Most developers will happily take good patches. So, if you have a particular problem, find a programmer who will work on the code and pay him to fix it for you, and send it back to the maintainer. It doesn't have to be the main developer - plenty of OSS support shops can handle this kind of work.
If most people get in the habit of doing this, then it all works up and down the stack.
The trick with the 'redhat' approach for 'projectfoo' is that the 'redhat' in that model represents centralized control/funding/responsibility and erects a barrier to entry for changes that aren't of immediate obvious benefit (some of the best). And if that entity goes away, you're likely to have to overcome huge momentum to find a new one. If there's a management failure, you're just stuck. De-centralization is one of the major benefits of OSS - git-tree-style branching allows for simultaneous cooperation and competition, which is really a unique and powerful development and potentially economic model.
Um, this is Slashdot? Old people with Alzheimers today wander off and are generally returned by a large society with docile environments. There's ample evidence for a low average life expectancy of early humans in the fossil record.
And no, I don't mean the misapplication of statistics that makes idiots think that "average" is the same as "median" or "mode".
I'm not sure what this has to do with the matter? They can tell by the bones how old somebody was when died. Mode probably doesn't matter, median and mean are probably about as useful as each other in this context.
It seems there are lots of folks who get the intrinsic benefits of FLOSS, get why there are benefits of paying for support, but then decline to pay for support for FLOSS.
One would think this would be a pure value judgement, but it's not. I'd be curious if some could offer a hypothesis.
Intel beat Apple last time around - now they're working together. I suspect Apple learned exactly what not to do with Firewire. At least we now know why Apple bought Zayante and then apparently let Firewire wither on the daisy chain.
Looking at the mess of wires on my desk at the moment, I say, "bring it". And, lord help us, make the connectors gender neutral so we can all just use a single cable for everything.
That's a non-sequitor. Grandparents or great-grandparents is irrelevant.
Of course it's relevant, it's the law of diminishing returns. Does a 75 year old really impart so much more wisdom to a tribe than a 60 year old does? A 90-year old? Does that wisdom imparted provide significant enough benefit to weigh against the resource requirements in order to achieve reproduction and child-rearing age improvements, say 35 years or so for an average lifespan?
Believe it or not, people used to value the elderly, and in many parts of the world, they still do.
There's no point in starting a modern cultural relativism argument. For the purposes of bioevolutionary selection pressures, the arguments need to be limited to the development of modern humans. That's mostly likely going to be 15,000-200,000 years ago or more. It's difficult to impossible to establish ideas about cultural norms that far back in history. If there's evidence to the contrary, I'd be quite interested in seeing it.
Yeah, you're absolutely right. I don't know what Google is like, but there are some tech companies where the lawyers are itchin' to sue everybody in sight and they get held back by more reasoned heads until some line is crossed. IP problems vs. operational problems could be such a line - given how engineer-heavy Google is, I can see a culture that would have decided that trust among engineers was lost with the decision to distribute a private beta of a client/server app. Since Google isn't want to sue everybody in sight, my speculation is that something like this was going on. Or maybe this will define a bright line in their descent into Evil - we'll see.
Tiredness is caused by a chemical regulated by the Circadian Rhythm.
Right, the release of melatonin will make you feel sleepy right now, but what causes the Circadian Rhythm to establish/maintain? I can't seem to find the right search on pubmed tonight, but there was recent research to suggest that an accumulation of misfolded proteins during the course of the day is at least one of the mediators, and that those are cleaned up during the rest cycle, making the rhythm self-establishing.
Healthy old folks are directly beneficial to social groups. They remember how to solve problems. They can take care of grandchildren while the parents are off gathering food. And many social groups had old folks. Sure, most people didn't live to get very old. But some did.
That's absolutely true of grandparents, but we're talking about great-great grandparents here, from a bio-evolutionary standpoint.
Could delaying the inevitable onset of Alzheimer's be the biological function of sleep? Last I heard, the purpose of sleep wasn't entirely clear
What? No. There are at least two functions of sleep that I know of: one is cleaning up misshapen proteins that accumulate during the day (and may be what causes tiredness). The other is transcription of short-term memory into long-term memory. Evolutionarily speaking, nobody ever lived long enough to get Alzheimers. Those who did wandered off into the tundra and didn't burden the tribe any longer.
The instructions stink too. I'm pretty consistently putting stuff together from bad instructions, but this one really takes the cake. I checked with my kid to make sure I wasn't just overthinking it. I suspect they make perfect sense for somebody who has been involved with the CAD design of the unit, but some user testing seems to have been left out. I bought two extras for the cousins and I know I'll have to be involved with the assembly.
people are going to beat down hijackers now, on their own
Shhh.. if you mention that the 9/11 problem was solved in a plane over a field in rural PA just over an hour after the first plane hit the towers by ordinary Americans (who can comprehend real security very well) then there's no need for massive expansion of government. Why aren't you patriotic?
What's next, are you going to tell us that with hardened cockpit doors there's absolutely no need to confiscate small pointy objects from passengers? You one of those terrists, son?
There are some videos of Eyefinity at work in this article, here is a direct link as well:
Hrm, this reminds me of playing AH-10 Attack on my Mac IIvx with a couple nuBus video cards added and some old monitors. Front screen got the front view; left screen, left window; right screen, right window. It was really quite fun for the time (c. 1992?). I had no idea a generation later the technology had gone missing (after college, game time dried up).
Google are taking the matter into their own hands and actually putting resources towards improving IE, because they know that MS will not do it in any reasonable way.
Prediction: when YouTube dumps Flash, the new 'YouTube installer' is this.
Heaven forbid that the geek should be encouraged to contribute new stories based on his own characters and settings. To throw away the crutch of fan fiction.
Apple uses a hybrid not-exactly-a-microkernel-anymore approach, but yeah, their drivers are out at an abstraction layer. There is a pretty large performance penalty with their I/O system, databases for instance are terrible. Anything that has to send a 'message' through the microkernel get slowed down. Databases may be on the order of 3-5x slower than on a linux box on the same hardware.
How much of this is Apple's code quality and architecture vs. being endemic to microkernels, though... there's probably some difference splitting to be done in that examination.
It's kind of like a crippled BSD server with weird management utilities and a lot of buggy modified utilities.
No, it's not like BSD. With BSD, real bugs embarrass developers and you stand a good chance of getting stuff fixed, or at least an apology and a workaround offered in the short term. With OSX, lots of people on the lists, most of whom are technically clueless about the inner-workings (though no doubt good at their primary jobs) will have the same problems, but there are few solutions. Apple is mostly non-responsive to problem reports and doesn't participate on the lists.
You might as well just use a normal Linux server, since all the same daemons are available, and much easier to manage.
I'm very interested to know how to do the network logins, distributed permissions, ARD-type management, etc. I didn't think those things existed.
perhaps there are incidental rewards to those resources having been used
Right - everybody who seriously competed greatly enhanced their own personal knowledge of the field. I'd bet that most of that new working knowledge is not left to waste. There is a ripe market for prediction systems, and even the worst of the entrants can probably fulfil somebody's small need.
...microkernels are better *in the long run*? You only add what you really need and want to, to the basic kernel? Sort of like FF as a browser, which comes stock with just pure browser functionality and no more, then you salt to taste with thousands of add-ons?
That was my thought as well - Linus has always been against microkernels, they have a performance penalty, but surely he's not arguing against features? Microkernels are the obvious way out of this box.
But then there's the performance problem. This makes me wonder - could hardware support for whatever it is that is the latency killer be developed that would support efficient microkernels? Virtualization was solved this way. I wonder if even those instructions would help. Just thinking out loud, I am not a kernel designer.
Slashdot Mirror
Most developers will happily take good patches. So, if you have a particular problem, find a programmer who will work on the code and pay him to fix it for you, and send it back to the maintainer. It doesn't have to be the main developer - plenty of OSS support shops can handle this kind of work.
If most people get in the habit of doing this, then it all works up and down the stack.
The trick with the 'redhat' approach for 'projectfoo' is that the 'redhat' in that model represents centralized control/funding/responsibility and erects a barrier to entry for changes that aren't of immediate obvious benefit (some of the best). And if that entity goes away, you're likely to have to overcome huge momentum to find a new one. If there's a management failure, you're just stuck. De-centralization is one of the major benefits of OSS - git-tree-style branching allows for simultaneous cooperation and competition, which is really a unique and powerful development and potentially economic model.
[citation needed]
Um, this is Slashdot? Old people with Alzheimers today wander off and are generally returned by a large society with docile environments. There's ample evidence for a low average life expectancy of early humans in the fossil record.
And no, I don't mean the misapplication of statistics that makes idiots think that "average" is the same as "median" or "mode".
I'm not sure what this has to do with the matter? They can tell by the bones how old somebody was when died. Mode probably doesn't matter, median and mean are probably about as useful as each other in this context.
You'd wish you could pay to have it maintained.
You can. Why don't you?
It seems there are lots of folks who get the intrinsic benefits of FLOSS, get why there are benefits of paying for support, but then decline to pay for support for FLOSS.
One would think this would be a pure value judgement, but it's not. I'd be curious if some could offer a hypothesis.
Intel beat Apple last time around - now they're working together. I suspect Apple learned exactly what not to do with Firewire. At least we now know why Apple bought Zayante and then apparently let Firewire wither on the daisy chain.
Looking at the mess of wires on my desk at the moment, I say, "bring it". And, lord help us, make the connectors gender neutral so we can all just use a single cable for everything.
They guy pushing the button my believe the 71 virgin shtick.
We just need to let them know that nobody ever said those 71 virgins were women.
That's a non-sequitor. Grandparents or great-grandparents is irrelevant.
Of course it's relevant, it's the law of diminishing returns. Does a 75 year old really impart so much more wisdom to a tribe than a 60 year old does? A 90-year old? Does that wisdom imparted provide significant enough benefit to weigh against the resource requirements in order to achieve reproduction and child-rearing age improvements, say 35 years or so for an average lifespan?
Believe it or not, people used to value the elderly, and in many parts of the world, they still do.
There's no point in starting a modern cultural relativism argument. For the purposes of bioevolutionary selection pressures, the arguments need to be limited to the development of modern humans. That's mostly likely going to be 15,000-200,000 years ago or more. It's difficult to impossible to establish ideas about cultural norms that far back in history. If there's evidence to the contrary, I'd be quite interested in seeing it.
Yeah, you're absolutely right. I don't know what Google is like, but there are some tech companies where the lawyers are itchin' to sue everybody in sight and they get held back by more reasoned heads until some line is crossed. IP problems vs. operational problems could be such a line - given how engineer-heavy Google is, I can see a culture that would have decided that trust among engineers was lost with the decision to distribute a private beta of a client/server app. Since Google isn't want to sue everybody in sight, my speculation is that something like this was going on. Or maybe this will define a bright line in their descent into Evil - we'll see.
Tiredness is caused by a chemical regulated by the Circadian Rhythm.
Right, the release of melatonin will make you feel sleepy right now, but what causes the Circadian Rhythm to establish/maintain? I can't seem to find the right search on pubmed tonight, but there was recent research to suggest that an accumulation of misfolded proteins during the course of the day is at least one of the mediators, and that those are cleaned up during the rest cycle, making the rhythm self-establishing.
Healthy old folks are directly beneficial to social groups. They remember how to solve problems. They can take care of grandchildren while the parents are off gathering food. And many social groups had old folks. Sure, most people didn't live to get very old. But some did.
That's absolutely true of grandparents, but we're talking about great-great grandparents here, from a bio-evolutionary standpoint.
way to poke the bull in the eyes there - so he's not just violating license, he's probably causing back-end problems too. That was dumb.
What the hell is going here? I see a story about Corona CA evading the law on red-light cameras and comments (and tags) are about some MS story?
Somebody please start some threads about random conspiracy theories - they'll get lots of hits and really confuse the hell out of everybody.
I doubt there's a single 30 line block of code anywhere that isn't violating someones patent.
TFTFY
Could delaying the inevitable onset of Alzheimer's be the biological function of sleep? Last I heard, the purpose of sleep wasn't entirely clear
What? No. There are at least two functions of sleep that I know of: one is cleaning up misshapen proteins that accumulate during the day (and may be what causes tiredness). The other is transcription of short-term memory into long-term memory. Evolutionarily speaking, nobody ever lived long enough to get Alzheimers. Those who did wandered off into the tundra and didn't burden the tribe any longer.
Thank you! I'll send printouts of one or both along with the gift 'scopes.
I doubt in 1789 that any customs people were reading the personal papers of travelers to see if they had anything naughty written on them.
The instructions stink too. I'm pretty consistently putting stuff together from bad instructions, but this one really takes the cake. I checked with my kid to make sure I wasn't just overthinking it. I suspect they make perfect sense for somebody who has been involved with the CAD design of the unit, but some user testing seems to have been left out. I bought two extras for the cousins and I know I'll have to be involved with the assembly.
people are going to beat down hijackers now, on their own
Shhh.. if you mention that the 9/11 problem was solved in a plane over a field in rural PA just over an hour after the first plane hit the towers by ordinary Americans (who can comprehend real security very well) then there's no need for massive expansion of government. Why aren't you patriotic?
What's next, are you going to tell us that with hardened cockpit doors there's absolutely no need to confiscate small pointy objects from passengers? You one of those terrists, son?
There are some videos of Eyefinity at work in this article, here is a direct link as well:
Hrm, this reminds me of playing AH-10 Attack on my Mac IIvx with a couple nuBus video cards added and some old monitors. Front screen got the front view; left screen, left window; right screen, right window. It was really quite fun for the time (c. 1992?). I had no idea a generation later the technology had gone missing (after college, game time dried up).
Prediction: when YouTube dumps Flash, the new 'YouTube installer' is this.
Heaven forbid that the geek should be encouraged to contribute new stories based on his own characters and settings. To throw away the crutch of fan fiction.
But that's all* Disney does.
* for large values of all
Wonder what Perimeter would have done if switched on back then.
No need to wonder, TFA explains how it actually works.
Apple uses a hybrid not-exactly-a-microkernel-anymore approach, but yeah, their drivers are out at an abstraction layer. There is a pretty large performance penalty with their I/O system, databases for instance are terrible. Anything that has to send a 'message' through the microkernel get slowed down. Databases may be on the order of 3-5x slower than on a linux box on the same hardware.
How much of this is Apple's code quality and architecture vs. being endemic to microkernels, though... there's probably some difference splitting to be done in that examination.
It really does suck.
seconded.
It's kind of like a crippled BSD server with weird management utilities and a lot of buggy modified utilities.
No, it's not like BSD. With BSD, real bugs embarrass developers and you stand a good chance of getting stuff fixed, or at least an apology and a workaround offered in the short term. With OSX, lots of people on the lists, most of whom are technically clueless about the inner-workings (though no doubt good at their primary jobs) will have the same problems, but there are few solutions. Apple is mostly non-responsive to problem reports and doesn't participate on the lists.
You might as well just use a normal Linux server, since all the same daemons are available, and much easier to manage.
I'm very interested to know how to do the network logins, distributed permissions, ARD-type management, etc. I didn't think those things existed.
perhaps there are incidental rewards to those resources having been used
Right - everybody who seriously competed greatly enhanced their own personal knowledge of the field. I'd bet that most of that new working knowledge is not left to waste. There is a ripe market for prediction systems, and even the worst of the entrants can probably fulfil somebody's small need.
...microkernels are better *in the long run*? You only add what you really need and want to, to the basic kernel? Sort of like FF as a browser, which comes stock with just pure browser functionality and no more, then you salt to taste with thousands of add-ons?
That was my thought as well - Linus has always been against microkernels, they have a performance penalty, but surely he's not arguing against features? Microkernels are the obvious way out of this box.
But then there's the performance problem. This makes me wonder - could hardware support for whatever it is that is the latency killer be developed that would support efficient microkernels? Virtualization was solved this way. I wonder if even those instructions would help. Just thinking out loud, I am not a kernel designer.