"Beacause you can't make money - only take it from someone else"
This is incorrect. If the productive capacity increases, one can print money (up to the amount of the productivity increase) without causing inflation. In fact, not doing so causes deflation (and probably a recession).
It is also worth noting that what is important is the *effective* money supply, not the printed money supply. A catastrophic drop in the effective money supply was the problem in the 1929-33 period in the US, exacerbated by the Fed doing the wrong thing (selling government bonds rather than buying them, further shrinking the money supply).
This is attacking them on the correct front. This is cancelling an abuse of monopoly power. That is the important precedent that is being set. The fact that it could allow other companies to abuse the patent system is irrelevant. If Microsoft does not have enough resources to protect itself from bad patents without abusing monopoly power, then what hope do the rest of us have?
To get back to the war analogy, this is like forcing a small country to disarm its nukes (monopoly power). This has the side effect of making the small country more vulnerable to attack (software patent lawsuits). However, that does not make nuclear disarmament (anti-trust) pro-war (software patents).
Your original point remains true. Some of the people who support this are doing so for the wrong reasons. However, it is perfectly reasonable to be glad that Japan is forcing Microsoft to clean up its contracts and be against software patents.
One of my mother's favorite stories is about how I used the neighbors' Ken doll to play Superman when I was four. That said, when I was on the math team in high school, I went to a competition with about twenty people. The two females tied for third, one point behind the two people who tied for first. Point being that the only females who participated were those who were at the upper end of the distribution group (one was the youngest person there, a sophmore in a group of mostly seniors).
Sample size is too small to make real assertions, but if the pattern holds over larger samples as well, it would suggest that only the excellent persist past some barrier. If it were a natural distribution of a group with a lower mean or variance, we would expect a more even distribution or one weighted towards the bottom.
"If you are against software patents, its hypocritical to want MS to suffer from them."
Anything that makes Microsoft oppose software patents is good in my book. I don't want a software patent system with which Microsoft is comfortable; I want software patents out of the system.
Its like waging war against a country that is on the offensive (e.g. Japan and Germany in the '40s; Iraq in the early '90s). The real issue here, as regards Microsoft, is abuse of monopoly power. I would be satisfied with this clause being removed from the contract and the software patents being ruled invalid. Software patents are wrong and monopoly abuse is wrong. Two wrongs does not make a right. There is no reason to give Microsoft a pass on monopoly abuse, even if they are using that monopoly abuse to protect themselves against patent system abuse.
Finally, are we sure that these are software patents (I didn't RTFA)? Electronics makers are saying that Microsoft is in violation. Microsoft does make electronics equipment (e.g. XBox). These could be patents on circuits, etc. Even if this regards software patents, is the clause restricted to software patents?
P.S. I will metamod your Troll as Unfair and Insightful or Interesting as Fair. While I disagree, you clearly weren't trolling.
"trying all the possible 2-5 digit extensions in hopes of a hit"
Further, even if it did become common enough to support this, brute force attempts are relatively easy to recognize. If a non-whitelisted IP tries to connect more than 5 times (adjust number as necessary), don't let it connect for 10 minutes. If it repeats, blacklist it. This would allow direct IP/IP connections without requiring the network authentication (e.g. Vonage).
"is there any reason to completely fill up the RAM slots with 256Mb DIMMs when 512MB are easily available?"
From the article: "ensuring that the chipset can run in dual-channel mode"
Dual channel mode requires an even number of memory bars, as it allows for two memory paths with balanced memory in each. Dual channel is faster than single channel memory, so one would always want to run in dual channel mode if the motherboard supports it.
"oh, and a google search will show you that, at least last year, only 6% of spam is Chinese , 58% was American."
I think that he was referring to relays rather than sources. More than 58% of spam advertises people with an American presence; otherwise, they couldn't get the money from Americans. It's the person who receives the American money that needs to be prosecuted. If spammers couldn't collect money, there would be no need for spam. Sending machines can be compromised. It's much harder to do so with bank accounts (or if you could, why would you waste time just using it for spam; take over Gates' account and spend his billions).
"never see any good mail end up in the SPAM folder - with the spam level set to 3.9."
I've seen good mail in my spam folder at both 4.0 and 5.0.
It's worth noting that different people get different levels of mail/spam. Your parent mentioned 100,000 spams per week. At 10% delivered, this would be 10,000 spams. Even if the filters block 99% of that, it's still 100 spams per week.
It also allows vote selling. Note that I can agree to vote in a certain way; do so; and prove that I did so (by giving the buyer my private key). Thus, the system is not fully anonymous; I can choose to give up the anonymity of my vote--something that is currently impossible (once the vote is cast, no one, not even me, can determine my actual vote). Heck, I could give the actual key to someone and let them place the vote.
Another issue is the possibility of hijacking uncast votes. Since those people weren't intending to vote, it is unlikely that they will check to make sure that they didn't. As such, it is not very secure.
I am also unconvinced that replacement keys could be provided safely without compromising anonymity. A replacement key could only be provided safely if the previous key is cancelled. To cancel the previous key, we need to know what it was. To know what it is, there must be some way of deriving it from a person's identity. If so, then it's not anonymous. The only way replacement keys work is if the key hasn't been activated yet. If it hasn't been activated, then there is no reason to disburse it.
Republicans have the same issues as democrats. Even if the resolution last time helped Republicans (by tossing out votes for the leading Democrat), it still was both scary and controversial. Say what you will about evoting, but it doesn't allow for 2000 to happen. No paper trail means no recounts and thus no hanging chad issues.
The 83% number alone pretty much precludes there being a big margin between democrats and republicans. Apparently a majority of each think that evoting is safe.
It doesn't matter how long the keys are if you have the private key (which must exist at the decryption point). Any evoting system is vulnerable to human attack. By centralizing the system, the whole system is vulnerable at that single point.
Not to mention that the individual votes are easily hijackable if the cards are sent by mail. Especially considering that only about 50% of people actually vote. IRS stuff is not equivalent, since their is no anonymity; the IRS knows who you are. If money disappears out of your account (or doesn't appear), you know that something went wrong. If your vote is miscast, there is no way to tell.
Paper ballots dropped in a box are anonymous, because one doesn't know exactly how the ballot fell inside the box. They are secure because one can keep the ballots physically secure. Further, the very distributed nature makes it more secure. One cannot cause massive changes without corrupting multiple locations.
We could do a straight vote now. We just need to change the constitution. Other than absentee ballots, the vote count is known next day anyway.
"established and reputable cryptographers have proved that it is possible to have anonymous auditable secure voting systems"
Please post citations (links) to such proofs.
The issue is that if I can't view how my vote and others are actually counted, there is no way to establish that the original set of votes was not just replaced by another set of votes. Alternately, if the proof exists that my vote was cast the way that I intended, then it be backtracked to my actual vote. Please demonstrate how it is possible to determine the effect of my individual vote (what I need to verify) without allowing someone other than me to determine the effect of the original vote (which breaks anonymity).
A paper is no less anonymous than evoting. A paper trail only creates problems if the votes are tagged by voter ID or are kept in vote order. Note that the same problems exist with evoting as well.
It's not a paper vs. electronic issue; it's a good vs. bad implementation issue.
I'm a US citizen, and I've never filled out a ballot with more than thirty choices. Further, it is quite possible to manually count the ballots within a day, as counting is simpler than actual voting (which is done within a day). That said, there's nothing wrong with automated ballot counting, so long as it is accurate, verifiable, and anonymous. The problem is that most e-voting systems lose one or more of those characteristics (in particular, a method that only relies on evoting with no paper trail is not verifiable).
The simple answer to this is to use a paper ballot and then run the ballot through a counting tool in the presence of the voter. This allows the voter to notice if they accidentally voted for Buchanan instead of Gore (or if it can't count the vote because there are marks for both). If so, then the ballot should be destroyed and redone. Alternately, the evoting machine can print out a ballot which the voter can verify and then drop in a box. With that method, an immediate count can be obtained at the end of the voting period, and a recount can be done from the paper ballots (if necessary). With the first method, the recount can be done quicker but the original count takes as long as a recount.
I think that you missed the point. Why did they pick a domain name that was in use for the book? They could have easily named the book something else (KTchat.net makes more sense to me in terms of the book's subject and is currently available). For that matter, the actual Katie Tarbox has the web site KatieT.com... why not use that for the book's name?
"Nothing else matters to Penguin"
Then perhaps a system change is needed? Perhaps one should only be able to link an extant domain name to a book about *that* domain and its actual content.
'(they're called Liberal, but really should be "wannabe-Republicans")'
In any country *except* the US, Liberal refers to someone with liberal economic views. In the US, liberal refers to someone with liberal morality views. Note that in both cases, liberal means "free from government interference." In the US, the libertarians are the only ones who are liberal in both senses.
It's just like how in almost any country but the US, football refers to a game where one mostly plays by kicking the ball, as opposed to a game that is mostly played by holding the ball in one's hands with the occasional kick as part of scoring and change of posession.
"would happily let people drive at 35mph in a 35mph limit in thick fog on a snowy road, but would stop them from driving at 40mph on the same road in clear weather"
Whereas the current system allows people to drive 70 mph in fog on a snowy road, because the speed traps don't work in fog. Meanwhile, it stops people in clear weather for going 11 mph over the limit (generally not 9, because 9 doesn't hold up in court). Even if the police can measure speed in bad conditions, they still don't have authority to adjust the speed limit in most states in the US.
Laws don't work when they are subjective. At least with this system, there is the possibility of using weather info as part of the determining factor for the speed limit. The current system does not: the sign always shows the same speed.
The black box method is not the same as traffic cameras. By their nature, traffic cameras are restricted in space and occasional in enforcement. The black box method is constant in enforcement and goes where the car goes.
"You see speeding tickets are another form of regressive tax, many local governments use them to fill a shortfall in revenue and they hit low and middle income people the hardest."
This would not be true if tickets were automatic. Low and middle income people would watch their speed (even high income people would need to watch their speed if they don't want their license suspended). Further, tickets would stop being revenue generators and speed limits would be created because they made sense rather than to encourage flounting (you know, like the four lane highway that drops to a speed limit of 25 MPH in one township; people go 40 hoping not to get caught). Speeding tickets as revenue generators only work when there is a possibility of not getting caught.
The privacy concerns are another issue, but this system would be fairer than what we have now, where the police can selectively enforce speed limits.
"one of the idiots that rear-ended me tried to claim that I was backing up"
I've got a better one. There was an accident where a car sideswiped a van. The driver of the car claimed that the car was stopped and the van moved *sideways* to hit it.
I'm disappointed that it isn't Debian. IIRC, HP has a representative on the Debian Desktop project. The worst part of working with Debian is the install and hardware configuration (both of which would be done for you here; just add a recovery disk and a few CDs as a local apt-get repository and off you go).
I would seriously consider a preinstalled basic Debian for a dual boot system with XP Pro. With Suse, I would just get XP Pro and add Suse afterwards...the preinstall isn't as helpful.
Debian is also free beer, which would allow the price to be lower than its MS Windows equivalent.
Slashdot Mirror
"Beacause you can't make money - only take it from someone else"
This is incorrect. If the productive capacity increases, one can print money (up to the amount of the productivity increase) without causing inflation. In fact, not doing so causes deflation (and probably a recession).
It is also worth noting that what is important is the *effective* money supply, not the printed money supply. A catastrophic drop in the effective money supply was the problem in the 1929-33 period in the US, exacerbated by the Fed doing the wrong thing (selling government bonds rather than buying them, further shrinking the money supply).
According to http://sp2torrent.com/index.php Microsoft already has stopped them (at least from publishing the torrent link at that site).
This is attacking them on the correct front. This is cancelling an abuse of monopoly power. That is the important precedent that is being set. The fact that it could allow other companies to abuse the patent system is irrelevant. If Microsoft does not have enough resources to protect itself from bad patents without abusing monopoly power, then what hope do the rest of us have?
To get back to the war analogy, this is like forcing a small country to disarm its nukes (monopoly power). This has the side effect of making the small country more vulnerable to attack (software patent lawsuits). However, that does not make nuclear disarmament (anti-trust) pro-war (software patents).
Your original point remains true. Some of the people who support this are doing so for the wrong reasons. However, it is perfectly reasonable to be glad that Japan is forcing Microsoft to clean up its contracts and be against software patents.
"but then I brought them Barbie dolls"
One of my mother's favorite stories is about how I used the neighbors' Ken doll to play Superman when I was four. That said, when I was on the math team in high school, I went to a competition with about twenty people. The two females tied for third, one point behind the two people who tied for first. Point being that the only females who participated were those who were at the upper end of the distribution group (one was the youngest person there, a sophmore in a group of mostly seniors).
Sample size is too small to make real assertions, but if the pattern holds over larger samples as well, it would suggest that only the excellent persist past some barrier. If it were a natural distribution of a group with a lower mean or variance, we would expect a more even distribution or one weighted towards the bottom.
"If you are against software patents, its hypocritical to want MS to suffer from them."
Anything that makes Microsoft oppose software patents is good in my book. I don't want a software patent system with which Microsoft is comfortable; I want software patents out of the system.
Its like waging war against a country that is on the offensive (e.g. Japan and Germany in the '40s; Iraq in the early '90s). The real issue here, as regards Microsoft, is abuse of monopoly power. I would be satisfied with this clause being removed from the contract and the software patents being ruled invalid. Software patents are wrong and monopoly abuse is wrong. Two wrongs does not make a right. There is no reason to give Microsoft a pass on monopoly abuse, even if they are using that monopoly abuse to protect themselves against patent system abuse.
Finally, are we sure that these are software patents (I didn't RTFA)? Electronics makers are saying that Microsoft is in violation. Microsoft does make electronics equipment (e.g. XBox). These could be patents on circuits, etc. Even if this regards software patents, is the clause restricted to software patents?
P.S. I will metamod your Troll as Unfair and Insightful or Interesting as Fair. While I disagree, you clearly weren't trolling.
"trying all the possible 2-5 digit extensions in hopes of a hit"
Further, even if it did become common enough to support this, brute force attempts are relatively easy to recognize. If a non-whitelisted IP tries to connect more than 5 times (adjust number as necessary), don't let it connect for 10 minutes. If it repeats, blacklist it. This would allow direct IP/IP connections without requiring the network authentication (e.g. Vonage).
"is there any reason to completely fill up the RAM slots with 256Mb DIMMs when 512MB are easily available?"
From the article: "ensuring that the chipset can run in dual-channel mode"
Dual channel mode requires an even number of memory bars, as it allows for two memory paths with balanced memory in each. Dual channel is faster than single channel memory, so one would always want to run in dual channel mode if the motherboard supports it.
"oh, and a google search will show you that, at least last year, only 6% of spam is Chinese , 58% was American."
I think that he was referring to relays rather than sources. More than 58% of spam advertises people with an American presence; otherwise, they couldn't get the money from Americans. It's the person who receives the American money that needs to be prosecuted. If spammers couldn't collect money, there would be no need for spam. Sending machines can be compromised. It's much harder to do so with bank accounts (or if you could, why would you waste time just using it for spam; take over Gates' account and spend his billions).
"never see any good mail end up in the SPAM folder - with the spam level set to 3.9."
I've seen good mail in my spam folder at both 4.0 and 5.0.
It's worth noting that different people get different levels of mail/spam. Your parent mentioned 100,000 spams per week. At 10% delivered, this would be 10,000 spams. Even if the filters block 99% of that, it's still 100 spams per week.
"Post-IPO gains in a stock's price doesn't [sic] put money into the corporation's hands."
Unless they sell more stock.
It also allows vote selling. Note that I can agree to vote in a certain way; do so; and prove that I did so (by giving the buyer my private key). Thus, the system is not fully anonymous; I can choose to give up the anonymity of my vote--something that is currently impossible (once the vote is cast, no one, not even me, can determine my actual vote). Heck, I could give the actual key to someone and let them place the vote.
Another issue is the possibility of hijacking uncast votes. Since those people weren't intending to vote, it is unlikely that they will check to make sure that they didn't. As such, it is not very secure.
I am also unconvinced that replacement keys could be provided safely without compromising anonymity. A replacement key could only be provided safely if the previous key is cancelled. To cancel the previous key, we need to know what it was. To know what it is, there must be some way of deriving it from a person's identity. If so, then it's not anonymous. The only way replacement keys work is if the key hasn't been activated yet. If it hasn't been activated, then there is no reason to disburse it.
That's not evidence. That's rampant speculation.
Republicans have the same issues as democrats. Even if the resolution last time helped Republicans (by tossing out votes for the leading Democrat), it still was both scary and controversial. Say what you will about evoting, but it doesn't allow for 2000 to happen. No paper trail means no recounts and thus no hanging chad issues.
The 83% number alone pretty much precludes there being a big margin between democrats and republicans. Apparently a majority of each think that evoting is safe.
It doesn't matter how long the keys are if you have the private key (which must exist at the decryption point). Any evoting system is vulnerable to human attack. By centralizing the system, the whole system is vulnerable at that single point.
Not to mention that the individual votes are easily hijackable if the cards are sent by mail. Especially considering that only about 50% of people actually vote. IRS stuff is not equivalent, since their is no anonymity; the IRS knows who you are. If money disappears out of your account (or doesn't appear), you know that something went wrong. If your vote is miscast, there is no way to tell.
Paper ballots dropped in a box are anonymous, because one doesn't know exactly how the ballot fell inside the box. They are secure because one can keep the ballots physically secure. Further, the very distributed nature makes it more secure. One cannot cause massive changes without corrupting multiple locations.
We could do a straight vote now. We just need to change the constitution. Other than absentee ballots, the vote count is known next day anyway.
"established and reputable cryptographers have proved that it is possible to have anonymous auditable secure voting systems"
Please post citations (links) to such proofs.
The issue is that if I can't view how my vote and others are actually counted, there is no way to establish that the original set of votes was not just replaced by another set of votes. Alternately, if the proof exists that my vote was cast the way that I intended, then it be backtracked to my actual vote. Please demonstrate how it is possible to determine the effect of my individual vote (what I need to verify) without allowing someone other than me to determine the effect of the original vote (which breaks anonymity).
A paper is no less anonymous than evoting. A paper trail only creates problems if the votes are tagged by voter ID or are kept in vote order. Note that the same problems exist with evoting as well.
It's not a paper vs. electronic issue; it's a good vs. bad implementation issue.
I'm a US citizen, and I've never filled out a ballot with more than thirty choices. Further, it is quite possible to manually count the ballots within a day, as counting is simpler than actual voting (which is done within a day). That said, there's nothing wrong with automated ballot counting, so long as it is accurate, verifiable, and anonymous. The problem is that most e-voting systems lose one or more of those characteristics (in particular, a method that only relies on evoting with no paper trail is not verifiable).
The simple answer to this is to use a paper ballot and then run the ballot through a counting tool in the presence of the voter. This allows the voter to notice if they accidentally voted for Buchanan instead of Gore (or if it can't count the vote because there are marks for both). If so, then the ballot should be destroyed and redone. Alternately, the evoting machine can print out a ballot which the voter can verify and then drop in a box. With that method, an immediate count can be obtained at the end of the voting period, and a recount can be done from the paper ballots (if necessary). With the first method, the recount can be done quicker but the original count takes as long as a recount.
Someone (who doesn't have any of these shares) might sue afterwards because Google misrepresented its outstanding shares (or something similar).
"(My apoligies to the D.C. resident/business whose number that is)."
So why didn't you use (202)555-0040? That would be a phone company number (if anything).
Or you could have just published Penguin's number, but that would have required research.
I think that you missed the point. Why did they pick a domain name that was in use for the book? They could have easily named the book something else (KTchat.net makes more sense to me in terms of the book's subject and is currently available). For that matter, the actual Katie Tarbox has the web site KatieT.com ... why not use that for the book's name?
"Nothing else matters to Penguin"
Then perhaps a system change is needed? Perhaps one should only be able to link an extant domain name to a book about *that* domain and its actual content.
'(they're called Liberal, but really should be "wannabe-Republicans")'
In any country *except* the US, Liberal refers to someone with liberal economic views. In the US, liberal refers to someone with liberal morality views. Note that in both cases, liberal means "free from government interference." In the US, the libertarians are the only ones who are liberal in both senses.
It's just like how in almost any country but the US, football refers to a game where one mostly plays by kicking the ball, as opposed to a game that is mostly played by holding the ball in one's hands with the occasional kick as part of scoring and change of posession.
"would happily let people drive at 35mph in a 35mph limit in thick fog on a snowy road, but would stop them from driving at 40mph on the same road in clear weather"
Whereas the current system allows people to drive 70 mph in fog on a snowy road, because the speed traps don't work in fog. Meanwhile, it stops people in clear weather for going 11 mph over the limit (generally not 9, because 9 doesn't hold up in court). Even if the police can measure speed in bad conditions, they still don't have authority to adjust the speed limit in most states in the US.
Laws don't work when they are subjective. At least with this system, there is the possibility of using weather info as part of the determining factor for the speed limit. The current system does not: the sign always shows the same speed.
The black box method is not the same as traffic cameras. By their nature, traffic cameras are restricted in space and occasional in enforcement. The black box method is constant in enforcement and goes where the car goes.
"You see speeding tickets are another form of regressive tax, many local governments use them to fill a shortfall in revenue and they hit low and middle income people the hardest."
This would not be true if tickets were automatic. Low and middle income people would watch their speed (even high income people would need to watch their speed if they don't want their license suspended). Further, tickets would stop being revenue generators and speed limits would be created because they made sense rather than to encourage flounting (you know, like the four lane highway that drops to a speed limit of 25 MPH in one township; people go 40 hoping not to get caught). Speeding tickets as revenue generators only work when there is a possibility of not getting caught.
The privacy concerns are another issue, but this system would be fairer than what we have now, where the police can selectively enforce speed limits.
"one of the idiots that rear-ended me tried to claim that I was backing up"
I've got a better one. There was an accident where a car sideswiped a van. The driver of the car claimed that the car was stopped and the van moved *sideways* to hit it.
"glad to see that they are using Suse"
I'm disappointed that it isn't Debian. IIRC, HP has a representative on the Debian Desktop project. The worst part of working with Debian is the install and hardware configuration (both of which would be done for you here; just add a recovery disk and a few CDs as a local apt-get repository and off you go).
I would seriously consider a preinstalled basic Debian for a dual boot system with XP Pro. With Suse, I would just get XP Pro and add Suse afterwards...the preinstall isn't as helpful.
Debian is also free beer, which would allow the price to be lower than its MS Windows equivalent.
"The fact remains that this concept had been percolating prior to his employment."
Is that clearly a fact? If so, why didn't he exclude the idea when signing the employment contract?