OTOH, Instant Runoff is simple enough that you don't need a computer to tally the votes.
I've always imagined what a nightmare it would be to audit an instant runoff election with a large number of candidates. In Condorcet, you simply have to tabulate the number of voters preferring X to Y vs. those preferring Y to X, for each pair of candidates X and Y. With 10 candidates, that's 180 pieces of information.
A 10-candidate IRV race can require elimination of up to 8 candidates before there is a guaranteed majority winner (or possibly a final 2-candidate tie if the number of ballots is even). The number of voters that prefer X to all remaining candidates shifts at each stage, so you have to record a set of scores for all remaining candidates at each stage. That's 10 + 9 + 8 +... + 2 or 54 pieces of information -- less than Condorcet.
But the kicker is this: In Condorcet, you create that 10x10x2 matrix (with 20 empty positions because you don't compare a candidate with himself) just once. It then contains all the information needed to complete the audit. Once you're satisfied that the matrix is correct, you don't have to look at the ballots again. In IRV, you have to have access to the full set of ballots at every stage. You count the ballots 8 times, each time ignoring the candidates eliminated by the previous counts. So you can't toss the ballots aside until you've completed the entire audit.
And forget about just aggregating the number of voters that vote a certain ranking. 10 factorial is quite a large number. You'd have to allocate a lot of memory (OK, in a computer program you'd use a hash table, but how much more efficient is that than simply having a record per ballot? My point is that Condorcet sidesteps these difficulties; you just read in a record, add the points to the matrix, and move on to the next record.)
California had an election not to long ago with 135 candidates. I'd hate to see that done in instant runoff.
Condorcet is not the simplest election method, but it is hands down the most accurate.
Well, I wouldn't call it most accurate. First, you've got to look at whether clones can affect the outcome. If some subset of the candidates have virtually indistinguishable positions, then adding or removing candidates in this subset can affect the outcome when there are nontransitive cycles such as in your example. They can also affect the way people vote, if people consider the possibility of such cycles (then they may try to game the cycle resolution method by reversing some of their preferences). On top of that, there are many, many ways to resolve cycles, and they will give different results in certain cases, so you would have to decide which resolution method is "most accurate" before you could start comparing Condorcet-class methods to other methods.
Then you've got to consider different categories of balloting. Basically, you've got ranked methods and rating methods. Condorcet, Borda, IRV, and many others are ranked methods. Plurality and approval are rating methods (though in those cases the ratings are binary, with plurality additionally constrained).
I tend to think that you can't compare methods for "accuracy" unless you can eliminate voter strategizing. Voters won't strategize if they have no clue how other voters are likely to vote. In those circumstances, I think approval is the most accurate rating method, and Borda is the most accurate ranked method. But those circumstances are rare to nonexistent in the real world.
But I like approval best (even over ranked methods), because any strategizing that occurs in approval tends to be benign.
Nah. Instant runoff is fine with three candidates. With more than three, you start having voters strategize their rankings. For example, if I like Kucinich > Edwards > Obama > Clinton (not my actual preference order, but it illustrates my case), I may nevertheless list Obama 2nd and Edwards 3rd in the primary because Obama has a better chance of beating the front runner Clinton. The more candidates you add, the more this kind of gaming of the vote becomes necessary.
Approval voting is much better. I could, for instance, approve Kucinich, Edwards, Obama, Dodd, and maybe Biden and Richardson, in order to maximize my insurance against Clinton.
Excuse me, but this is an open forum. If you don't like that, or if you can't take the criticism that goes with the territory, you don't have to post here. I haven't made any statements I'm not willing to defend, so I don't see why you have a problem with me or why you have to label my "style" as "hit-and-run". Where did I run to? I'm still here!
I've come to the conclusion that you're just plain rude, and trollish too, since you don't seem to be willing to back up your assertions.
First, I cite common sense. There is a difference between being a racist and acting out a racist agenda. The first doesn't (of itself) land you in jail.
Additionally I cite Rehnquist in Wisconsin v. Mitchell. If you don't care to read it all, here's the executive version: The high court did not find that the defendant's First Amendment rights were infringed by the legislation, and found the sort of chilling effect on expression that you envision too speculative for consideration in their opinion.
Surely, if the expression of racist thoughts isn't substantially chilled by legislation of this nature, then the thoughts themselves (minus the expression) are not chilled, either.
Of course, it's possible to envision legislation that is overly broad in this area. But I'm not advocating that kind of legislation, nor am I aware of any proposed or on the books.
No, the difference there is not motivation but intent. Did you intend to kill the person? If you did intend to do so, did you do it on the spur of the moment or was it done in a planned out, cold-blooded fashion?
You've made a distinction without a difference. Both motivation and intent require a jury to make a determination of the defendant's state of mind. In fact juries are allowed to consider motivation in their deliberations; if prosecution can't establish a motive then it may cast reasonable doubt on the case (depending on the strength of other evidence, of course).
No, if he were just on trial for his actions then it would be a normal assault/robbery/arson/murder trial, not a hate crime trial.
It *is* a "normal assault/robbery/arson/murder trial", with the additional charges that the defendant is acting out a specific agenda. Both require a jury determination that the standard burden of proof is met. A jury can still convict on the former charge and acquit on the latter. If prosecution doesn't think they can meet the burden for the latter then they shouldn't bring that charge in the first place. If prosecution does bring the hate crime charges while having insufficient evidence then that is a separate issue, because we already have similar instances of overzealous prosecution with all manner of crimes. If you are looking for a perfect world where that doesn't happen, I can't help you.
At the end of the day, what good does hate crime legislation do? All it means is that a racist might spend 20 years in jail for the same exact crime that a non-racist would only spend 15 years in for. So, when it comes down to that, he's spending 5 years in jail just for being a racist.
No, he's spending the extra 5 years for acting out a racist agenda.
And HOW is that not "thought crime?"
Because there is no law against his thoughts, and his thoughts aren't policed. I think you are using a very loose definition of "thought crime".
I think slashdot moderators hate me, why am I modded a 1 and the comment I'm replying to a 2??
I think that's called a karma bonus modifier; my post wasn't moderated either way.
We're both right. It just depends on what is meant by "delivery". Delivery is out of order on the wire, and that's what I (and I suspect the original poster) meant. Of course the protocol stack guarantees in-order delivery to the higher layers. That means that the slowest packet is going to dictate the latency of an entire transfer, which I think is what he was getting at.
That said, I don't think ordering is a big deal for FCoE. For FC, in-order delivery (on the wire, that is) isn't necessarily required. Class-2 and Class-3 FC do not guarantee it. So at least these two classes should map to FCoE easily enough, and I'm sure a TCP-like reordering capability in the FCoE encapsulation could enable mapping of Classes 1, 4, and 5.
You've got some serious issues if you have packets that are out of order on your local network.
I don't think that would necessarily indicate *serious* issues. A local switched network for a mid-sized company or college campus could certainly be complex enough that a little temporary congestion could affect arrival order.
Disclaimer: I'm neither an FC nor Ethernet expert, but I've had just enough exposure to both to be dangerous.
It doesn't *deliver* packets in order (at least, not unless the underlying network does). It provides the capability to reconstruct the original order. GP was talking about *delivery* of packets.
Single-platform?
OpenFCoE may be a Linux software initiative, but I think the T11 FCoE standard on which it is based is being developed for all modern major server platforms.
In a legal sense, a "fact" is something to be determined by a jury. A jury may look at evidence for and against the prosecution's theory of what happened (including the motivation) and decide that it fits the facts of the case. If they can't find reasonable doubt, they accept a proposition as factual in producing their verdict. This is the power given to juries.
IANAL, but come on, we've all watched enough TV to know this stuff...
So in your mind, juries should never take into account a defendant's state of mind? That would wipe out many of the distinctions commonly made today between first and second degree murder and manslaughter, as well as consideration of mitigating and aggravating circumstances in violent crimes in general.
Contrary to the popular notion, hate crimes are not thought crimes. The defendant in such crimes is always on trial for his actions, not his thoughts. Juries are allowed to make factual findings in their deliberations, subject to the usual burden of proof ("beyond reasonable doubt"), and state of mind is a factual finding of this nature. But I know of no case in the U.S. where someone has been brought to trial just for harboring hateful thoughts.
You're right, of course, they are separate issues. However, Blockbuster's business model doesn't have any significant advantage (for me) over Netflix's. Between the two busines models, I'm neutral as a customer. Given my prior reason not to use Blockbuster, I rent from Netflix.
As an investor, I might also find that the lower overhead of the Netflix model is an advantage.
I object to government censorship of movies and ideas, as well as to the crowd-hysteria brand of censorship. If the government were to prevent a studio from releasing a film, I would object just as strongly. However, the citations you gave seem to tell of one individual who put public pressure on a studio to bow out. And the individual happened to be a state assemblyman. Is this an instance of one individual exercising his free speech rights, or is this government censorship? Maybe it's the latter, but somehow I doubt that the state of New York really has that kind of power.
More accurately, this would be the power of the bully pulpit, which of course falls under my original objection. I can't tell from those two blurbs whether the assemblyman's comments were a factor in Fox's decision, or whether they made the decision for internal reasons of their own. If it's the former, then we can say Fox caved just like Blockbuster, which would be a strike against Fox (and not necessarily against all of Hollywood as you originally posted).
I don't know if they "banded together" or not. I won't ask for a citation because such banding is not relevant; if they did then they were all presumably willing participants. Under that presumption none of the studios were censored by this "band". Ask anyone -- or any group/band/cartel or whatever noun you care to apply -- to publish something; they are free to say yes or no. And Blockbuster is free to choose the movies they carry, I'll grant them that. But someone else -- not Blockbuster -- decided that Blockbuster shouldn't carry a certain movie, and Blockbuster caved. That is enough of a flaw in Blockbuster's moral fiber to motivate me to rent elsewhere.
These two cases aren't the same. One case is self-censorship. I'm not even sure that such a broad use of the term "censorship" is a good thing. Maybe "self-representation" is more accurate. I fully support anyone's right to publish or not publish submitted materials, as they see fit. Call it an agenda if you like, but it doesn't matter: it's a basic freedom. The other case is caving to blackmail. It only opens the door to further blackmail and reduction of freedom.
I don't have a problem with Company X choosing not to produce Movie Y because they don't see commercial potential. Studios have to make those decisions all the time (and bad business decisions are inevitable). After all, they have to put up a lot of resources to get a movie made. But if Movie Y does get made, and Company Z, that is in the mainstream movie rental business, refuses to make that movie available to its customers, only to appease some rabid bunch of loonies, then I have a problem with Company Z. I can also understand it if Movie Y is in some niche market that Company Z doesn't play in. It's the caving in to special interest threats like these that I don't like (this holds, whether or not I had any interest in seeing Movie Y).
I never liked Blockbuster because of their past history of censorship (see "Last Temptation of Christ"). Maybe they don't do that anymore but there's still a bad vibe for me there. Before Netflix, I always preferred Hollywood Video. And with Netflix, I've never had a problem getting the movies I want. Why take the trouble with going to the video store -- one or two movies delivered to my mailbox every week is as much as I ever find time to watch these days, anyway.
Ah, but the irony would be lost. We could name it after some innocent victim like Laci Peterson and it would pass for sure. But the same law named after a victim who also happened to be guilty of a sex crime (even a less serious one) would almost certainly fail, even though it covers exactly the same people.
I hope that post is in the spirit of a "modest proposal". Let's have a look: jail overcrowding -- Decriminalize recreational drug use; revoke three-strikes laws. inmate abuse -- Since jail overcrowding is a factor in this, see above. Also, death row inmates will still spend a long time time in prison so your proposal isn't much of a solution. general overpopulation -- Wait, you want to use the death penalty as population control? That's one of the worst ideas I've heard in a long time. Let's try free condoms for everyone and a few other simple things before we go all Logan's Run.
Slashdot Mirror
OTOH, Instant Runoff is simple enough that you don't need a computer to tally the votes.
... + 2 or 54 pieces of information -- less than Condorcet.
I've always imagined what a nightmare it would be to audit an instant runoff election with a large number of candidates. In Condorcet, you simply have to tabulate the number of voters preferring X to Y vs. those preferring Y to X, for each pair of candidates X and Y. With 10 candidates, that's 180 pieces of information.
A 10-candidate IRV race can require elimination of up to 8 candidates before there is a guaranteed majority winner (or possibly a final 2-candidate tie if the number of ballots is even). The number of voters that prefer X to all remaining candidates shifts at each stage, so you have to record a set of scores for all remaining candidates at each stage. That's 10 + 9 + 8 +
But the kicker is this: In Condorcet, you create that 10x10x2 matrix (with 20 empty positions because you don't compare a candidate with himself) just once. It then contains all the information needed to complete the audit. Once you're satisfied that the matrix is correct, you don't have to look at the ballots again. In IRV, you have to have access to the full set of ballots at every stage. You count the ballots 8 times, each time ignoring the candidates eliminated by the previous counts. So you can't toss the ballots aside until you've completed the entire audit.
And forget about just aggregating the number of voters that vote a certain ranking. 10 factorial is quite a large number. You'd have to allocate a lot of memory (OK, in a computer program you'd use a hash table, but how much more efficient is that than simply having a record per ballot? My point is that Condorcet sidesteps these difficulties; you just read in a record, add the points to the matrix, and move on to the next record.)
California had an election not to long ago with 135 candidates. I'd hate to see that done in instant runoff.
Condorcet is not the simplest election method, but it is hands down the most accurate.
Well, I wouldn't call it most accurate. First, you've got to look at whether clones can affect the outcome. If some subset of the candidates have virtually indistinguishable positions, then adding or removing candidates in this subset can affect the outcome when there are nontransitive cycles such as in your example. They can also affect the way people vote, if people consider the possibility of such cycles (then they may try to game the cycle resolution method by reversing some of their preferences). On top of that, there are many, many ways to resolve cycles, and they will give different results in certain cases, so you would have to decide which resolution method is "most accurate" before you could start comparing Condorcet-class methods to other methods.
Then you've got to consider different categories of balloting. Basically, you've got ranked methods and rating methods. Condorcet, Borda, IRV, and many others are ranked methods. Plurality and approval are rating methods (though in those cases the ratings are binary, with plurality additionally constrained).
I tend to think that you can't compare methods for "accuracy" unless you can eliminate voter strategizing. Voters won't strategize if they have no clue how other voters are likely to vote. In those circumstances, I think approval is the most accurate rating method, and Borda is the most accurate ranked method. But those circumstances are rare to nonexistent in the real world.
But I like approval best (even over ranked methods), because any strategizing that occurs in approval tends to be benign.
Nah. Instant runoff is fine with three candidates. With more than three, you start having voters strategize their rankings. For example, if I like Kucinich > Edwards > Obama > Clinton (not my actual preference order, but it illustrates my case), I may nevertheless list Obama 2nd and Edwards 3rd in the primary because Obama has a better chance of beating the front runner Clinton. The more candidates you add, the more this kind of gaming of the vote becomes necessary.
Approval voting is much better. I could, for instance, approve Kucinich, Edwards, Obama, Dodd, and maybe Biden and Richardson, in order to maximize my insurance against Clinton.
Excuse me, but this is an open forum. If you don't like that, or if you can't take the criticism that goes with the territory, you don't have to post here. I haven't made any statements I'm not willing to defend, so I don't see why you have a problem with me or why you have to label my "style" as "hit-and-run". Where did I run to? I'm still here!
I've come to the conclusion that you're just plain rude, and trollish too, since you don't seem to be willing to back up your assertions.
First, I cite common sense. There is a difference between being a racist and acting out a racist agenda. The first doesn't (of itself) land you in jail.
Additionally I cite Rehnquist in Wisconsin v. Mitchell. If you don't care to read it all, here's the executive version: The high court did not find that the defendant's First Amendment rights were infringed by the legislation, and found the sort of chilling effect on expression that you envision too speculative for consideration in their opinion.
Surely, if the expression of racist thoughts isn't substantially chilled by legislation of this nature, then the thoughts themselves (minus the expression) are not chilled, either.
Of course, it's possible to envision legislation that is overly broad in this area. But I'm not advocating that kind of legislation, nor am I aware of any proposed or on the books.
Just as I expected, you've come up empty.
No, the difference there is not motivation but intent. Did you intend to kill the person? If you did intend to do so, did you do it on the spur of the moment or was it done in a planned out, cold-blooded fashion?
You've made a distinction without a difference. Both motivation and intent require a jury to make a determination of the defendant's state of mind. In fact juries are allowed to consider motivation in their deliberations; if prosecution can't establish a motive then it may cast reasonable doubt on the case (depending on the strength of other evidence, of course).
No, if he were just on trial for his actions then it would be a normal assault/robbery/arson/murder trial, not a hate crime trial.
It *is* a "normal assault/robbery/arson/murder trial", with the additional charges that the defendant is acting out a specific agenda. Both require a jury determination that the standard burden of proof is met. A jury can still convict on the former charge and acquit on the latter. If prosecution doesn't think they can meet the burden for the latter then they shouldn't bring that charge in the first place. If prosecution does bring the hate crime charges while having insufficient evidence then that is a separate issue, because we already have similar instances of overzealous prosecution with all manner of crimes. If you are looking for a perfect world where that doesn't happen, I can't help you.
Thank you for one of the clearest and most reasoned posts in this thread.
But in federal hate crime legislation, all you have to prove is a prior disposition towards a protected class.
Citation, please? Until you provide one, I'll maintain that GP is correct.
At the end of the day, what good does hate crime legislation do? All it means is that a racist might spend 20 years in jail for the same exact crime that a non-racist would only spend 15 years in for. So, when it comes down to that, he's spending 5 years in jail just for being a racist.
No, he's spending the extra 5 years for acting out a racist agenda.
And HOW is that not "thought crime?"
Because there is no law against his thoughts, and his thoughts aren't policed. I think you are using a very loose definition of "thought crime".
I think slashdot moderators hate me, why am I modded a 1 and the comment I'm replying to a 2??
I think that's called a karma bonus modifier; my post wasn't moderated either way.
We're both right. It just depends on what is meant by "delivery". Delivery is out of order on the wire, and that's what I (and I suspect the original poster) meant. Of course the protocol stack guarantees in-order delivery to the higher layers. That means that the slowest packet is going to dictate the latency of an entire transfer, which I think is what he was getting at.
That said, I don't think ordering is a big deal for FCoE. For FC, in-order delivery (on the wire, that is) isn't necessarily required. Class-2 and Class-3 FC do not guarantee it. So at least these two classes should map to FCoE easily enough, and I'm sure a TCP-like reordering capability in the FCoE encapsulation could enable mapping of Classes 1, 4, and 5.
You've got some serious issues if you have packets that are out of order on your local network.
I don't think that would necessarily indicate *serious* issues. A local switched network for a mid-sized company or college campus could certainly be complex enough that a little temporary congestion could affect arrival order.
Disclaimer: I'm neither an FC nor Ethernet expert, but I've had just enough exposure to both to be dangerous.
It doesn't *deliver* packets in order (at least, not unless the underlying network does). It provides the capability to reconstruct the original order. GP was talking about *delivery* of packets.
Single-platform? OpenFCoE may be a Linux software initiative, but I think the T11 FCoE standard on which it is based is being developed for all modern major server platforms.
You know, I agree with your post except for one thing. As a Southern California resident, I hate to see any act of arson described as trivial.
In a legal sense, a "fact" is something to be determined by a jury. A jury may look at evidence for and against the prosecution's theory of what happened (including the motivation) and decide that it fits the facts of the case. If they can't find reasonable doubt, they accept a proposition as factual in producing their verdict. This is the power given to juries.
IANAL, but come on, we've all watched enough TV to know this stuff...
So in your mind, juries should never take into account a defendant's state of mind? That would wipe out many of the distinctions commonly made today between first and second degree murder and manslaughter, as well as consideration of mitigating and aggravating circumstances in violent crimes in general.
Contrary to the popular notion, hate crimes are not thought crimes. The defendant in such crimes is always on trial for his actions, not his thoughts. Juries are allowed to make factual findings in their deliberations, subject to the usual burden of proof ("beyond reasonable doubt"), and state of mind is a factual finding of this nature. But I know of no case in the U.S. where someone has been brought to trial just for harboring hateful thoughts.
You're right, of course, they are separate issues. However, Blockbuster's business model doesn't have any significant advantage (for me) over Netflix's. Between the two busines models, I'm neutral as a customer. Given my prior reason not to use Blockbuster, I rent from Netflix.
As an investor, I might also find that the lower overhead of the Netflix model is an advantage.
I object to government censorship of movies and ideas, as well as to the crowd-hysteria brand of censorship. If the government were to prevent a studio from releasing a film, I would object just as strongly. However, the citations you gave seem to tell of one individual who put public pressure on a studio to bow out. And the individual happened to be a state assemblyman. Is this an instance of one individual exercising his free speech rights, or is this government censorship? Maybe it's the latter, but somehow I doubt that the state of New York really has that kind of power.
More accurately, this would be the power of the bully pulpit, which of course falls under my original objection. I can't tell from those two blurbs whether the assemblyman's comments were a factor in Fox's decision, or whether they made the decision for internal reasons of their own. If it's the former, then we can say Fox caved just like Blockbuster, which would be a strike against Fox (and not necessarily against all of Hollywood as you originally posted).
I don't know if they "banded together" or not. I won't ask for a citation because such banding is not relevant; if they did then they were all presumably willing participants. Under that presumption none of the studios were censored by this "band". Ask anyone -- or any group/band/cartel or whatever noun you care to apply -- to publish something; they are free to say yes or no. And Blockbuster is free to choose the movies they carry, I'll grant them that. But someone else -- not Blockbuster -- decided that Blockbuster shouldn't carry a certain movie, and Blockbuster caved. That is enough of a flaw in Blockbuster's moral fiber to motivate me to rent elsewhere.
These two cases aren't the same. One case is self-censorship. I'm not even sure that such a broad use of the term "censorship" is a good thing. Maybe "self-representation" is more accurate. I fully support anyone's right to publish or not publish submitted materials, as they see fit. Call it an agenda if you like, but it doesn't matter: it's a basic freedom. The other case is caving to blackmail. It only opens the door to further blackmail and reduction of freedom.
I don't have a problem with Company X choosing not to produce Movie Y because they don't see commercial potential. Studios have to make those decisions all the time (and bad business decisions are inevitable). After all, they have to put up a lot of resources to get a movie made. But if Movie Y does get made, and Company Z, that is in the mainstream movie rental business, refuses to make that movie available to its customers, only to appease some rabid bunch of loonies, then I have a problem with Company Z. I can also understand it if Movie Y is in some niche market that Company Z doesn't play in. It's the caving in to special interest threats like these that I don't like (this holds, whether or not I had any interest in seeing Movie Y).
That will open the door, and Home Depot will sue "Reaper".
I never liked Blockbuster because of their past history of censorship (see "Last Temptation of Christ"). Maybe they don't do that anymore but there's still a bad vibe for me there. Before Netflix, I always preferred Hollywood Video. And with Netflix, I've never had a problem getting the movies I want. Why take the trouble with going to the video store -- one or two movies delivered to my mailbox every week is as much as I ever find time to watch these days, anyway.
Ah, but the irony would be lost. We could name it after some innocent victim like Laci Peterson and it would pass for sure. But the same law named after a victim who also happened to be guilty of a sex crime (even a less serious one) would almost certainly fail, even though it covers exactly the same people.
I hope that post is in the spirit of a "modest proposal". Let's have a look:
jail overcrowding -- Decriminalize recreational drug use; revoke three-strikes laws.
inmate abuse -- Since jail overcrowding is a factor in this, see above. Also, death row inmates will still spend a long time time in prison so your proposal isn't much of a solution.
general overpopulation -- Wait, you want to use the death penalty as population control? That's one of the worst ideas I've heard in a long time. Let's try free condoms for everyone and a few other simple things before we go all Logan's Run.