he would still have to spend some time in prison with much more serious offenders
That's the law. Take it up with lawmakers. Don't blame the prosecutor when the public demands stupid laws (and some of these laws are stupid). In addition, Swartz knew the law and deliberately broke it.
Quick quiz: when he gets out he is viewed by potential employers as a) a brilliant young man, who just made some wrong decisions in the past, but it's all forgiven and forgotten; or b) a felon, found guilty of several computer-related crimes? Guess which viewpoint would be prevalent?
Doesn't seem to have hurt Robert Morris. He did roughly the same thing minus the copyright infringement, got convicted, and now is a professor at MIT.
Even in your adversarial system, there is neither a reason nor an expectation that the prosecutor should initially be making what were basically frivolous claims that go beyond all sane reason.
The reason the prosecutor can make those charges is because that's the law. I agree that the law is too strict, but you can't blame the prosecutor for that.
Furthermore, the prosecutor can't just go out and charge whoever he likes, he needs to convince a grand jury that the charges are reasonable. That means a majority of about 20 regular people have to agree that the person should get charged.
Sure enough a prosecutor will initially often enough be asking for more than is strictly warranted, but 35 years for using a script to download some files that were intentionally freely accessible within the uni network?
(1) Swartz was not a student at the university; he broke in and physically hacked into their network.
(2) The files were not "freely accessible"; they were available only under license, and Swartz repeatedly circumvented attempts to kick him off the network. Also, it was likely Swartz's intent to redistribute them.
(3) 35 years is the theoretical maximum when you total up all the charges with maximum penalties. He would likely have faced a few years in prison if found guilty of all charges, about the same as in many European countries.
We have an adversarial legal system: the prosecutor throws what he can get away with at the defendant, the defense attorney tries to defend against it, and the judge and the jury make a decision. If the prosecutor overreaches, the case will collapse very quickly.
(I can't quite tell: are you simply not familiar with how the US legal system works, or are you deliberately misrepresenting it?)
I once saw a (server-side) project of about 100k classes go from Java 1.3 32bit on Windows to Java 1.4 64bit on Solaris without even needing a recompilation. The stuff just worked.
That level of portability and cross-platform compatibility is nothing special; C/C++ is the rare exception because it exposes so many unportable features and is so completely lacking in runtime error checking.
but at least your most common data type doesn't change whenever you change the number of fucking bits in your CPU.
C99 has int32_t and similar types. If you want guaranteed sizes, use them. However, that makes your code less future-proof, since a lot of people expect that with a bigger CPU, they can also process more data. When used properly, C/C++'s CPU scaling of types is the right thing to do. What C/C++ get wrong is the lack of error checking by default; but, then, so does Java.
Java is a better language for the masses than C/C++. But that's kind of like saying that being hanged is better than being poisoned.
Old researchers (and both qualify as such) tend to have an excellent idea of what doesn't work, but they are often blind to the things that do work. Both these people have worked creating an AI for decades, neither has had a lack of resources, and neither has succeeded.
Replicating published experiments is a worthwhile effort, and there should be more of it. However, it's already pretty well known that scientific papers have a high rate of irreproducible results, due to both fraud and error. If they fail to replicate particular experiments, it's not an indictment of the scientific method, but of the particular scientists.
(Also, some experiments in the natural sciences are tricky to do and require experience to work, but they can still be replicated.)
Did I make a distinction between selling/stealing? Is that relevant to any argument I made? No. You're fixating on irrelevant minutiae to avoid the real question, namely providing any sound evidence that gun control has a demonstrably positive effect. Of course, you can't provide such evidence because it doesn't exist:
I conclude that every gun from which the serial number has been removed is almost certain evidence that gun was sold by its last legal owner to person prohibited from owning one, either in a straw purchase or an unrecorded or falsified sale from a gun shop or show.
Why are you even arguing that point? All guns had a "legal owner" at some point who then sold it to an "illegal owner". Isn't that bloody obvious? The question is whether the person committing the crime was in legal possession of the gun at the time they committed the crime, and as your own data shows, very few were.
I believe what you're arguing is that if we put in an intrusive tracking system, we can somehow reduce these transfers. Perhaps we could, but you'd have to take draconian measures. The reduction in gun violence is not going to be proportional to the reduction in gun availability because until you have achieved dramatic reductions, criminals will still be able to get them, and you are mostly reducing guns among harmless, law abiding, and responsible gun owners.
But the effect of such a tracing system is enormously intrusive; nearly half the US households would be registered and tracked in such a system, without ever having committed a criminal act. Furthermore, if you accept this principle, why not put into law complete tracing of prescription drugs and sexual contacts? Why not eliminate all cash transactions and make all financial transactions traceable by the government? Why not track, record, and send to the government all Internet connections? All of those can be argued to allow the government to reduce crime and death, so why not implement those as well? Or is that next?
You're starting with the wrong premise, namely the premise that it is the federal government's responsibility to make my life safer, and it can do whatever it wants to in order to achieve that goal. No, the federal government has a small set of enumerated powers. To the degree that any governmental entity has that duty or right, it's the state and local governments, but they are limited by the Constitution as well. "It makes you safer/healthier/whatever" by itself simply is not justification for federal law or federal intervention, even if it were unequivocally true (which, in the case of guns, it isn't).
I realized this was a tall order when I asked you and because I knew you didn't have data to back up your assertion. The reason is nobody collects really good statistics
Don't twist my words. Go to the Wikipedia page on "gun control", it tells you that people have looked at tons of data and never been able to find a strong effect. Some studies have shown small statistically significant effects, but not consistently, and the effects were generally so small that they don't justify gun control.
Seems to support what I was saying: if you add that up, about 90% of the guns used or possessed by state inmates were in their possession illegally. That's what I was referring to by "illegal gun", because a gun becomes an illegal gun when it is transferred from an owner who may possess it legally to someone who may not possess it legally.
If you think that you can choke the supply of illegal guns by choking the supply of legal guns, you're right, you can do that. It worked, for example, in the GDR. In fact, the GDR achieved very low overall crime rates. Of course, everybody's life was completely supervised by their government, they could not travel abroad, and any behavior or ideas that were thought to threaten the security of the state or society resulted in being sent to a psychiatric institution. That is exactly the direction we are heading, and there is a continuum between where we are now (more liberty, more crime, and more inequality than most other developed nations) to where nations like the GDR are (no liberty, nearly no crime, nearly no inequality). Where people like you and Obama are vague, naive, or dishonest (it is hard to tell which), is that you promise to deliver the benefits without the loss of liberty. You haven't experienced it, I have: believe me, it is not a good direction to go into.
That clearly isn't true. If you banned guns entirely it would quickly become harder to get one to commit suicide with, and the supply to criminals would be reduced as well since merely trying to manufacture or import a gun would be suspicious.
If you take very strong measures, yes, you could reduce suicide and homicide rates. I had relatives that lived in a country where that succeeded: the GDR. It is not an acceptable approach. Any actually feasible level of gun control, policing, and border control will not result in significant reductions.
3D printers complicate the issue even further, because in less than a decade, most people will be able to print working guns.
Yes, I deliberately used the ambiguous term "illegal gun" because addressing all those points in a single sentence would be pointless. If you really want to know, do some Google searches.
Furthermore you should go down to specifics: you want to pass laws to restrict our liberties, you need to show that these laws are in fact effective. Unless you can show clear evidence for a strong reduction in homicides through (more) gun control, there should be no additional gun control.
Let me add that I have never owned a gun, am not an NRA member, and never intend to own a gun. But the reasoning and swiftness with which Obama wants to take away liberties, as well as the lack of any sound scientific evidence, scares me, because the same approach can be taken to issues I do care about.
Furthermore, none of those studies take into account the tradeoffs. If you impose enough controls, you can indeed lower the crime rate; the GDR had a very low crime rate, but it wasn't a free society (I actually had relatives there). I do not want to trade liberties for safety.
And almost every one of those is a case where the gun is being used as the manufacturer intended, not an accident.
Yes. About 2/3 of those uses are suicides, and the rest are almost all homicides with illegal guns. Gun control has no significant effect reducing either of these numbers. There is a small remainder of homicides committed with legally owned guns and accidents, but many legal products are far more dangerous. Furthermore, there is no justification for creating intrusive government regulation that prevents me from committing suicide with a gun.
It's really not like the war on drugs at all. The war on drugs failed because many millions of folks in every country routinely take drugs and circumvent controls to get hold of them. That is patently not the case for gun restrictions, which vanishingly few people try to circumvent in the UK and other Western European countries.
How the hell do you know? Why do you keep fabricating these facts out of thin air?
You've assumed that the vast majority of gun homicides in the US are caused by criminals.
I'm not assuming anything. I'm observing that there is no data showing that gun control reduces homicides, I'm merely suggesting why that is a plausible result. And by definition, gun homicides are caused by criminals.
Neatly ignoring the many deaths caused by terrified home defenders, kids who get access to guns, etc.
Let's put these numbers into perspective. There are 16000 homicides in the US each year, 11000 gun related homicides and 19000 suicides with guns. Both criminals and suicidal persons will seek out and obtain alternatives to achieve their goal.
There are about 120 accidental deaths from firearms in children. These are due to parental negligence, and they are a miniscule fraction of overall preventable accidental deaths. Are you saying that all products that might cause accidental death in children should be controlled? Why?
You have presented no evidence that wrongful killing of home invaders by homeowners is a significant cause of death. Where is your data? And what's the alternative? Why should I be forced to face an armed invader into my home unarmed? And what makes you think that home owners won't find lethal alternatives?
You and people like you are arguing just like the Christian conservatives argue against abortion and gay marriage: you can't provide a shred of evidence that the restrictions you want to impose are effective, but you are ideologically opposed to it, so you keep making up hypotheticals and vague stories about how children are threatened.
And you have another thing in common with those Christian conservatives: given that guns are primarily used for suicide, much of what the gun control debate is really about is to deprive people of control over their own life and body, putting obstacles in their way for committing suicide.
Java is a progamming language, like C. It has access to the filesystem and can fork processes. Security is handled by the operating system, just like C.
That's false. Java claims to provide sandboxing, runtime safety, and fault isolation, something C never provided. In different words, security can be handled by the OS, but it can also be handled by the language and runtime. Providing it in the language and runtime has lots of potential advantages. Java tried to provide support in the language and runtime, but failed. Other languages and runtimes have succeeded at accomplishing what Java failed to do with sandboxing, runtime safety, and fault isolation, so it's certainly possible to do in principle.
Of course there's no correlation between countries in Western Europe.
You said that there was a "pattern", but there is no pattern in the data. And you refuse to put up any data showing your "pattern".
but the great majority of guns in Western European countries will be held by the police, the army, gamekeepers, criminals who've obtained them through smuggling and a few other groups.
Yes, and that's why gun control is ineffective: criminals can get them no matter how much you restrict legal gun ownership. It's the same reason the war on drugs has been a failure.
Sometimes less is more. I have a tablet hooked up to a monitor, but Android constantly gets confused about the two screens and their resolutions. That means you keep having to fiddle with the touch screen. A dedicated device like this always uses the connected monitor for its output, and the mouse and keyboard for its input. Also, this has better specs than low-end tablets; in different words, at the same price, you get better performance for not paying for a screen.
Well, the morale of Apple workers is clearly generally great. And their products are pretty good, but you also pay for it. But that's no different from most other manufacturers of expensive luxury products, and that's what Apple effectively is.
Maintaining an all-Linux backend usually requires a lot of high level administrative support and the administrators I can guarantee you are looking at COST first THEN functionality and they will ALWAYS see a Microsoft-based solution as inherently cheaper "because we already do that."
That's utter nonsense. In reality, Linux is no harder (and probably easier) to maintain than Windows: it's less complicated, its interfaces and tools change less over time, and there are numerous ready-made, turnkey server solutions that you basically just boot up and use. More CS grads will know and will have administered Linux systems than Windows systems because administration for Linux systems is much more similar between single user and multi-user.
School administrators may choose Windows because they believe it's easier to maintain, but they are mistaken.
When talking about large-scale websites the language is hardly relevent. There are as many high-traffic sites running on C#, Java, PHP or whatever.
When talking about small scale websites, language isn't relevant, because such sites usually operate far below server capacity, and small scale web sites make up the bulk of the web. But for large scale web sites, and in particular web sites that have to perform compute intensive operations for users (image processing, speech recognition, etc.), performance, and hence language, matters a great deal: a factor of two speedup cuts your hardware costs in half.
In fact, little performance critical code is written in Java, Ruby, or languages like that because their performance is lousy. As you note yourself, people convert performance critical parts to C or C++. But such mixed language systems are hard to maintain, and C/C++ programming itself is costly and risky. A language that combines runtime safety with near-C performance is an excellent compromise. There are several languages that offer that, including C# and D, but industry inertia keeps them from being adopted.
I had done all the graphs. There is no correlation for all nations. There is no correlation for all nations with homicide rates less than 0.6. There is no correlation for OECD countries. There is no correlation for OECD countries without outliers (US, Mexico). There is no correlation for Western Europe. There is no correlation for US states.
That first graph is probably one of the most stupid fucking graphs I've ever seen. Have you never heard of "keep all other factors constant"? There is no point whatsoever in including conflict countries like Afghanistan or drug-supplier countries like El Salvador on the same graph as Western Europe, the US, Canada etc
I had done all the graphs. There is no correlation for all nations. There is no correlation for all nations with homicide rates http://minus.com/lddhssi1nFFVu
Yes, there are too many confounding variables. There are more confounding variables than there are countries. That's why the hypothesis "higher gun ownership causes higher homicide rates" wouldn't be supported even if there was a correlation and even if you think you can justify a choice of a particular subset to look at. And there are far fewer confounding variables among US states, and there still is no correlation.
Sorry, but there simply is no statistical support for the hypothesis that higher gun ownership rates causes higher homicide rates, nor for the hypothesis that gun control lowers homicide rates. If you disagree, the burden is really on you to provide statistically significant evidence for the hypothesis.
Slashdot Mirror
That's the law. Take it up with lawmakers. Don't blame the prosecutor when the public demands stupid laws (and some of these laws are stupid). In addition, Swartz knew the law and deliberately broke it.
Doesn't seem to have hurt Robert Morris. He did roughly the same thing minus the copyright infringement, got convicted, and now is a professor at MIT.
The reason the prosecutor can make those charges is because that's the law. I agree that the law is too strict, but you can't blame the prosecutor for that.
Furthermore, the prosecutor can't just go out and charge whoever he likes, he needs to convince a grand jury that the charges are reasonable. That means a majority of about 20 regular people have to agree that the person should get charged.
(1) Swartz was not a student at the university; he broke in and physically hacked into their network.
(2) The files were not "freely accessible"; they were available only under license, and Swartz repeatedly circumvented attempts to kick him off the network. Also, it was likely Swartz's intent to redistribute them.
(3) 35 years is the theoretical maximum when you total up all the charges with maximum penalties. He would likely have faced a few years in prison if found guilty of all charges, about the same as in many European countries.
We have an adversarial legal system: the prosecutor throws what he can get away with at the defendant, the defense attorney tries to defend against it, and the judge and the jury make a decision. If the prosecutor overreaches, the case will collapse very quickly.
(I can't quite tell: are you simply not familiar with how the US legal system works, or are you deliberately misrepresenting it?)
That level of portability and cross-platform compatibility is nothing special; C/C++ is the rare exception because it exposes so many unportable features and is so completely lacking in runtime error checking.
C99 has int32_t and similar types. If you want guaranteed sizes, use them. However, that makes your code less future-proof, since a lot of people expect that with a bigger CPU, they can also process more data. When used properly, C/C++'s CPU scaling of types is the right thing to do. What C/C++ get wrong is the lack of error checking by default; but, then, so does Java.
Java is a better language for the masses than C/C++. But that's kind of like saying that being hanged is better than being poisoned.
Old researchers (and both qualify as such) tend to have an excellent idea of what doesn't work, but they are often blind to the things that do work. Both these people have worked creating an AI for decades, neither has had a lack of resources, and neither has succeeded.
Replicating published experiments is a worthwhile effort, and there should be more of it. However, it's already pretty well known that scientific papers have a high rate of irreproducible results, due to both fraud and error. If they fail to replicate particular experiments, it's not an indictment of the scientific method, but of the particular scientists.
(Also, some experiments in the natural sciences are tricky to do and require experience to work, but they can still be replicated.)
Did I make a distinction between selling/stealing? Is that relevant to any argument I made? No. You're fixating on irrelevant minutiae to avoid the real question, namely providing any sound evidence that gun control has a demonstrably positive effect. Of course, you can't provide such evidence because it doesn't exist:
http://en.wikipedia.org/wiki/Gun_control
Why are you even arguing that point? All guns had a "legal owner" at some point who then sold it to an "illegal owner". Isn't that bloody obvious? The question is whether the person committing the crime was in legal possession of the gun at the time they committed the crime, and as your own data shows, very few were.
I believe what you're arguing is that if we put in an intrusive tracking system, we can somehow reduce these transfers. Perhaps we could, but you'd have to take draconian measures. The reduction in gun violence is not going to be proportional to the reduction in gun availability because until you have achieved dramatic reductions, criminals will still be able to get them, and you are mostly reducing guns among harmless, law abiding, and responsible gun owners.
But the effect of such a tracing system is enormously intrusive; nearly half the US households would be registered and tracked in such a system, without ever having committed a criminal act. Furthermore, if you accept this principle, why not put into law complete tracing of prescription drugs and sexual contacts? Why not eliminate all cash transactions and make all financial transactions traceable by the government? Why not track, record, and send to the government all Internet connections? All of those can be argued to allow the government to reduce crime and death, so why not implement those as well? Or is that next?
You're starting with the wrong premise, namely the premise that it is the federal government's responsibility to make my life safer, and it can do whatever it wants to in order to achieve that goal. No, the federal government has a small set of enumerated powers. To the degree that any governmental entity has that duty or right, it's the state and local governments, but they are limited by the Constitution as well. "It makes you safer/healthier/whatever" by itself simply is not justification for federal law or federal intervention, even if it were unequivocally true (which, in the case of guns, it isn't).
Don't twist my words. Go to the Wikipedia page on "gun control", it tells you that people have looked at tons of data and never been able to find a strong effect. Some studies have shown small statistically significant effects, but not consistently, and the effects were generally so small that they don't justify gun control.
And how do those statistics relate to the inference that gun control works? Or that guns are in the legal possession of perpetrators?
Seems to support what I was saying: if you add that up, about 90% of the guns used or possessed by state inmates were in their possession illegally. That's what I was referring to by "illegal gun", because a gun becomes an illegal gun when it is transferred from an owner who may possess it legally to someone who may not possess it legally.
If you think that you can choke the supply of illegal guns by choking the supply of legal guns, you're right, you can do that. It worked, for example, in the GDR. In fact, the GDR achieved very low overall crime rates. Of course, everybody's life was completely supervised by their government, they could not travel abroad, and any behavior or ideas that were thought to threaten the security of the state or society resulted in being sent to a psychiatric institution. That is exactly the direction we are heading, and there is a continuum between where we are now (more liberty, more crime, and more inequality than most other developed nations) to where nations like the GDR are (no liberty, nearly no crime, nearly no inequality). Where people like you and Obama are vague, naive, or dishonest (it is hard to tell which), is that you promise to deliver the benefits without the loss of liberty. You haven't experienced it, I have: believe me, it is not a good direction to go into.
If you take very strong measures, yes, you could reduce suicide and homicide rates. I had relatives that lived in a country where that succeeded: the GDR. It is not an acceptable approach. Any actually feasible level of gun control, policing, and border control will not result in significant reductions.
3D printers complicate the issue even further, because in less than a decade, most people will be able to print working guns.
Yes, I deliberately used the ambiguous term "illegal gun" because addressing all those points in a single sentence would be pointless. If you really want to know, do some Google searches.
Furthermore you should go down to specifics: you want to pass laws to restrict our liberties, you need to show that these laws are in fact effective. Unless you can show clear evidence for a strong reduction in homicides through (more) gun control, there should be no additional gun control.
Let me add that I have never owned a gun, am not an NRA member, and never intend to own a gun. But the reasoning and swiftness with which Obama wants to take away liberties, as well as the lack of any sound scientific evidence, scares me, because the same approach can be taken to issues I do care about.
You're cherry-picking data, inferring causation from temporal correlation, and making leaps of logic.
In reality, there is no clear evidence that gun control has any positive (or negative) effect:
http://en.wikipedia.org/wiki/Gun_control
Furthermore, none of those studies take into account the tradeoffs. If you impose enough controls, you can indeed lower the crime rate; the GDR had a very low crime rate, but it wasn't a free society (I actually had relatives there). I do not want to trade liberties for safety.
Yes. About 2/3 of those uses are suicides, and the rest are almost all homicides with illegal guns. Gun control has no significant effect reducing either of these numbers. There is a small remainder of homicides committed with legally owned guns and accidents, but many legal products are far more dangerous. Furthermore, there is no justification for creating intrusive government regulation that prevents me from committing suicide with a gun.
How the hell do you know? Why do you keep fabricating these facts out of thin air?
I'm not assuming anything. I'm observing that there is no data showing that gun control reduces homicides, I'm merely suggesting why that is a plausible result. And by definition, gun homicides are caused by criminals.
Let's put these numbers into perspective. There are 16000 homicides in the US each year, 11000 gun related homicides and 19000 suicides with guns. Both criminals and suicidal persons will seek out and obtain alternatives to achieve their goal.
There are about 120 accidental deaths from firearms in children. These are due to parental negligence, and they are a miniscule fraction of overall preventable accidental deaths. Are you saying that all products that might cause accidental death in children should be controlled? Why?
You have presented no evidence that wrongful killing of home invaders by homeowners is a significant cause of death. Where is your data? And what's the alternative? Why should I be forced to face an armed invader into my home unarmed? And what makes you think that home owners won't find lethal alternatives?
You and people like you are arguing just like the Christian conservatives argue against abortion and gay marriage: you can't provide a shred of evidence that the restrictions you want to impose are effective, but you are ideologically opposed to it, so you keep making up hypotheticals and vague stories about how children are threatened.
And you have another thing in common with those Christian conservatives: given that guns are primarily used for suicide, much of what the gun control debate is really about is to deprive people of control over their own life and body, putting obstacles in their way for committing suicide.
That's false. Java claims to provide sandboxing, runtime safety, and fault isolation, something C never provided. In different words, security can be handled by the OS, but it can also be handled by the language and runtime. Providing it in the language and runtime has lots of potential advantages. Java tried to provide support in the language and runtime, but failed. Other languages and runtimes have succeeded at accomplishing what Java failed to do with sandboxing, runtime safety, and fault isolation, so it's certainly possible to do in principle.
You said that there was a "pattern", but there is no pattern in the data. And you refuse to put up any data showing your "pattern".
Yes, and that's why gun control is ineffective: criminals can get them no matter how much you restrict legal gun ownership. It's the same reason the war on drugs has been a failure.
Sometimes less is more. I have a tablet hooked up to a monitor, but Android constantly gets confused about the two screens and their resolutions. That means you keep having to fiddle with the touch screen. A dedicated device like this always uses the connected monitor for its output, and the mouse and keyboard for its input. Also, this has better specs than low-end tablets; in different words, at the same price, you get better performance for not paying for a screen.
Well, the morale of Apple workers is clearly generally great. And their products are pretty good, but you also pay for it. But that's no different from most other manufacturers of expensive luxury products, and that's what Apple effectively is.
That's utter nonsense. In reality, Linux is no harder (and probably easier) to maintain than Windows: it's less complicated, its interfaces and tools change less over time, and there are numerous ready-made, turnkey server solutions that you basically just boot up and use. More CS grads will know and will have administered Linux systems than Windows systems because administration for Linux systems is much more similar between single user and multi-user.
School administrators may choose Windows because they believe it's easier to maintain, but they are mistaken.
When talking about small scale websites, language isn't relevant, because such sites usually operate far below server capacity, and small scale web sites make up the bulk of the web. But for large scale web sites, and in particular web sites that have to perform compute intensive operations for users (image processing, speech recognition, etc.), performance, and hence language, matters a great deal: a factor of two speedup cuts your hardware costs in half.
In fact, little performance critical code is written in Java, Ruby, or languages like that because their performance is lousy. As you note yourself, people convert performance critical parts to C or C++. But such mixed language systems are hard to maintain, and C/C++ programming itself is costly and risky. A language that combines runtime safety with near-C performance is an excellent compromise. There are several languages that offer that, including C# and D, but industry inertia keeps them from being adopted.
One more thing: gun homicide rates are irrelevant. What matters is total homicide rates, since that's what we want to lower.
If outlawing guns merely causes people to shift to other means of committing murder, we don't need to bother outlawing guns.
Sorry, here's the correct link for Western Europe.
http://minus.com/lbrJeBeSvITMig
It ate part of my comment...
I had done all the graphs. There is no correlation for all nations. There is no correlation for all nations with homicide rates less than 0.6. There is no correlation for OECD countries. There is no correlation for OECD countries without outliers (US, Mexico). There is no correlation for Western Europe. There is no correlation for US states.
I had done all the graphs. There is no correlation for all nations. There is no correlation for all nations with homicide rates
http://minus.com/lddhssi1nFFVu
http://minus.com/lXDFvzhp8KK7
http://minus.com/lbyuu4u73nhKcE
Yes, there are too many confounding variables. There are more confounding variables than there are countries. That's why the hypothesis "higher gun ownership causes higher homicide rates" wouldn't be supported even if there was a correlation and even if you think you can justify a choice of a particular subset to look at. And there are far fewer confounding variables among US states, and there still is no correlation.
Sorry, but there simply is no statistical support for the hypothesis that higher gun ownership rates causes higher homicide rates, nor for the hypothesis that gun control lowers homicide rates. If you disagree, the burden is really on you to provide statistically significant evidence for the hypothesis.