Would you prefer he'd used a macro? Having looked at the code I think he could have. That would eliminate the function call, but it would be just as unintelligible and ugly. (Actually, if the compiler respects inline in this case it could even generate the same code.)
FWIW, the very code used in the accepted handler could be implemented as a macro, and I think the only gain would be unintelligibility. And the danger that sometime in the future someone might change the macro with undetected side effects.
Even more to the point, if you have just reason and get mad at someone, and they do the same thing again, and you get mad again, they are likely to leave of their own free will. As long as you have more people trying to join your group than you can handle, this will lead to an increasingly skilled group, as the ones who remain will be skilled, and able to learn to avoid repeating stupidities.
But for best results you've got to be good enough to not get wrong incorrectly, and if (when) you do to be willing to apologize in public.
I've got a bit of a memory about the state of C++ when kernel development was started, and it wasn't good. Or standardized. And IIRC it depended upon cfront, when translated the C++ code into C.
Years later C++ code was considerably bulky after being compiled and linked, and not at all suitable for many of the uses of the kernel.
It's my understanding that currently C++ code can be equally good, and the source more compact...but I'm not really certain this is true, and in any case that's not enough to justify a conversion effort.
Yah, Linux was a noob when he started the project, and he started it in C because that was what he knew. But it happened to be the best available choice, and may still be the best choice.
You are, of course, correct about why Linux was successful. But you need to remember that BCD Unix was not the only Unix around, so it's not as if there weren't competitors. But they wanted to charge exorbitant prices, and weren't interested in the microcomputer market anyway. So the lawsuits weren't the only reason, greed played a major role.
That's a selection of quotes, however, and occasionally he's been quite a bit more negative.
Still, the ones I almost remember were about how unresponsive the systemd maintainers were when an error was reported rather than about the basic idea.
No. The $500 hammer was because the vendor didn't want to do the paperwork to sell just one hammer, and the purchaser wasn't footing the bill. So the vendor charged enough to make it worth his while to do the paperwork.
Compiler bugs are one thing, but I once ran on a computer that had occasional transient memory errors. Just *TRY* to track that one down. The only thing I could do was try running the program multiple times...and we paid for time by the second.
I frequently wish there were some decent way to do tri-state logic. Possibly because my original language was FortranIV which didn't have a logical if, only the tri-state arithmetic if. But there's no variable type that's good for storing tri-state variables. (Yeah, I could code it for byte, but that always feels like such a waste.) And what I really want is a suite of tri-state operations.
As a compromise I usually substitute "yes, no, maybe, error", (in some order), but often either "maybe" or "error" is unreasonable.
When something's not broken, don't try to fix it. You may not want to work there, but there are lots of projects with different management styles. Many of them work...and working is important.
It's a style that I don't particularly like, but it works. It's one of several that work. If you decide to contribute to a project, pick one that's not only important to you, but whose management style you find acceptable. There are many to choose between. But check and be sure that not only do you like it, but that they style works.
If you're thinking about kernel development, follow the development list for awhile before committing yourself. (You'll probably find that it's a pretty reasonable style, even if it's occasionally newsworthy.)
It has been reported previously that Linux first asks politely and only if this is ignored does he ask impolitely. This has been said often enough, in enough contexts, that I believe it.
Also, we do not know what private communication has gone on. Why do you assume that he hasn't asked privately? But if there is no private communication, then what is the alternative to asking noisily and publicly?
Whatever, this is an approach that works. It's not the one that Guido uses for Python, but Guido's approach also works. It's not the one that Walter Bright uses for D, but that approach also works. You've got lots of choices for a project to work on, if that's your choice.
I don't know who "Dave cutler" is, but MSWindows has an EULA that, as far as I am concerned, makes it unusable.
Currently I use Linux, and I've considered various BSDs. Actually, I still consider them every time there's a new update, but I require that I be able to use my current file system, ext4, with them, and so far that's been a deal breaker. If BSD had gotten to me first, I'd be requiring that Linux read the native BSD file system before I considered it. MSWindows got to me first, and I required that Linux have read/write access to the MSWindows file system. Fortunately it was VAT. NT would have taken a few more years...and the MSWind EULA became unusable with MSWindows2000.
Is your firmware microcode? Otherwise I agree that firmware is just code stored on an ROM. Still, this doesn't mean it isn't something to be afraid of it you don't trust it.
With a Carrington Event the governors next door will have the same problem as the local ones, and so will the other countries.
The post was exaggerated when talking about past disasters. But we haven't had an major disasters. Actually, humanity hasn't had any major disasters within historical record that weren't caused by other people. The eruption of a volcano in the area of the Mediterranean is an example that's just slightly prehistoric, and that caused immense loss of life and trauma in the survivors. It probably caused the fall of Crete, among many other problems.
There definitely are many people who claim to be systems administrators who claim that systemd makes their job easier. I expect that for certain, perhaps many, use cases it's true. That is shouldn't be true for all use cases is hardly a surprise.
I've seen this put "One size doesn't fit all."
And BusyBox is aiming at SMALL computers. So systemd is probably not even usable by many of their target audience. And for most it's probably unreasonably excessive overhead.
She's a scapegoat. Not that she didn't do what was said...
If you want to deal properly with this figure the amount of time served by the prisoners (at 24 hours/day) times the median wage/hour in the country and fine the company that amount. Then divvy that up among the prisoners based on the amount of time served.
DON'T give the money to the govt. That a malign reinforcement. Perhaps you should take some of it from the prosecutors where there is evidence that they knew or had reason to know that the evidence was tainted.
That would align results with proper action, but it still wouldn't fix what is a basically broken system.
It's not just due to the "war on drugs", though that's a big part. Prisons run for profit is another part of the problem. And the system was in major trouble before the "war on drugs" was officially proclaimed. (If you want to go back to the start in the 1920's or 30's, perhaps you have a point, but I don't know enough to either agree *or* disagree.)
Sorry, but while the first step you propose is correct (and allow them to sue the individuals involved in railroading them for false imprisonment), the rest doesn't work.
The system as designed is basically flawed. Probably by intent. Punishment, while a necessary component, becomes destructive of rehabilitation when overdone, as it is in every prison I've ever heard described.
But the right approach wouldn't be cheap. First you need a thorough physical examination to deal with any problems. Then there should be a period of complete isolation. And complete means both that you can't talk to anyone, nobody can hear you, and the guards don't come in and beat you up (i.e., this is NOT solitary confinement as currently implemented). This needs to be broken after a couple of weeks by visits from a trained clinician...trained in, probably, Rational Emotive Behavior Therapy. Then small group workshops in an appropriate trade. Appropriate here means appropriate for their personality, not for earning a living. Auto mechanics will probably often be a good choice, or wood working, or home economics. Then develop along any "major" that they are willing to commit themselves to. During this time they need to be continually protected both from injuring someone else and from someone injuring them. Consider that they are "as if" young children living in the bodies of adults. And children can be vicious until they have learned better. Eventually they should be trained for release. I'd like to say trained for a job, and placed by a placement agency, but with unemployment as high as it is, and projected to increase substantially, I see no way that this is plausible. And I don't know what release training is plausible.
The course I propose is, of course, totally illegal. It involves cutting off all communications with their past for a period of time in a way that is only legal for the military, and in other ways it is similar to basic training. OTOH, it's quite different in other ways...and I hope that the parts I've left out aren't crucial, but they may be. One of the things left out is believable threat of death, but many traditions find this to be an essential component. Perhaps it's necessary to cut the ties to the prior life, and bind them to the new one.
You could have stopped after "People who think rationally". The rest isn't needed. I never expect to actually serve on a jury (though I did once, to my surprise), because as soon as the lawyers notice that you're paying attention to what they say they find some cause to exclude you.
P.S.: Don't assume it's for your political beliefs. The cases I've observed don't confirm that. The conservatives were as readily kicked off the jury as the liberals if they showed any awareness of rationality rather than emotional reactions. I *guess* that emotional reactions are more predictably manipulable. Cicero certainly thought so.
Well, it has to have been after the creation of language. Probably after the creation of pronouns.
OTOH, IIUC most tribal groups DON'T believe in an all-powerful being. Just a mighty ancestor, whose deeds become increasingly magnified as he retreats further into the past. Maui had a mother, e.g., and did what she told him to...usually.
So I'd guess that the "all powerful being" doesn't much predate Ikhnaton. And that it was usually a personification of the sun. (Note that early Judaic writings say things like "Thou shalt have no other gods before me.", not "There is no god but me."
OTOH, the idea of a mightier than plausible entity probably dates back far before language. I've heard this called the "Big Baboon" theory of religion. It mixes religion and politics and claims that gods evolved out of submission to the Alpha Ape of the pack. This is plausible, but is a far cry from omnipotence, which even Zeus and Jupiter didn't have claimed for them.
Slashdot Mirror
You cannot accurately assert that anymore than any particular Anonymous group can claim to speak for all anonymous groups.
You don't know the particular members of this group. They may all be following a consistent moral standard. It may even be a laudable moral standard.
OTOH, some reports have questioned the accuracy of some of the data. If those are accurate then the posting group is at minimum criminally reckless.
Actually, I do. But then I'm not in favor of capital punishment.
P.S.: I don't consider government approval to be a moral sanction.
Would you prefer he'd used a macro? Having looked at the code I think he could have. That would eliminate the function call, but it would be just as unintelligible and ugly. (Actually, if the compiler respects inline in this case it could even generate the same code.)
FWIW, the very code used in the accepted handler could be implemented as a macro, and I think the only gain would be unintelligibility. And the danger that sometime in the future someone might change the macro with undetected side effects.
Even more to the point, if you have just reason and get mad at someone, and they do the same thing again, and you get mad again, they are likely to leave of their own free will. As long as you have more people trying to join your group than you can handle, this will lead to an increasingly skilled group, as the ones who remain will be skilled, and able to learn to avoid repeating stupidities.
But for best results you've got to be good enough to not get wrong incorrectly, and if (when) you do to be willing to apologize in public.
I've got a bit of a memory about the state of C++ when kernel development was started, and it wasn't good. Or standardized. And IIRC it depended upon cfront, when translated the C++ code into C.
Years later C++ code was considerably bulky after being compiled and linked, and not at all suitable for many of the uses of the kernel.
It's my understanding that currently C++ code can be equally good, and the source more compact...but I'm not really certain this is true, and in any case that's not enough to justify a conversion effort.
Yah, Linux was a noob when he started the project, and he started it in C because that was what he knew. But it happened to be the best available choice, and may still be the best choice.
You are, of course, correct about why Linux was successful. But you need to remember that BCD Unix was not the only Unix around, so it's not as if there weren't competitors. But they wanted to charge exorbitant prices, and weren't interested in the microcomputer market anyway. So the lawsuits weren't the only reason, greed played a major role.
That's a selection of quotes, however, and occasionally he's been quite a bit more negative.
Still, the ones I almost remember were about how unresponsive the systemd maintainers were when an error was reported rather than about the basic idea.
No. The $500 hammer was because the vendor didn't want to do the paperwork to sell just one hammer, and the purchaser wasn't footing the bill. So the vendor charged enough to make it worth his while to do the paperwork.
Compiler bugs are one thing, but I once ran on a computer that had occasional transient memory errors. Just *TRY* to track that one down. The only thing I could do was try running the program multiple times...and we paid for time by the second.
I frequently wish there were some decent way to do tri-state logic. Possibly because my original language was FortranIV which didn't have a logical if, only the tri-state arithmetic if. But there's no variable type that's good for storing tri-state variables. (Yeah, I could code it for byte, but that always feels like such a waste.) And what I really want is a suite of tri-state operations.
As a compromise I usually substitute "yes, no, maybe, error", (in some order), but often either "maybe" or "error" is unreasonable.
So pick a different project to work on.
When something's not broken, don't try to fix it. You may not want to work there, but there are lots of projects with different management styles. Many of them work...and working is important.
It's a style that I don't particularly like, but it works. It's one of several that work. If you decide to contribute to a project, pick one that's not only important to you, but whose management style you find acceptable. There are many to choose between. But check and be sure that not only do you like it, but that they style works.
If you're thinking about kernel development, follow the development list for awhile before committing yourself. (You'll probably find that it's a pretty reasonable style, even if it's occasionally newsworthy.)
It has been reported previously that Linux first asks politely and only if this is ignored does he ask impolitely. This has been said often enough, in enough contexts, that I believe it.
Also, we do not know what private communication has gone on. Why do you assume that he hasn't asked privately? But if there is no private communication, then what is the alternative to asking noisily and publicly?
Whatever, this is an approach that works. It's not the one that Guido uses for Python, but Guido's approach also works. It's not the one that Walter Bright uses for D, but that approach also works. You've got lots of choices for a project to work on, if that's your choice.
I don't know who "Dave cutler" is, but MSWindows has an EULA that, as far as I am concerned, makes it unusable.
Currently I use Linux, and I've considered various BSDs. Actually, I still consider them every time there's a new update, but I require that I be able to use my current file system, ext4, with them, and so far that's been a deal breaker. If BSD had gotten to me first, I'd be requiring that Linux read the native BSD file system before I considered it. MSWindows got to me first, and I required that Linux have read/write access to the MSWindows file system. Fortunately it was VAT. NT would have taken a few more years...and the MSWind EULA became unusable with MSWindows2000.
While that's generally true, there have been cases where common code has had potential exploits that have hung around for decades.
Is your firmware microcode? Otherwise I agree that firmware is just code stored on an ROM. Still, this doesn't mean it isn't something to be afraid of it you don't trust it.
With a Carrington Event the governors next door will have the same problem as the local ones, and so will the other countries.
The post was exaggerated when talking about past disasters. But we haven't had an major disasters. Actually, humanity hasn't had any major disasters within historical record that weren't caused by other people. The eruption of a volcano in the area of the Mediterranean is an example that's just slightly prehistoric, and that caused immense loss of life and trauma in the survivors. It probably caused the fall of Crete, among many other problems.
There definitely are many people who claim to be systems administrators who claim that systemd makes their job easier. I expect that for certain, perhaps many, use cases it's true. That is shouldn't be true for all use cases is hardly a surprise.
I've seen this put "One size doesn't fit all."
And BusyBox is aiming at SMALL computers. So systemd is probably not even usable by many of their target audience. And for most it's probably unreasonably excessive overhead.
He's also an anonymous coward, and probably a troll.
Who pays? Most people have to struggle to afford even a basic defense lawyer. If you don't see that as a problem, then you are a part of the problem.
She's a scapegoat. Not that she didn't do what was said...
If you want to deal properly with this figure the amount of time served by the prisoners (at 24 hours/day) times the median wage/hour in the country and fine the company that amount. Then divvy that up among the prisoners based on the amount of time served.
DON'T give the money to the govt. That a malign reinforcement. Perhaps you should take some of it from the prosecutors where there is evidence that they knew or had reason to know that the evidence was tainted.
That would align results with proper action, but it still wouldn't fix what is a basically broken system.
It's not just due to the "war on drugs", though that's a big part. Prisons run for profit is another part of the problem. And the system was in major trouble before the "war on drugs" was officially proclaimed. (If you want to go back to the start in the 1920's or 30's, perhaps you have a point, but I don't know enough to either agree *or* disagree.)
Sorry, but while the first step you propose is correct (and allow them to sue the individuals involved in railroading them for false imprisonment), the rest doesn't work.
The system as designed is basically flawed. Probably by intent. Punishment, while a necessary component, becomes destructive of rehabilitation when overdone, as it is in every prison I've ever heard described.
But the right approach wouldn't be cheap. First you need a thorough physical examination to deal with any problems.
Then there should be a period of complete isolation. And complete means both that you can't talk to anyone, nobody can hear you, and the guards don't come in and beat you up (i.e., this is NOT solitary confinement as currently implemented). This needs to be broken after a couple of weeks by visits from a trained clinician...trained in, probably, Rational Emotive Behavior Therapy. Then small group workshops in an appropriate trade. Appropriate here means appropriate for their personality, not for earning a living. Auto mechanics will probably often be a good choice, or wood working, or home economics. Then develop along any "major" that they are willing to commit themselves to. During this time they need to be continually protected both from injuring someone else and from someone injuring them. Consider that they are "as if" young children living in the bodies of adults. And children can be vicious until they have learned better. Eventually they should be trained for release. I'd like to say trained for a job, and placed by a placement agency, but with unemployment as high as it is, and projected to increase substantially, I see no way that this is plausible. And I don't know what release training is plausible.
The course I propose is, of course, totally illegal. It involves cutting off all communications with their past for a period of time in a way that is only legal for the military, and in other ways it is similar to basic training. OTOH, it's quite different in other ways...and I hope that the parts I've left out aren't crucial, but they may be. One of the things left out is believable threat of death, but many traditions find this to be an essential component. Perhaps it's necessary to cut the ties to the prior life, and bind them to the new one.
While the evidence seems to show she was guilty as charged, do you really believe that her management didn't know or have reason to know?
You could have stopped after "People who think rationally". The rest isn't needed. I never expect to actually serve on a jury (though I did once, to my surprise), because as soon as the lawyers notice that you're paying attention to what they say they find some cause to exclude you.
P.S.: Don't assume it's for your political beliefs. The cases I've observed don't confirm that. The conservatives were as readily kicked off the jury as the liberals if they showed any awareness of rationality rather than emotional reactions. I *guess* that emotional reactions are more predictably manipulable. Cicero certainly thought so.
Well, it has to have been after the creation of language. Probably after the creation of pronouns.
OTOH, IIUC most tribal groups DON'T believe in an all-powerful being. Just a mighty ancestor, whose deeds become increasingly magnified as he retreats further into the past. Maui had a mother, e.g., and did what she told him to...usually.
So I'd guess that the "all powerful being" doesn't much predate Ikhnaton. And that it was usually a personification of the sun. (Note that early Judaic writings say things like "Thou shalt have no other gods before me.", not "There is no god but me."
OTOH, the idea of a mightier than plausible entity probably dates back far before language. I've heard this called the "Big Baboon" theory of religion. It mixes religion and politics and claims that gods evolved out of submission to the Alpha Ape of the pack. This is plausible, but is a far cry from omnipotence, which even Zeus and Jupiter didn't have claimed for them.