this address the mirror problem, you could also use this technology to address a serious workplace health issue. You could make it so the tracking chemical has antibacterial properties. This would help keep shared computers sanitary during cold season, reducing sick days and boosting overall productivity.
I didn't say you could take out everything you don't need for your specific application. On Windows, you can take out nearly all of the bundled user applications (Outlook Express, games, etc), including Internet Explorer. (You can't take out all the DLLs associated with IE, but you can get rid of the application).
When you do this, you end up with about the same level of security in an embedded Windows system as you do with an embedded Linux system. Of course I can't prove that, is it is a statement of experimental fact, not a statement of a proposed theorem.
My point, though, is that a voting machine should not be running either Linux of Windows. Even stripped down, you are left with a known insecure kernel.
A voting machine should run a custom kernel, designed just for that, with a custom voting application on top of that, and both the kernel and the application should be designed and implemented using state of the art formal methods, with formal proofs of correctness for both.
Linux in embedded applications is not necessarily any more secure than Windows. On both, if you take out things you don't need, so just run the kernel, minimal support applications, plus the custom application for your embedded device, you end up with about the same level of security.
And that level is NOT good enough for voting machines. The right way to do a voting machine is to design a system (hardware and software) specifically for this one task. This system should be subjected to state of the art formal methods, form the specification through the implementation (with all steps open for public review by experts). This would be hard, and might take a few years, but it would be worth it. Voting machine can have a very long service life, on the order of decades or even a century or more, so taking a few years to get it right up front is justified.
I don't think so. However, there is good news. On the forums on the OO site, the developers finally acknowledged that it's "navigator" mode (or whatever it's called--the thing they kept telling people to use who said they wanted something like Word's outline mode) is not an adequate substitute, and said that a proper outline mode is a high priority.
Unfortunately, they also said that implementing this will be a lot of work, so will take a while to get in.
According to the submitter's blurb:
According to TFA:
For a long time, my plan has been to build something like this into my house. When I want to adjust my thermostat, for example, I want the house to give me a quick little test to see if I'm mentally alert. Regardless of the outcome, the house would then let me adjust the thermostat.
However, it would remember the results. And if it determines over a long period (say, a couple months), that I'm suffering significant mental degradation, to the point where I'm likely to not be able to take care of myself, the house will wait until I'm sound asleep one night (which it can determine by monitoring my temperature with infrared sensors, and listening to my respiration, for example), and then do something to kill me in my sleep (gas, probably).
When the house is sure that I'm dead (no breathing for a long time, and body temperature down to ambient room temperature), it will then call the coroner's office to report my death.
He definitely has played Everquest. He was the leader of the guild Legacy of Steel on The Nameless. They were one of the top three uberguilds across all EQ servers (the other two were Fires of Heaven and Afterlife).
Uberguild web sites are usually full of somewhat childish bragging about their accomplishments. LoS's site was notable for having much less of that, and having interesting articles written by Tigole about the problems with the EQ high end game, and how to fix them. They were very good suggestions and would have vastly improved the high end game, but of course Sony ignored them. I bet they now wish they had listened.
How so? I recently built a new gaming rig, and it has Vista. I last seriously used XP on my previous gaming rig, but in the last two years, I haven't done much PC gaming, so haven't used that system much.
I've had no trouble figuring out Vista. When I have some problem, or want to make configuration change, or something like that, I've found that if I ask myself "what would I do in XP?", and then try to apply that answer in Vista, I end up at the right place. And when I get there, I almost always have found that the way Vista actually handles accomplishing what I want to do is much better than XP's way.
That's a bit of an overstatement. If an open source license is treated in court as a contract, it may or may not be revokable.
Furthermore, it is important how third parties receive their license. Suppose a license, is non-revokable. If the owner of the software, O, licenses the software to a licensee, L, and L gives the software to a third party, T, where does T get his license? If the license is written so that it gives L the right to grant licenses, and T's license comes from L, no problem. However, if the way the license works is that T receives his license from O when L gives T the software, then O could refuse to grant T a license.
The effect of this later situation would be that, from the time O stops granting new licenses, people who already have licenses could redistribute, as they have licenses to do so. Their recipients would not have a license, but would be owners of their particular copies, and so able to use them, but not redistribute copies themselves.
But wait. The license to L might include an implicit promise that O will continue to grant licenses. That's part of the value of an open source license--you can redistribute the (possibly modified) software, and you know that your recipients can modify and redistribute. L could be relying on that, O could know that, and this could be part of the consideration for the contract.
For licenses that are not interpreted as contracts (not many, as it doesn't take much to be seen as a contract--you pretty much have to go out of your way to avoid it), but rather are seen as bare copyright licenses, they can be revoked at any time. Every act a recipient does that requires copyright permission is a new occasion on which permission must be obtained from the copyright owner, and the owner can stop granting that permission at any time. This is one of the big reasons that Larry Rosen, in his book on open source licensing, recommends that authors of new open source licenses be explicit about making them contracts, not bare licenses.
Anyway, the bottom line here is that this area is unclear, and is going to remain so until we have some serious attempts at revocation of open source licenses, and these result in litigation. That will at least provide some clarity (although it might only provide clarity for the specific licenses involved--we may not get any good general principles, but that could be a good thing, as it would encourage people to use one of the licenses where this is settled, rather than writing new licenses)
In particular, how much is applicable to the Mac versions? 2008 dropped VBA support, I believe, which sounds like it could wipe out much of the applicability of this book.
I was referring to the first claim (patents in general).
Companies cannot be inventors, as far as the patent system is concerned. If the company wants a patent, it has to have the inventors apply, and have the inventors assign the patent to the company.
The vast majority of published research on this disagrees with you.
Note that most of his research seems to be about properties of fluids and modifying them electromagnetically, and he has quite a few publications in peer-reviewed journals in this area.
This seems quite likely to be legitimate. You are making the elementary logic error of assuming that because many crackpots have claimed to accomplish a certain result, anyone who claims to accomplish that result must be a crackpot.
Finally! Now we'll have something to pass the time between World Car Free Day (Sept 22) and International Grab Hand Day (Sept 25).
Hell, if I had it to do over again, I wouldn't have bought my DVD collection in the first place. Most of my hundreds of DVDs, just like most of my hundred or so Laserdiscs before them, have not been off the shelf since their first viewing.
LSB brings the distros all together--it gives them something in common to ignore.
No copyright protection? So they are only releasing music that is in the public domain!?
Or did the newspaper screw up, and mean to write "no copy protection"?
Why GPL? I would expect that when one is publishing academic research, one would want to license any accompanying code under a license that is as compatible as possible with whatever code other researchers are using. That would be something like BSD.
EA does do Mac games. Spore and several of the sports games, for example. Furthermore, while they have declined to comment on whether or not they are going to do a Mac version of Warhammer, Mark Jacob's has strongly hinted. E.g., he's said no comment, and then added that the other computer next to him is not a PC.
WAR has no ganking, and an extensive PvE game.
They got people talking. The last two TWiT podcasts, for example, included long discussions of the ads. The discussion was critical of them--but the ads got Leo Laporte and his guests to spend something like 30 minutes of some of the most valuable podcast time on the...er......uhm..,pod(?) talking about them.
One of the guests, not quite seriously, did a detailed symbolic analysis of the second ad. He said the old lady represented Steve Jobs--she had been living with the family for the same amount of time since Steve came back to Apple.
How about making a capacitor that uses the platters of a hard disk as the plates? The idea would be to combine the power source and hard disk on a laptop, to save space and weight.
They do pay. Copyright licensing for most radio broadcast music is handled by BMI and ASCAP. The radio stations report what they play, and BMI and ASCAP collect the royalties for the copyright owners.
I think many people are a little unclear on how compulsory licenses work. Once that is understood, a lot becomes clearer.
For most things you might want to do that require permission of the copyright owner, how much that costs you, if anything is entirely up to whatever agreement you and the copyright own make. If you are unable to come to an agreement, you can't go ahead with your proposed use of the copyrighted material.
For some particular uses of some kinds of work, copyright law provides a different default. If you and the copyright holder can't agree to terms, you can pay an amount determined by the Copyright Royalty Board, and go ahead with your proposed use, even if the copyright owner does not want to give you a license.
The Board generally sets these rates higher than what the market rate would be. In most cases, the party that wants to use the work and the copyright owner comes to terms, and agree on a rate well below the compulsory rate.
It is quite sensible that the compulsory rate is set high, as its a rate of last resort, for the case where the copyright owner is going to be forced against their will to license the work. Also, it would probably not be possible for the Board to set a rate that is close to market rate, even if they wanted to, as market rate would vary quite a bit from work to work. At best, they could approximate the average market rate for the type of work in question, but then that would greatly mess up the market, as for about half the works, the compulsory rate would be lower than the market rate!
In the Pandora case, we have that rare situation where the copyright owners and the party that wants a license are not able to agree to terms, leaving Pandora with the choice of stopping, or taking the expensive compulsory license.