I think you have valid concerns, but you are also mixing up severals meanings of "Linux is better, you should try it".
First, it is (more or less) generally admitted that Linux is technically better than Windows. For parts, this is something you can experience as a "general public" user: less system crashes, no need to reboot all the time,... But it's also many things you might not experience yourself, like ease of administration, in particular with a large number of machines,...
Now, the problems you mention rather fall into the daily use pattern and hardware support. That it did not work as easily as you expected is a valid reproach. I can assure you there are many people out there working on improving this, and if you have been following Linux you know it has already come a long way. I also think there is a way to make everything work, but yes, it might take some effort, that should be improved. But if you are honest, you should recognize that part of the reason for this is that the companies you bought hardware and services from "do not support Linux". Why? Because the market is too small? Because they have eclusivity deals with Microsoft? In any case, that looks a lot like a vicious circle.
Which brings to the third aspect. If Linux is "better", it is also from a freedom/philosophical point of view. It gives you choice. It allows you to avoid being locked in a single system (windows) , which is very valuable if you understand the risks of such a situation. This is what makes it worthwile to support Linux.
Does it mean you should use Linux? It really depends on your situation, and for you it looks like you might indeed want a bit more for a more user-friendly version/distribution (assuming you did try a reasonable distribution). Or get help from somebody to get those things working. In any case, just don't dismiss the "philosophical" aspect of "better".
I think it is quite obvious that if he or she keeps the copyright under his or her name and release the code under a specific GPL version (say, v.2 instead of "v.2 or any later version") s/he will always be in control.
Let me recall the fears of the OP:
Maybe eventually FSF will prevent me from using my own code in commercial products or something.
This fear is unfounded as long as the code is release under at least GPL v2, so advising to release under a specific GPL version does not seem appropriate.
The only "risk" I see in having "or any later version" is that the later versions will be more permissive. The risk of not having "or any later version" is that if/when the copyright holder goes AWOL/dies and GPL v2 becomes unacceptable for some reason (new laws, incompatible with some other license) nothing can be done about it. I suppose this is especially problematic when there are so many contributors that contacting all of them becomes virtually impossible.
Your advice does not match your analysis. Your are stating that the problem is that FSF asks for assignments. Then you advise to keep the copyright under his name and to license under a specific GPL version. You gave no motivation for the latter.
I will never put an "... or later" clause. Maybe eventually FSF will prevent me from using my own code in commercial products or something.
I think you have two misunderstandings:
As the copyright holder, you can always decide to release your code under another license.
The "or later" clause is at your option, "you" being the licensee. This means that that clause can only grant new rights, not remove rights, since anybody can always decide to chose to see the software as licensed under GPL v2. This is similar to dual licensing.
The fact that is does clustering proves that if was designed for it? Man, you either need to look up the word "design" in the dictionary or to take a course in logics.
Correct. Although ideally, if it tells you the variable may be null, then there exists some situation where the variable *will* be null. So you better handle this case.
Clearly, the problem of reporting an issue exactly in the cases where it will happen is undecidable, so a safe compiler has to be a bit restrictive. In practice, it seems that most cases where it reports a spurious problem is when it depends on the way clients call your code. And in that case, even if it cannot happen now, it could later happen because of a client change, so it's a good idea to handle that case now anyway.
Did you check the links? I'm actually pointing to a specific language where null pointer (OK, references) errors at caught at compile time. It *is* possible.
And if sun did something really awful to the language, which make no mistake they will do sooner or later simply because they're human, it would be good to have an alternative.
Great. But the whole point of election reform is to make it possible to have more choice, and I would guess you need at least 5-8 to cover a broad range of ideas. So why not chose a system that works well in all cases, not just for 3?
Great idea. Which makes me think it would be useful to have a place on the web to put presentation material, to make the task easier. Does such a place exist already?
My gut feeling is that ties would indeed be rare, but probably less so than with plurality. Is there any data or model to assess approximately how rare?
You've also got a very good point that Condorcet at least has a method to solve ties, while other methods simply hide the problem under the carpet. Ties are not a problem of Condorcet, it's a fact in the voters' intentions that has to be dealt with (should it ever occur).
This does not mean that it would be easy to sell Condorcet. The media is not always rational... Hence, it might useful to support Approval now, while still supporting Condorcet for the long run. What are the ways to do so? I know of Citizens for approval voting, although I'm not sure how active they are. Any other proposal?
If I'm not mistaken, the Avy method of IRV passes the Monotonicity Criterion.
From what looks like avy's author's webpage:
A simple way to measure that unfairness of a method, is to show the smallest example exhibiting unfairness or non-monotonicty, when the counts of ballots are integers. For the Alternative Vote, the smallest number is -9 when 3 candidates. [I have not computed the number for the Avy example.
This seems to suggest that avy is not monotonic.
http://electionmethods.org/evaluation.htm says both IRV and Condorcet fail Participation Criterion. As to whether the Avy method fails PC, I don't know.
Right, Condorcet fails PC. This illustrates that no method is perfect (which was proved, for a certain set of criteria).
Still, it seems that this should be a killer requirement: "If one candidate is preferred over each of the other candidates, then he should win". It actually sounds like a self-evident proposition. Well, only Condorcet satisfies this requirement.
The winning candidate is the one who can beat every other in a head-to-head race. Just rank your choices 1, 2, 3, etc. Your preferences indicate who wins in a head-to-head in your opinion. Tally those 'wins' for everyone's ballot."
Right, but you know you are ignoring the case of ties. Based on every serious comparison I read, it seems clear that Condorcet is the most fair system, with Approval still a significant improvement over plurality, while IRV is not. The problem is that when election reform comes under the spotlight, it would be easy to run stories describing tie-breaking algorithms, and scare people about the mess if a tie occurs and the lack of confidence in the result.
I wish there would be a more united election reform front, with more energy spent spreading the word than fighting each other. The choice between Approval and Condorcet is simply a compromise between simplicity and fairness.
Note that votergate's server has been attacked (looks like some people don't want this message to be heard) so they are only diffusing medium quality video at the moment, but you can get a high quality one on the edonkey network:
Yes, Diebold's marketing speach is just ridiculous when confronted to these facts. It's very striking if you watch the votergate documentary.
Note that their server has been attacked (looks like some people don't want this message to be heard) so they are only diffusing medium quality video at the moment, but you can get a high quality one on the edonkey network:
homosexuality is unarguably regarded as badly wrong
The problem is, many christians think this is not unarguable. In particular, it seems that the idea of homosexuality itself as a sexual orientation was not common at the time those passages were written, so it seems perfectly arguable to understand them as condemning some practices (homosexual rape, pedophily, religious orgies,...) but not homosexuality in a loving relationship. A few translations seem to be very misleading in this respect, which is why it's interesting to compare them and look at the original text. That's what the religioustolerance.org articles are trying to do. In this context, I cannot agree that there is an unarguable interpretation.
Concerning the strawman, it surely is a risk in this kind of debate. It struck me that the liberal view proposed by the grand-parent is actually proposed by somebody who is not a liberal himself, so that might as well be a strawman. Maybe it helps to simply avoid this kind of labels, which are indeed easily misused.
1. in 2Tim 3:16 Paul couldn't have been talking about the New Testament, as it didn't exist yet, only the Old Testament. Yet it is hard to find Christians today who would feel they have to keep all the regulations of the OT. Do they wear clothes woven of two separate materials? Do women keep themselves secluded during their period? Do we not eat blood sausages or bloody steaks? Does this make the Bible (or the OT) any less inspired?
2. My pastor told me that it is important to find out the context of the specific passages. For example, in his letter to the Corinthians, Paul says that women shoudln't have their hair down in church. Is this an absolute (yet mysterious) law of God? or, as the pastor suggested, was it a specific piece of advice to those women in those times, where the only women who wore their hair down in public were prostitutes?
3. This supports the idea that verses are not simply either true or false, but they have to be interpreted: what do they mean today?
4. To me this does not make the Bible useless as you suggest. It just means that you need to take the time to search what it can teach us. So it's not about perfectly knowing what God wants but thinking one knows better. It's about honestly seeking the truth, through the study of the bible of a personal relationship with God. Actually, this is similar to your idea that you should cross-check your interpretation with other passages. However, those other passages also have to be understood correctly. The ultimate check is what Jesus himself taught as the greatest commandement: love God, and your neightbour as yourself.
What the Bible says about any subject always depends on the way you read and interpret it. Just think about how differently Jesus lived and interpreted the scriptures compared to the Pharisees.
Slashdot Mirror
I think you have valid concerns, but you are also mixing up severals meanings of "Linux is better, you should try it".
... But it's also many things you might not experience yourself, like ease of administration, in particular with a large number of machines, ...
First, it is (more or less) generally admitted that Linux is technically better than Windows. For parts, this is something you can experience as a "general public" user: less system crashes, no need to reboot all the time,
Now, the problems you mention rather fall into the daily use pattern and hardware support. That it did not work as easily as you expected is a valid reproach. I can assure you there are many people out there working on improving this, and if you have been following Linux you know it has already come a long way. I also think there is a way to make everything work, but yes, it might take some effort, that should be improved. But if you are honest, you should recognize that part of the reason for this is that the companies you bought hardware and services from "do not support Linux". Why? Because the market is too small? Because they have eclusivity deals with Microsoft? In any case, that looks a lot like a vicious circle.
Which brings to the third aspect. If Linux is "better", it is also from a freedom/philosophical point of view. It gives you choice. It allows you to avoid being locked in a single system (windows) , which is very valuable if you understand the risks of such a situation. This is what makes it worthwile to support Linux.
Does it mean you should use Linux? It really depends on your situation, and for you it looks like you might indeed want a bit more for a more user-friendly version/distribution (assuming you did try a reasonable distribution). Or get help from somebody to get those things working. In any case, just don't dismiss the "philosophical" aspect of "better".
The idea of 20% is 1 day per week (1/5). Not sure if one specific day is recommended (friday?).
Your advice does not match your analysis. Your are stating that the problem is that FSF asks for assignments. Then you advise to keep the copyright under his name and to license under a specific GPL version. You gave no motivation for the latter.
- As the copyright holder, you can always decide to release your code under another license.
- The "or later" clause is at your option, "you" being the licensee. This means that that clause can only grant new rights, not remove rights, since anybody can always decide to chose to see the software as licensed under GPL v2. This is similar to dual licensing.
Both points mean that your fears are not founded.Which OS we are talking about is irrelevant. I'm saying that argument of the great-grandparent (of this post) is logically flawed.
The fact that is does clustering proves that if was designed for it? Man, you either need to look up the word "design" in the dictionary or to take a course in logics.
Correct. Although ideally, if it tells you the variable may be null, then there exists some situation where the variable *will* be null. So you better handle this case. Clearly, the problem of reporting an issue exactly in the cases where it will happen is undecidable, so a safe compiler has to be a bit restrictive. In practice, it seems that most cases where it reports a spurious problem is when it depends on the way clients call your code. And in that case, even if it cannot happen now, it could later happen because of a client change, so it's a good idea to handle that case now anyway.
Did you check the links? I'm actually pointing to a specific language where null pointer (OK, references) errors at caught at compile time. It *is* possible.
What about using a language where null pointer errors are caught at compile time? Oh, I guess our compiler must have a brain the size of a whole solar system to be able to do that.
You already have the choice.
Great. But the whole point of election reform is to make it possible to have more choice, and I would guess you need at least 5-8 to cover a broad range of ideas. So why not chose a system that works well in all cases, not just for 3?
Great idea. Which makes me think it would be useful to have a place on the web to put presentation material, to make the task easier. Does such a place exist already?
You've also got a very good point that Condorcet at least has a method to solve ties, while other methods simply hide the problem under the carpet. Ties are not a problem of Condorcet, it's a fact in the voters' intentions that has to be dealt with (should it ever occur).
This does not mean that it would be easy to sell Condorcet. The media is not always rational... Hence, it might useful to support Approval now, while still supporting Condorcet for the long run. What are the ways to do so? I know of Citizens for approval voting, although I'm not sure how active they are. Any other proposal?
From what looks like avy's author's webpage:
This seems to suggest that avy is not monotonic.http://electionmethods.org/evaluation.htm says both IRV and Condorcet fail Participation Criterion. As to whether the Avy method fails PC, I don't know.
Right, Condorcet fails PC. This illustrates that no method is perfect (which was proved, for a certain set of criteria).
Still, it seems that this should be a killer requirement: "If one candidate is preferred over each of the other candidates, then he should win". It actually sounds like a self-evident proposition. Well, only Condorcet satisfies this requirement.
I wish there would be a more united election reform front, with more energy spent spreading the word than fighting each other. The choice between Approval and Condorcet is simply a compromise between simplicity and fairness.
How is this idea formalized? Can anybody comment if this is indeed a problem with Condorcet?
On the other hand, there seems to be loads of problems with IRV. In particular, with IRV:
-
voting a candidate higher can cause the candidate to lose, and voting a candidate lower can cause the candidate to win.
-
adding one or more ballots that vote X over Y can change the winner from X to Y.
Based on these facts, why do you think IRV is a good method?ed2k://|file|votergate.avi|253406440|FCB6A52253FD5 0D6BB9F4E5823F755E2|
Note that their server has been attacked (looks like some people don't want this message to be heard) so they are only diffusing medium quality video at the moment, but you can get a high quality one on the edonkey network:
ed2k://|file|votergate.avi|253406440|FCB6A52253FD5 0D6BB9F4E5823F755E2|
Concerning the strawman, it surely is a risk in this kind of debate. It struck me that the liberal view proposed by the grand-parent is actually proposed by somebody who is not a liberal himself, so that might as well be a strawman. Maybe it helps to simply avoid this kind of labels, which are indeed easily misused.
1. in 2Tim 3:16 Paul couldn't have been talking about the New Testament, as it didn't exist yet, only the Old Testament. Yet it is hard to find Christians today who would feel they have to keep all the regulations of the OT. Do they wear clothes woven of two separate materials? Do women keep themselves secluded during their period? Do we not eat blood sausages or bloody steaks? Does this make the Bible (or the OT) any less inspired?
2. My pastor told me that it is important to find out the context of the specific passages. For example, in his letter to the Corinthians, Paul says that women shoudln't have their hair down in church. Is this an absolute (yet mysterious) law of God? or, as the pastor suggested, was it a specific piece of advice to those women in those times, where the only women who wore their hair down in public were prostitutes?
3. This supports the idea that verses are not simply either true or false, but they have to be interpreted: what do they mean today?
4. To me this does not make the Bible useless as you suggest. It just means that you need to take the time to search what it can teach us. So it's not about perfectly knowing what God wants but thinking one knows better. It's about honestly seeking the truth, through the study of the bible of a personal relationship with God. Actually, this is similar to your idea that you should cross-check your interpretation with other passages. However, those other passages also have to be understood correctly. The ultimate check is what Jesus himself taught as the greatest commandement: love God, and your neightbour as yourself.
What the Bible says about homosexuality on religioustolerance.org analyses the various texts and tries to show the different points of views.
OK, that would be a good point for it, but that would still make it objectively inferior to both condorcet and approval.