Actually, the big language culprits would be those with auto-garbage collection, etc. as they tend to have lazier programmers that don't "need" to manage their own resources,
People keep saying this. I wonder if it's true -- I haven't seen any evidence of it. I've seen lazy/stupid programmers in all kinds of languages. Wouldn't the same argument work just as well against memory protection, and preemptive multitasking, and compilers?
No, no, and yes (sort of).
Memory protection and preemptive multitasking are entirely different animals. Their job is not to help the programmer program the application to a working state. Their job is to protect the system from a bad (or hostile) program.
Compilers do count to the degree that the writing programmer doesn't understand what the compiler is doing. One of the greatest disservices I have seen in college is the glossing over of how c++ code is translated into machine code. My first brush with what is actually going on was given to me by an upper-classman TA! The teachers themselves never have sufficiently covered the topic. The principle applies to other languages too, of course.
Excuse my ignorance, but isn't there a huge difference between {removing a frame from the stack && explicit object destruction} and {garbage collection}?
Garbage collection is usually defined as automatic object destruction when there are no more pointers to it within the application. C++ doesn't do that. It can't, without taking a performance hit and taking control away from the programmer. This is not to say that garbage collection is evil (I use python regularly), only that I seriously doubt your claim to a garbage collector in gcc compiled programs. If there is one, it is probably a different definition than normal for garbage collection.
Perhaps men should just quit being such ASSHOLES towards women?
I wonder in anyone ever considered that concept.
I've concidered it. In fact, I don't remember ever being rude (etc.) towards women specifically. I think most guys I know haven't either (since junior high, at least).
(although there is a fair percentage who haven't been notified yet)
Whatever you experience that you think is discrimination, most of it ISN'T. And yes, some of it is. Most of the time, just pick yourself up and realize that there are idiots in the world, and you can't fix them.
Everyone who is genuinely discriminated against has equal right to do something about it. This is true to the degree of discrimination. To suggest otherwise is both stupid, and tears at the fabric of society.
To borrow from your example: Two men are trying to get into different colleges. One is white trying to get into a black college, and the other is black trying to get into a white college. Both are denied, and both have EQUAL right to act/complain. Why equal? Because the white college is guilty of NOTHING other than denying a black student (_assuming_ it is racism). The college is NOT guilty of any other discrimination that black student has suffered. If you do not hold to this truth, then you are guilty of misplacing blame. Nor is the white student guilty of anything.
I don't know about their new model. The flash version that they have had is actually pretty good. It gives you one commercial about six times throughout the show (instead of dozens). If you don't like the commercial, you can refresh the page and see a different one (oh what a breath of fresh air). The comercials are only aired at the beginning of a segment. You can fast-forward, rewind, pause, whatever within a segment. You can skip between segments (like on a DVD), but need to rewatch the comercial again.
You will find information on the basic stats, but nothing that describes how to generate the ability scores themselves. In aggregate, it comes close. The SRD is not entirely sufficient for character creation (although it comes very close).
WoTC has provided an open source license that they call "open game license" (for, get this "open game content"). They have NOT provided the core books under this license.
What they have done however (and I applaud) is provide a vast majority of the core rules under the OGL. This is referred to as the SRD (System Reference Document). The SRD does not contain everything that the books do. In general, the books are more verbose, but the SRD relates the rules just as well. The SRD rarely says "why" (as the books occasionally do). I believe there are a rare few few key rules not covered in the SRD.
The best place to read the SRD? In my opinion it is at www.d20srd.org (no connection to myself whatsoever). Everything is HYPERLINKED! The site is amazing in terms of quick access to everything. It even includes spell and monster filters to help navigate those areas. This can make building adventures much easier at times. (Example: the SRD includes 5 Undead monsters between CR 5 and 15 who have a lawful alignment; beats flipping through the book and finding one or two.)
This is a common misconception. You CAN sell GPL code. The GPL specifically allows it. The only thing you must do is provide the source code (in one way or another) and the only thing you can't do is include additional restrictions.
As the other posters have pointed out, the copyright owner has the right to dual license the software (offer the software under more than one license). The GPL is not law, but an open ended contract. It is permission to copy the code if certain conditions are met. This doesn't mean that the copyright owner can give permission under other conditions as well.
On the other hand, who is the copyright owner of patches and bug fixes submitted to them? IANAL, but it would seem to me that this could be a legal problem. For example, Sun requires anybody working on/patching Open Office to have a record on file delegating copyright of patches back to Sun. If MySQL AB doesn't do something like this, how do they plan to fend off lawsuits? (frivolous or legitimate)
Sorry, I can't help but point out that for, while, do, case, break, etc. all use a jump instruction. In essence, they all rely on goto. What makes these special is that they are designed to make it hard to write nonsense code. In that respect, label breaks are perfectly legitimate control flow elements NOT equivalent to goto (as they are controlled and specific meaning).
I'm not real strong on the math needed to evaluate this. I wish I was, and probably will be at some point in the future.
I still don't believe that the "Condorcet Winner" can be gamed (again, note that this is distinct from just someone who wins an election that uses a Condorcet Method.) The Condorcet Winner is defined as one candidate that would beat each other candidate in a one-on-one election (there may be no clear Condorcet winner in a given election).
Which Condorcet methods fail the independence axiom? How so? I will admit that some completion methods must, but is it somehow inherent in Condorcet (as you seem to suggest). I doubt it is generic to Condorcet, but I am willing to listen to a well formulated layman's proof.
I didn't know that. Although I did know that IRV has problems, I still support it. (although I do prefer a Condorcet Method, which eliminates most of the problems that IRV has)
With Survivor, its the candidates themselves that are voting. Furthermore, people change tactics constantly between rounds. IRV is, well, "Instant" Runoff Voting. It's not "slow, candidates only runoff voting".
The AC is right, that site on range voting is inaccurate on many levels. At the very least, it will break down into preferential voting. The IRV example is contrived (if not by percentages, then by candidates chosen). In that example, Gore is the clear Condorcet winner. Please look up Condorcet methods (which are like IRV in some directions, but much better).
I might be reaching, but I think you're wrong. This is not to say Arrow's Theorem is wrong, only that I think you are misapplying it.
Arrows Theorem discuses voting systems that take preferential ballots in an attempt to create an ordered list of candidates. These systems are useful when choosing a city council, for example. It is very rare in our politics (that I've seen) to have an election where more than one candidate is elected at a time. Even times when we could do so (such as electing house representatives and senators), we typically break up into districts.
Within the context of electing one person per election, it IS possible to be fair (even to a limited degree). The Condorcet Criterion is clearly fair, any way you examine it (at least for single office elections). I will never be entirely comfortable with an election system that does not choose the winner out of the Smith set (as per the Condorcet matrix).
If you wish to argue Arrows Theorem outside of that, fine.
While I disagree with dynamo, his post is an honest one that deserves some respect. It is representative of someone who cares politically and has a valid point.
It is not flamebait. Would someone please mod him (+1 underrated) to get him back up to '1' (I'm not suggesting higher).
The two party system is inherent in the plurality voting system that we employ. Nevertheless, I believe that both parties enjoy the false dichotomy of "Republicans v. Democrats". It's easier to get elected when you have only one other serious contender.
Oh, and if the Electoral College where to work correctly (and it wont), you could put IRV or Condorcet vote between them. Today, that would of course be a farce. It might not be if we could get the EC reps out of the pockets of the political parties (I'd still far prefer a popular-Smith/Condorcet vote).
I don't see "all things" as being that naive (no offense intended). By this definition, you could select the median candidate (round up) as the winner. That would make no sense, but would be "equal" in your framework.
The purpose of voting is to: (1) select between candidates for an office (2) select based on voters votes (3) select the candidate that the larges number of voters can agree should have the job.
It is on point 3 that Condorcet and Plurality are not equal. If you had one candidate who would win a one-on-one election with each other candidate, the Condorcet Criterion specifies that that candidate should win. This is clearly superior to plurality. Therefore, all things are NOT equal.
If only this was proven (and not very, very likely speculation). I would love to see Diebold thrown entirely out of the voting system market. They have certainly treated our votes with careless contempt, as evidenced by the voting systems that malfunctioned because Diebold loaded uncertified code onto them.
Again, not all Condorcet methods use beat-paths. The Condorcet paradox is not only inherent to Condorcet methods, it's inherent to voting itself. In other methods this isn't always obvious because they aren't sufficiently fair. More of them would be if they identifed the Condorcet winner (who should ALWAYS win any election; note that this is DIFFERENT than just someone who merely wins an election by a Condorcet Method).
Occam's Razor is a good rule of thumb, but one should never try to live by it exclusively. People who do have a serious problem (same with Murphy's Law). There are exceptions, and this is clearly one of them.
To further push the point "All things being equal, the simplest solution tends to be the best one." Condorcet (and even IRV) is clearly not equal to plurality. Therefore, Occam's Razor (as often quoted) does not even apply.
You're right, but it might not be obvious from your explanation as to why Condorcet is better in this case. (by the way, I like IRV, but agree that Condorcet is better.)
If: you have 39 ballots as: a b c you have 20 ballots as: b a c you have 41 ballots as: c b a
(this is the parent post example)
The second round of IRV would be : 59 for a 41 for c
While with a Condorcet method, b would win. Why (and this is important)? In a one-on-one election between b and either a or c, b would win. b is therefore the Condorcet winner (by definition). It's easy to see why b should be elected: he's preferred over both a AND c!
Condorcet matrix: (made ugly to get past lameness filter; looks a little cleaner if copied into a text editor.)
a b c a -- 39 59 b 61 -- 59 c 41 41 --
You've massively oversimplified the situation. The post you're responding to is much closer to accurate. There are very few ways for a third party candidate to win in our current plurality system. One, if we constantly hear from others (read: media) that they have a chance (Liberman in the last election). Two, if the vast majority of people don't care which candidate gets elected and throw their votes away together (unbelievable, but hypothetical). Three, I don't know. I think there is no three.
As long as the majority both care who is elected, and don't think a third party has a chance, then the third party candidate has no chance. People will always vote against the party they dislike most by voting for the party they dislike least.
(shameless plug: that's why we need a different voting method in the US; examples: Instant Runoff (IRV), Condorcet)
The only reason I think you are voting Libertarian (it sounds like) is because you care more about the principle of the thing than about who actually wins. You sir, are a rare minority (for better or worse).
Slashdot Mirror
Nah. Remember, they planned on using reentry to cook their product. Efficient.
He was at least half joking (though I believe it).
The joke is, if Microsoft products cannot agree with themselves, then at least one of them IS broken.
fascinating. How in the world does it handle an array of pointers?
http://www.hpl.hp.com/personal/Hans_Boehm/gc/gcdescr.html
No, no, and yes (sort of).
Memory protection and preemptive multitasking are entirely different animals. Their job is not to help the programmer program the application to a working state. Their job is to protect the system from a bad (or hostile) program.
Compilers do count to the degree that the writing programmer doesn't understand what the compiler is doing. One of the greatest disservices I have seen in college is the glossing over of how c++ code is translated into machine code. My first brush with what is actually going on was given to me by an upper-classman TA! The teachers themselves never have sufficiently covered the topic. The principle applies to other languages too, of course.
Excuse my ignorance, but isn't there a huge difference between {removing a frame from the stack && explicit object destruction} and {garbage collection}?
Garbage collection is usually defined as automatic object destruction when there are no more pointers to it within the application. C++ doesn't do that. It can't, without taking a performance hit and taking control away from the programmer. This is not to say that garbage collection is evil (I use python regularly), only that I seriously doubt your claim to a garbage collector in gcc compiled programs. If there is one, it is probably a different definition than normal for garbage collection.
And don't confuse biological gender with social gender.
There ARE certain traits that show up in different proportions between men and women. The only question is what causes the difference.
I've concidered it. In fact, I don't remember ever being rude (etc.) towards women specifically. I think most guys I know haven't either (since junior high, at least).
(although there is a fair percentage who haven't been notified yet)
You clearly belong to a "minority group".
Whatever you experience that you think is discrimination, most of it ISN'T. And yes, some of it is. Most of the time, just pick yourself up and realize that there are idiots in the world, and you can't fix them.
Everyone who is genuinely discriminated against has equal right to do something about it. This is true to the degree of discrimination. To suggest otherwise is both stupid, and tears at the fabric of society.
To borrow from your example: Two men are trying to get into different colleges. One is white trying to get into a black college, and the other is black trying to get into a white college. Both are denied, and both have EQUAL right to act/complain. Why equal? Because the white college is guilty of NOTHING other than denying a black student (_assuming_ it is racism). The college is NOT guilty of any other discrimination that black student has suffered. If you do not hold to this truth, then you are guilty of misplacing blame. Nor is the white student guilty of anything.
Please give us some sanity!
I don't know about their new model. The flash version that they have had is actually pretty good. It gives you one commercial about six times throughout the show (instead of dozens). If you don't like the commercial, you can refresh the page and see a different one (oh what a breath of fresh air). The comercials are only aired at the beginning of a segment. You can fast-forward, rewind, pause, whatever within a segment. You can skip between segments (like on a DVD), but need to rewatch the comercial again.
Yeah, not great, but it's much better than TV.
Other arguments entirely aside, I dare you to point out where in the SRD it describes character creation (in sufficient detail).
Post a link.
You will find information on the basic stats, but nothing that describes how to generate the ability scores themselves. In aggregate, it comes close. The SRD is not entirely sufficient for character creation (although it comes very close).
Not quite, but close.
WoTC has provided an open source license that they call "open game license" (for, get this "open game content"). They have NOT provided the core books under this license.
What they have done however (and I applaud) is provide a vast majority of the core rules under the OGL. This is referred to as the SRD (System Reference Document). The SRD does not contain everything that the books do. In general, the books are more verbose, but the SRD relates the rules just as well. The SRD rarely says "why" (as the books occasionally do). I believe there are a rare few few key rules not covered in the SRD.
The best place to read the SRD? In my opinion it is at www.d20srd.org (no connection to myself whatsoever). Everything is HYPERLINKED! The site is amazing in terms of quick access to everything. It even includes spell and monster filters to help navigate those areas. This can make building adventures much easier at times. (Example: the SRD includes 5 Undead monsters between CR 5 and 15 who have a lawful alignment; beats flipping through the book and finding one or two.)
This is a common misconception. You CAN sell GPL code. The GPL specifically allows it. The only thing you must do is provide the source code (in one way or another) and the only thing you can't do is include additional restrictions.
As the other posters have pointed out, the copyright owner has the right to dual license the software (offer the software under more than one license). The GPL is not law, but an open ended contract. It is permission to copy the code if certain conditions are met. This doesn't mean that the copyright owner can give permission under other conditions as well.
On the other hand, who is the copyright owner of patches and bug fixes submitted to them? IANAL, but it would seem to me that this could be a legal problem. For example, Sun requires anybody working on/patching Open Office to have a record on file delegating copyright of patches back to Sun. If MySQL AB doesn't do something like this, how do they plan to fend off lawsuits? (frivolous or legitimate)
Sorry, I can't help but point out that for, while, do, case, break, etc. all use a jump instruction. In essence, they all rely on goto. What makes these special is that they are designed to make it hard to write nonsense code. In that respect, label breaks are perfectly legitimate control flow elements NOT equivalent to goto (as they are controlled and specific meaning).
I'm not real strong on the math needed to evaluate this. I wish I was, and probably will be at some point in the future.
I still don't believe that the "Condorcet Winner" can be gamed (again, note that this is distinct from just someone who wins an election that uses a Condorcet Method.) The Condorcet Winner is defined as one candidate that would beat each other candidate in a one-on-one election (there may be no clear Condorcet winner in a given election).
Which Condorcet methods fail the independence axiom? How so? I will admit that some completion methods must, but is it somehow inherent in Condorcet (as you seem to suggest). I doubt it is generic to Condorcet, but I am willing to listen to a well formulated layman's proof.
I didn't know that. Although I did know that IRV has problems, I still support it. (although I do prefer a Condorcet Method, which eliminates most of the problems that IRV has)
Good observation, but no.
With Survivor, its the candidates themselves that are voting. Furthermore, people change tactics constantly between rounds. IRV is, well, "Instant" Runoff Voting. It's not "slow, candidates only runoff voting".
Both of these skew the results horribly.
The AC is right, that site on range voting is inaccurate on many levels. At the very least, it will break down into preferential voting. The IRV example is contrived (if not by percentages, then by candidates chosen). In that example, Gore is the clear Condorcet winner. Please look up Condorcet methods (which are like IRV in some directions, but much better).
I might be reaching, but I think you're wrong. This is not to say Arrow's Theorem is wrong, only that I think you are misapplying it.
Arrows Theorem discuses voting systems that take preferential ballots in an attempt to create an ordered list of candidates. These systems are useful when choosing a city council, for example. It is very rare in our politics (that I've seen) to have an election where more than one candidate is elected at a time. Even times when we could do so (such as electing house representatives and senators), we typically break up into districts.
Within the context of electing one person per election, it IS possible to be fair (even to a limited degree). The Condorcet Criterion is clearly fair, any way you examine it (at least for single office elections). I will never be entirely comfortable with an election system that does not choose the winner out of the Smith set (as per the Condorcet matrix).
If you wish to argue Arrows Theorem outside of that, fine.
While I disagree with dynamo, his post is an honest one that deserves some respect. It is representative of someone who cares politically and has a valid point.
It is not flamebait. Would someone please mod him (+1 underrated) to get him back up to '1' (I'm not suggesting higher).
For my argument against him, see my own post.
The two party system is inherent in the plurality voting system that we employ. Nevertheless, I believe that both parties enjoy the false dichotomy of "Republicans v. Democrats". It's easier to get elected when you have only one other serious contender.
Oh, and if the Electoral College where to work correctly (and it wont), you could put IRV or Condorcet vote between them. Today, that would of course be a farce. It might not be if we could get the EC reps out of the pockets of the political parties (I'd still far prefer a popular-Smith/Condorcet vote).
I don't see "all things" as being that naive (no offense intended). By this definition, you could select the median candidate (round up) as the winner. That would make no sense, but would be "equal" in your framework.
The purpose of voting is to:
(1) select between candidates for an office
(2) select based on voters votes
(3) select the candidate that the larges number of voters can agree should have the job.
It is on point 3 that Condorcet and Plurality are not equal. If you had one candidate who would win a one-on-one election with each other candidate, the Condorcet Criterion specifies that that candidate should win. This is clearly superior to plurality. Therefore, all things are NOT equal.
If only this was proven (and not very, very likely speculation). I would love to see Diebold thrown entirely out of the voting system market. They have certainly treated our votes with careless contempt, as evidenced by the voting systems that malfunctioned because Diebold loaded uncertified code onto them.
Again, not all Condorcet methods use beat-paths. The Condorcet paradox is not only inherent to Condorcet methods, it's inherent to voting itself. In other methods this isn't always obvious because they aren't sufficiently fair. More of them would be if they identifed the Condorcet winner (who should ALWAYS win any election; note that this is DIFFERENT than just someone who merely wins an election by a Condorcet Method).
Occam's Razor is a good rule of thumb, but one should never try to live by it exclusively. People who do have a serious problem (same with Murphy's Law). There are exceptions, and this is clearly one of them.
To further push the point "All things being equal, the simplest solution tends to be the best one." Condorcet (and even IRV) is clearly not equal to plurality. Therefore, Occam's Razor (as often quoted) does not even apply.
You're right, but it might not be obvious from your explanation as to why Condorcet is better in this case. (by the way, I like IRV, but agree that Condorcet is better.)
If:
you have 39 ballots as: a b c
you have 20 ballots as: b a c
you have 41 ballots as: c b a
(this is the parent post example)
The second round of IRV would be :
59 for a
41 for c
While with a Condorcet method, b would win. Why (and this is important)? In a one-on-one election between b and either a or c, b would win. b is therefore the Condorcet winner (by definition). It's easy to see why b should be elected: he's preferred over both a AND c!
Condorcet matrix:
(made ugly to get past lameness filter; looks a little cleaner if copied into a text editor.)
a b c
a -- 39 59
b 61 -- 59
c 41 41 --
You've massively oversimplified the situation. The post you're responding to is much closer to accurate. There are very few ways for a third party candidate to win in our current plurality system. One, if we constantly hear from others (read: media) that they have a chance (Liberman in the last election). Two, if the vast majority of people don't care which candidate gets elected and throw their votes away together (unbelievable, but hypothetical). Three, I don't know. I think there is no three.
As long as the majority both care who is elected, and don't think a third party has a chance, then the third party candidate has no chance. People will always vote against the party they dislike most by voting for the party they dislike least.
(shameless plug: that's why we need a different voting method in the US; examples: Instant Runoff (IRV), Condorcet)
The only reason I think you are voting Libertarian (it sounds like) is because you care more about the principle of the thing than about who actually wins. You sir, are a rare minority (for better or worse).