Thundercats? Sesame Street isn't commercial TV. I would guess that he is referring to afternoon cartoons: Thundercats, Masters of the Universe, and Transformers come to mind from my teen years. Presumably there are new ones now. Notice how all the females in Thundercats and Masters of the Universe wear bikinis and the men are bare chested. The ads are all for junk food. The plots are violence based (even if there is an anti-violence message).
Any system that allows you to verify your vote allows you to verify to a third party that you voted in a certain way. That is inherently bad, as it allows vote selling.
We don't need a complicated eVoting system. We need a system that has the voter verify that the vote counter is counting the vote correctly prior to storing the vote. That was the only problem in the Florida election: people thought that their votes were being cast in a different way than they actually were. Simply putting the ballot through a reader would have told the voter how the vote was actually to be cast. The vote could then be put in a locked box for later recounts (or destroyed if it did not match the voter's wishes).
Open source is not the fix here. What if they replace the *entire* system with another binary that produces the same hash?
The problem is that these systems aren't auditable because there is no voter verified record. To fix this, you add a voter verified record (i.e. print out the ballot). If you have this record, you don't need to verify the machine. The record is sufficient.
You missed a step. After the paper ballot is printed, put it in whatever will be reading the ballot and verify that it is reading it the way that you expect. This ensures that the vote is understandable (and can count the votes so that the end of night count can be read off the reader).
Note that this step was what was missing from the problematic Florida ballots. No one ever verified that the vote was being read properly before the voter left. If they had, they would have caught both problems: votes that were indeterminate (due to hanging chads, etc.--these ballots should have been destroyed and replaced); and votes that were placed for Buchanan but meant for Gore.
"He assumes that the patent office, which can't even seem to read the patent applications, would be able to test the software and determine whether it was indeed a functional and useful implementaion"
Even if they fail at this, there are still two things that this does:
1. A defendant in an infringement suit can read the actual software and use it in the defense. If the code is non-functional, then it will be easier to invalidate the patent in court (since a valid implementation is required).
2. After the patent finishes, the code is available. Currently, a software patent offers no new information.
"Sounds more like a "shared source" (look, don't touch) than "open source" to me"
Yes, but after the patent period expires, the source would be public domain. I.e. during the period of the patent use of the source is restricted (and would be even if the source was not in the patent). Afterwards, use of that source would be unrestricted.
Also, if you obtain a patent license, there is nothing keeping you from modifying the code during the patent period. It's not Free, but it is open. However, I think that Mr. Bray was using open source only to mean visible source, not modifiable source. One would not be able to use the code in F/OSS projects.
"Under the GPL, you CANNOT distribute a work if you cannot give people the rights to *redistribute* it."
A patent does not prevent someone from redistributing a work, just using it. Further, as the Sveasoft case demonstrated, the GPL does *NOT* prevent someone from making a separate agreement that prevents redistribution. However, there is actually a section in the GPL explicitly about patents though. If Novell had software that was patent encumbered, they could not distribute it under the GPL.
While open source programmers and projects probably won't be able to take advantage of defensive patent pools on an individual basis (due to the cost issues you cite), this might be useful for the open source IP indemnification company. They have to audit code for patent infringement anyway, so if they find anything potentially patentable, it might make sense to patent it. This would both prevent that method being patented by someone else (after the code audit) and potentially allow it to be used in a countersuit.
This would not be practical in all cases, but IIRC the indemnification could be for amounts greater than $3 million (and might involve a patent defense anyway). In those situations, a patent counterattack might be cheaper than paying out the suit.
I would think that the argument against this would be based more on the question of validity of open source patents. How can one patent something that has already been published in an open source project. Of course, the Kodak/Sun case had a patent on something that looked suspiciously like Smalltalk to me, so the mere prior publication of a system may not be relevant.
That would require that software patents be legitimate. Further, as others have pointed out, this *is* a legitimate use of patents currently. Note that IBM is using patents exactly this way against SCO.
"it is certainly safe to say that through at least 26 weeks (as another poster pointed out) the womb is necessary. Why can't a women decide if before that time she wants to terminate the pregnancy? The use of said technology is a choice."
It is certainly safe to say that through the age of six months, the child cannot support itself and requires constant care. Why can't the parents decide to terminate the relationship? Doing so would be a choice. Instead, if either parent abandons the baby, we make that parent pay child support. If they kill the child (e.g. by abandoning it in a dumpster), we charge them with murder.
While abortion views are heavily influenced by religious views, there are still philosophical/legal/ethical questions involved. One, is the fetus a human being. Two, if so, is the mother obligated (by being involved in the creation of this human being) to try to bring it to term? When does this obligation start?
Note that a father is obligated to provide child support even if he has no contact with the woman after conception. He is bound by that decision.
"Actually, it is a commonly-held belief that sex crimes (ie: rape) are more about power and control than sexual gratification. I doubt that porn has much of an impact in that respect."
Porn replacements like being a peeping tom are known gateway crimes (i.e. many rapists start as peeping toms). People get into it for the sex, but stay for the power/control. Porn is a passive resource; it does not feed power/control issues.
The only observed relations between porn and sex crimes suggest that availability of porn decreases sex crimes.
"I'm sorry, getting paid overtime is a bad thing now?"
I think that you missed your parent's point. If you don't punch a time clock, then overtime is *UNpaid*, not paid. Thus, punching a time clock is better for the IT worker, as it results in wages with overtime. Not punching a clock is bad because it usually results in lots of overtime with no increase in pay.
"that's hard to refute if you look at how those voters would otherwise have leaned."
No it isn't. In both cases, exit polls showed that the third party candidate drew roughly equally from both parties. In general, most voters who strongly preferred Gore to Bush or Bush to Clinton voted for Gore or Bush Sr. What was left were a bunch of people who didn't really care.
Nader might possibly have cost Gore the Florida vote. The overall margin was so small that even a small change could have made a difference. However, most voters who would have voted for Gore if Nader weren't in the race did vote for Gore (since Nader wasn't an effective candidate). It is just as likely that Nader voters would have not voted as have voted for Gore (or Bush).
To get back to the previous point, once Perot re-entered the race, he never had a shot. At best, he was the first choice of around 25% of voters. He would have lost handily in IRV voting, as he would have been eliminated in favor of Bush and Clinton (both of whom had more first place votes). He might have won in Condorcet voting, as he was probably the second choice of most voters for the other candidates; he was a relatively centrist candidate.
In no way did Perot lose because voters did not take him seriously. Perot lost because he abandoned the election in the middle. Until then, he had been in a dead heat with the other candidates.
In a two candidate election, both candidates will race for the middle. That way, each voter counts twice, once for the candidate and once for not voting for the opponent. This is not a bad thing. It causes policy to maintain a relatively constant path. If anything, the bad part is the primary system, which puts forth more partisan candidates.
And all of your examples are for uses that are meaningless without a technical society to support them. In case of a technological collapse, you would be better off hoarding canned goods or low tech tools (e.g. a smithy) than gold. You can't eat gold, nor can you use it to make tools (it's too soft).
Gold is a luxury metal. It requires rich people (who can afford luxuries) to give it value. One of your examples is as an alternative to silver in dental fillings; yet gold is a hundred times as expensive as silver. Take away the jewelry uses of gold, and it will only have a fraction of its current value.
Yes, I saw that afterwards. Note that it bolsters my original point: that the advertised increase does not prove that the sky has started falling.
I had assumed that the argument that a 4PPM increase was unaturally high compared to 1-2PPM/year was based on real info. Instead, we see that the 4PPM increase is hardly higher than normal. My bad in assuming that they at least got the base data correct.
"this has resulted in nonsensical results like someone getting more time for breaking into a gas station and making off with a couple of bags of chips"
This is only nonsensical if you regard the three strikes law as punishment. It is not. The three strikes law is saying that this is a career criminal who will commit more crimes to spend more time in jail; why not just skip the "commit more crimes" part and let them spend more time in jail? Murderers are less likely to commit additional crimes in the future; murder is often a crime of passion by an otherwise law abiding citizen. Thus, it makes more sense to put the career criminal in jail (where he would likely end up anyway) than the murderer (who may have honestly reformed while in jail; i.e. it is possible that the system worked).
I'm generally against mandatory sentencing. I would prefer to leave flexibility in the system. However, the arguments that you are using suggest that mandatory sentencing may be necessary in these cases because judges (and you) are missing the point. The point is not to punish or deter crime; the point is to find recurrent criminals and get them off the streets.
Your example of the guy who got drunk and stole some chips is exactly the kind of situation that I mean. Here is someone who is clearly an idiot. I would rather catch this with a meaningless crime than by him hitting a child or driving while drunk. Clearly, jail is not a deterrent and did not have a rehabilitative effect. Jail does have a preventative effect; while there, he isn't committing crimes against the general population.
Any system is going to have overdone punishments. For example, while delivering pizzas, I was held up twice. The first time, they caught four juveniles and charged them with armed robbery. For what? Five ones and a bunch of change (about $1.50). Fortunately for them, they were juveniles, but if they had been adults, they could have gotten a long prison sentence. Because they stole $6.50? No, because they could have killed me or someone else through their foolish decision to use a gun in their robbery.
For someone who is considering a crime, jail may have a deterrent effect. For the first or second time convict, jail may have a rehabilitative effect. For the third time convict, one might as well put away those hopes. If the threat of jail didn't deter three times (possibly more; they didn't necessarily get caught for all their crimes) and the reality of jail didn't rehabilitate twice, then there is no reason to think that jail will work for them in the future. Give up on them and remove them from contact with society.
Now, if you want to argue that the war on drugs is a failed policy, that's a different matter entirely. However, I would point out that even if drugs are legalized, it is still possible to have an illegal drug problem (e.g. the UK). Because of this, you will still have the problem of young kids selling illegal drugs to bypass taxes and age limits. It is also worth noting that even legal drugs provide an incentive for crimes. Since it is hard to maintain a normal job while on high levels of drugs (and the drugs themselves are not free), there is a strong incentive to get a high profit job, e.g. theft or fraud, to pay for one's habit.
"So what makes you so sure 4PPM is insignificant here?"
The link your parent posted showing that 4 PPM increases have been seen in a single month? The problem is not the 4PPM increase, it's the possibility that the increases will have permanently changed from 1-2 PPM/year to 4PPM/year or more.
"we'll have stupidity such as the three strikes law. This is law based on a baseball term! Talk about stupidity ruling the unwashed masses!"
By your own standards, the three strikes law is not stupid. The three strikes law is not intended to be punitive or rehabilitative; it is purely preventative. The issue is that the majority of crimes are committed by repeat criminals. The three strikes rule takes someone who is recidivist (i.e. has a history of committing crimes after being released from jail) and stops releasing them to commit more crimes. It has a much lower incidence of false positives (innocents imprisoned), because it requires multiple convictions; it is unlikely that a totally innocent defendant will be convicted three times.
Good marketing says nothing about an idea's merits: positive nor negative. The basis of the name is irrelevant. What's important is that this actually addresses the problems of recidivist criminals, something that existing punitive and rehabilitative actions do not.
It's not so much efficiency as recognition of desires. It is somewhat similar to the issue with IRV. IRV is bad because some of the time, it eliminates a candidate on whom voters would be more willing to compromise. Similarly, the reverse commons effect underspends on a particular service even though all taxpayers would be willing to pay more for more service. The IRV issue can be solved by switching to Condorcet voting, which scores the same votes, only better.
Another example of a similar problem is blind auctions (like eBay). There is a strong incentive to lie about the price you are willing to pay in an auction. If your bid is higher than everyone else's, you win the auction, regardless of whether it is 1 penny higher or a million dollars. As a result, there is a strong tendency to bid just a little higher than you expect everyone else to bid. Unfortunately, everyone else has the same incentive, so they also underbid. As a result, the seller gets screwed. eBay solved this by awarding to the highest bidder at the price of the second highest bid (a Vichy auction; named after the Nobel laureate who thought of it).
In a Vichy auction, there is no incentive to underbid, as your bid *never* determines the price you pay (except as an upper limit), just whether or not you win the bid. As a result, both sides bid higher and the seller makes more money.
What I am saying is that the reverse commons effect does not take into full account the preferences of the purchasers. As a result, they will tend to prefer the government run system which better reflects their preferences. As such, it is harder to switch systems (practical arguments tend to trump moral arguments; e.g. many northern abolititionists were swayed more by a fear of slave labor competing in factories than by the moral wrongness of slavery). One should thus keep an eye out for solutions to this.
"Simply edit the url you are currenlty in and remove the part before slashdot.org"
But then it appears separately in my browser history, which makes it harder to find the sites that I would really like to visit. If it is a clickable link (like the five I made), then it doesn't appear separately. Given the way that I use my browser (frequent visits to a constantly changing list of recently visited sites), it is actually easier for me to post my own clickable link and use that than it is for me to edit the URL manually in the location bar.
It would of course be a lot easier for me if I could set my preferences to use a particular color scheme instead of the IT color scheme. However, there doesn't seem to be much interest in that.
Is slashdot really better off making people find a CSS hack that changes the color scheme (and disables all the ads) on it. pages? Wouldn't it make more sense to enourage the simple use of alternative color schemes and avoid more complicated methods that also interfere with revenue? If I patch slashcode to allow the color scheme to be user specified, will/. adopt it into the main trunk? What is needed here?
P.S. I'm holding out for a phone that runs embedded Linux. Do I really need a phone that runs embedded Linux? Of course not, but I want one anyway.
There is alrady a patent clause in the GPL. The current clause says that software cannot be distributed under the GPL if it is patent encumbered. It seems (did not RTFL) this license goes a step further and includes (and requires!) a patent license as part of it.
Not sure how the FSF will proceed here. The comments they sent me regarding the Sveasoft abuses of the GPL implied that they did not consider a license that virally applies contract terms to be valid. I.e. they would not be able to enforce a license that demands that patent rights be waived if you redistribute. The current GPL works the other way; if you hold patent rights, you can't redistribute under the GPL.
Note: the issue here is not Microsoft. The issue arises when IBM redistributes the software. Does it automatically get protection from IBM's patent portfolio?
Neither is a browser. P2P allows for three classes of exploits, all of which they describe.
The main reason that businesses should keep P2P software off their machines is that it makes them vulnerable to lawsuits by copyright holders (the legal exploit). *All* P2P programs are vulnerable to that and the spoofed content (social) exploit. The technical exploits (which are program specific) are much less serious in comparison.
This is not a technical article. They aren't giving prizes to the best exploits. It is simply a discussion of various things people can do to minimize the exploitability of their computer systems. It includes things like being sued for copyright infringement as an "exploit" (presumably because it does harm to the system owner).
Slashdot Mirror
Thundercats? Sesame Street isn't commercial TV. I would guess that he is referring to afternoon cartoons: Thundercats, Masters of the Universe, and Transformers come to mind from my teen years. Presumably there are new ones now. Notice how all the females in Thundercats and Masters of the Universe wear bikinis and the men are bare chested. The ads are all for junk food. The plots are violence based (even if there is an anti-violence message).
Any system that allows you to verify your vote allows you to verify to a third party that you voted in a certain way. That is inherently bad, as it allows vote selling.
We don't need a complicated eVoting system. We need a system that has the voter verify that the vote counter is counting the vote correctly prior to storing the vote. That was the only problem in the Florida election: people thought that their votes were being cast in a different way than they actually were. Simply putting the ballot through a reader would have told the voter how the vote was actually to be cast. The vote could then be put in a locked box for later recounts (or destroyed if it did not match the voter's wishes).
Open source is not the fix here. What if they replace the *entire* system with another binary that produces the same hash?
The problem is that these systems aren't auditable because there is no voter verified record. To fix this, you add a voter verified record (i.e. print out the ballot). If you have this record, you don't need to verify the machine. The record is sufficient.
You missed a step. After the paper ballot is printed, put it in whatever will be reading the ballot and verify that it is reading it the way that you expect. This ensures that the vote is understandable (and can count the votes so that the end of night count can be read off the reader).
Note that this step was what was missing from the problematic Florida ballots. No one ever verified that the vote was being read properly before the voter left. If they had, they would have caught both problems: votes that were indeterminate (due to hanging chads, etc.--these ballots should have been destroyed and replaced); and votes that were placed for Buchanan but meant for Gore.
"Ctrl+P Choose Acrobat PDFWriter as printer."
That's not in MS Word... That's an add on.
"He assumes that the patent office, which can't even seem to read the patent applications, would be able to test the software and determine whether it was indeed a functional and useful implementaion"
Even if they fail at this, there are still two things that this does:
1. A defendant in an infringement suit can read the actual software and use it in the defense. If the code is non-functional, then it will be easier to invalidate the patent in court (since a valid implementation is required).
2. After the patent finishes, the code is available. Currently, a software patent offers no new information.
"Sounds more like a "shared source" (look, don't touch) than "open source" to me"
Yes, but after the patent period expires, the source would be public domain. I.e. during the period of the patent use of the source is restricted (and would be even if the source was not in the patent). Afterwards, use of that source would be unrestricted.
Also, if you obtain a patent license, there is nothing keeping you from modifying the code during the patent period. It's not Free, but it is open. However, I think that Mr. Bray was using open source only to mean visible source, not modifiable source. One would not be able to use the code in F/OSS projects.
http://www.fsf.org/licenses/gpl.html -- sections 7 and 8 mention patents. 7 would be the section of which I was thinking.
"Under the GPL, you CANNOT distribute a work if you cannot give people the rights to *redistribute* it."
A patent does not prevent someone from redistributing a work, just using it. Further, as the Sveasoft case demonstrated, the GPL does *NOT* prevent someone from making a separate agreement that prevents redistribution. However, there is actually a section in the GPL explicitly about patents though. If Novell had software that was patent encumbered, they could not distribute it under the GPL.
While open source programmers and projects probably won't be able to take advantage of defensive patent pools on an individual basis (due to the cost issues you cite), this might be useful for the open source IP indemnification company. They have to audit code for patent infringement anyway, so if they find anything potentially patentable, it might make sense to patent it. This would both prevent that method being patented by someone else (after the code audit) and potentially allow it to be used in a countersuit.
This would not be practical in all cases, but IIRC the indemnification could be for amounts greater than $3 million (and might involve a patent defense anyway). In those situations, a patent counterattack might be cheaper than paying out the suit.
I would think that the argument against this would be based more on the question of validity of open source patents. How can one patent something that has already been published in an open source project. Of course, the Kodak/Sun case had a patent on something that looked suspiciously like Smalltalk to me, so the mere prior publication of a system may not be relevant.
"what if someone did have a legitimate claim?"
That would require that software patents be legitimate. Further, as others have pointed out, this *is* a legitimate use of patents currently. Note that IBM is using patents exactly this way against SCO.
"it is certainly safe to say that through at least 26 weeks (as another poster pointed out) the womb is necessary. Why can't a women decide if before that time she wants to terminate the pregnancy? The use of said technology is a choice."
It is certainly safe to say that through the age of six months, the child cannot support itself and requires constant care. Why can't the parents decide to terminate the relationship? Doing so would be a choice. Instead, if either parent abandons the baby, we make that parent pay child support. If they kill the child (e.g. by abandoning it in a dumpster), we charge them with murder.
While abortion views are heavily influenced by religious views, there are still philosophical/legal/ethical questions involved. One, is the fetus a human being. Two, if so, is the mother obligated (by being involved in the creation of this human being) to try to bring it to term? When does this obligation start?
Note that a father is obligated to provide child support even if he has no contact with the woman after conception. He is bound by that decision.
"Actually, it is a commonly-held belief that sex crimes (ie: rape) are more about power and control than sexual gratification. I doubt that porn has much of an impact in that respect."
Porn replacements like being a peeping tom are known gateway crimes (i.e. many rapists start as peeping toms). People get into it for the sex, but stay for the power/control. Porn is a passive resource; it does not feed power/control issues.
The only observed relations between porn and sex crimes suggest that availability of porn decreases sex crimes.
"I'm sorry, getting paid overtime is a bad thing now?"
I think that you missed your parent's point. If you don't punch a time clock, then overtime is *UNpaid*, not paid. Thus, punching a time clock is better for the IT worker, as it results in wages with overtime. Not punching a clock is bad because it usually results in lots of overtime with no increase in pay.
"that's hard to refute if you look at how those voters would otherwise have leaned."
No it isn't. In both cases, exit polls showed that the third party candidate drew roughly equally from both parties. In general, most voters who strongly preferred Gore to Bush or Bush to Clinton voted for Gore or Bush Sr. What was left were a bunch of people who didn't really care.
Nader might possibly have cost Gore the Florida vote. The overall margin was so small that even a small change could have made a difference. However, most voters who would have voted for Gore if Nader weren't in the race did vote for Gore (since Nader wasn't an effective candidate). It is just as likely that Nader voters would have not voted as have voted for Gore (or Bush).
To get back to the previous point, once Perot re-entered the race, he never had a shot. At best, he was the first choice of around 25% of voters. He would have lost handily in IRV voting, as he would have been eliminated in favor of Bush and Clinton (both of whom had more first place votes). He might have won in Condorcet voting, as he was probably the second choice of most voters for the other candidates; he was a relatively centrist candidate.
In no way did Perot lose because voters did not take him seriously. Perot lost because he abandoned the election in the middle. Until then, he had been in a dead heat with the other candidates.
In a two candidate election, both candidates will race for the middle. That way, each voter counts twice, once for the candidate and once for not voting for the opponent. This is not a bad thing. It causes policy to maintain a relatively constant path. If anything, the bad part is the primary system, which puts forth more partisan candidates.
And all of your examples are for uses that are meaningless without a technical society to support them. In case of a technological collapse, you would be better off hoarding canned goods or low tech tools (e.g. a smithy) than gold. You can't eat gold, nor can you use it to make tools (it's too soft).
Gold is a luxury metal. It requires rich people (who can afford luxuries) to give it value. One of your examples is as an alternative to silver in dental fillings; yet gold is a hundred times as expensive as silver. Take away the jewelry uses of gold, and it will only have a fraction of its current value.
Yes, I saw that afterwards. Note that it bolsters my original point: that the advertised increase does not prove that the sky has started falling.
I had assumed that the argument that a 4PPM increase was unaturally high compared to 1-2PPM/year was based on real info. Instead, we see that the 4PPM increase is hardly higher than normal. My bad in assuming that they at least got the base data correct.
Why does this story exist again?
"this has resulted in nonsensical results like someone getting more time for breaking into a gas station and making off with a couple of bags of chips"
This is only nonsensical if you regard the three strikes law as punishment. It is not. The three strikes law is saying that this is a career criminal who will commit more crimes to spend more time in jail; why not just skip the "commit more crimes" part and let them spend more time in jail? Murderers are less likely to commit additional crimes in the future; murder is often a crime of passion by an otherwise law abiding citizen. Thus, it makes more sense to put the career criminal in jail (where he would likely end up anyway) than the murderer (who may have honestly reformed while in jail; i.e. it is possible that the system worked).
I'm generally against mandatory sentencing. I would prefer to leave flexibility in the system. However, the arguments that you are using suggest that mandatory sentencing may be necessary in these cases because judges (and you) are missing the point. The point is not to punish or deter crime; the point is to find recurrent criminals and get them off the streets.
Your example of the guy who got drunk and stole some chips is exactly the kind of situation that I mean. Here is someone who is clearly an idiot. I would rather catch this with a meaningless crime than by him hitting a child or driving while drunk. Clearly, jail is not a deterrent and did not have a rehabilitative effect. Jail does have a preventative effect; while there, he isn't committing crimes against the general population.
Any system is going to have overdone punishments. For example, while delivering pizzas, I was held up twice. The first time, they caught four juveniles and charged them with armed robbery. For what? Five ones and a bunch of change (about $1.50). Fortunately for them, they were juveniles, but if they had been adults, they could have gotten a long prison sentence. Because they stole $6.50? No, because they could have killed me or someone else through their foolish decision to use a gun in their robbery.
For someone who is considering a crime, jail may have a deterrent effect. For the first or second time convict, jail may have a rehabilitative effect. For the third time convict, one might as well put away those hopes. If the threat of jail didn't deter three times (possibly more; they didn't necessarily get caught for all their crimes) and the reality of jail didn't rehabilitate twice, then there is no reason to think that jail will work for them in the future. Give up on them and remove them from contact with society.
Now, if you want to argue that the war on drugs is a failed policy, that's a different matter entirely. However, I would point out that even if drugs are legalized, it is still possible to have an illegal drug problem (e.g. the UK). Because of this, you will still have the problem of young kids selling illegal drugs to bypass taxes and age limits. It is also worth noting that even legal drugs provide an incentive for crimes. Since it is hard to maintain a normal job while on high levels of drugs (and the drugs themselves are not free), there is a strong incentive to get a high profit job, e.g. theft or fraud, to pay for one's habit.
"So what makes you so sure 4PPM is insignificant here?"
The link your parent posted showing that 4 PPM increases have been seen in a single month? The problem is not the 4PPM increase, it's the possibility that the increases will have permanently changed from 1-2 PPM/year to 4PPM/year or more.
"we'll have stupidity such as the three strikes law. This is law based on a baseball term! Talk about stupidity ruling the unwashed masses!"
By your own standards, the three strikes law is not stupid. The three strikes law is not intended to be punitive or rehabilitative; it is purely preventative. The issue is that the majority of crimes are committed by repeat criminals. The three strikes rule takes someone who is recidivist (i.e. has a history of committing crimes after being released from jail) and stops releasing them to commit more crimes. It has a much lower incidence of false positives (innocents imprisoned), because it requires multiple convictions; it is unlikely that a totally innocent defendant will be convicted three times.
Good marketing says nothing about an idea's merits: positive nor negative. The basis of the name is irrelevant. What's important is that this actually addresses the problems of recidivist criminals, something that existing punitive and rehabilitative actions do not.
It's not so much efficiency as recognition of desires. It is somewhat similar to the issue with IRV. IRV is bad because some of the time, it eliminates a candidate on whom voters would be more willing to compromise. Similarly, the reverse commons effect underspends on a particular service even though all taxpayers would be willing to pay more for more service. The IRV issue can be solved by switching to Condorcet voting, which scores the same votes, only better.
Another example of a similar problem is blind auctions (like eBay). There is a strong incentive to lie about the price you are willing to pay in an auction. If your bid is higher than everyone else's, you win the auction, regardless of whether it is 1 penny higher or a million dollars. As a result, there is a strong tendency to bid just a little higher than you expect everyone else to bid. Unfortunately, everyone else has the same incentive, so they also underbid. As a result, the seller gets screwed. eBay solved this by awarding to the highest bidder at the price of the second highest bid (a Vichy auction; named after the Nobel laureate who thought of it).
In a Vichy auction, there is no incentive to underbid, as your bid *never* determines the price you pay (except as an upper limit), just whether or not you win the bid. As a result, both sides bid higher and the seller makes more money.
What I am saying is that the reverse commons effect does not take into full account the preferences of the purchasers. As a result, they will tend to prefer the government run system which better reflects their preferences. As such, it is harder to switch systems (practical arguments tend to trump moral arguments; e.g. many northern abolititionists were swayed more by a fear of slave labor competing in factories than by the moral wrongness of slavery). One should thus keep an eye out for solutions to this.
"Simply edit the url you are currenlty in and remove the part before slashdot.org"
But then it appears separately in my browser history, which makes it harder to find the sites that I would really like to visit. If it is a clickable link (like the five I made), then it doesn't appear separately. Given the way that I use my browser (frequent visits to a constantly changing list of recently visited sites), it is actually easier for me to post my own clickable link and use that than it is for me to edit the URL manually in the location bar.
It would of course be a lot easier for me if I could set my preferences to use a particular color scheme instead of the IT color scheme. However, there doesn't seem to be much interest in that.
"I hate the IT color scheme, too."
What happened to all the posts with links to alternative color schemes? I used to be able to quickly scan through the story and find links like
Readable if garish
Traditional
Readable and pretty
Valiantly defiant
Gross but accurate
all over the place. Where have they gone?
Is slashdot really better off making people find a CSS hack that changes the color scheme (and disables all the ads) on it. pages? Wouldn't it make more sense to enourage the simple use of alternative color schemes and avoid more complicated methods that also interfere with revenue? If I patch slashcode to allow the color scheme to be user specified, will /. adopt it into the main trunk? What is needed here?
P.S. I'm holding out for a phone that runs embedded Linux. Do I really need a phone that runs embedded Linux? Of course not, but I want one anyway.
There is alrady a patent clause in the GPL. The current clause says that software cannot be distributed under the GPL if it is patent encumbered. It seems (did not RTFL) this license goes a step further and includes (and requires!) a patent license as part of it.
Not sure how the FSF will proceed here. The comments they sent me regarding the Sveasoft abuses of the GPL implied that they did not consider a license that virally applies contract terms to be valid. I.e. they would not be able to enforce a license that demands that patent rights be waived if you redistribute. The current GPL works the other way; if you hold patent rights, you can't redistribute under the GPL.
Note: the issue here is not Microsoft. The issue arises when IBM redistributes the software. Does it automatically get protection from IBM's patent portfolio?
Neither is a browser. P2P allows for three classes of exploits, all of which they describe.
The main reason that businesses should keep P2P software off their machines is that it makes them vulnerable to lawsuits by copyright holders (the legal exploit). *All* P2P programs are vulnerable to that and the spoofed content (social) exploit. The technical exploits (which are program specific) are much less serious in comparison.
This is not a technical article. They aren't giving prizes to the best exploits. It is simply a discussion of various things people can do to minimize the exploitability of their computer systems. It includes things like being sued for copyright infringement as an "exploit" (presumably because it does harm to the system owner).