can be found in a post in his live journal. He reports that with every new kernel release, the number of kernel related bug reports in the Fedora bugzilla goes up substantially.
(Davej is a long time kernel hacker and currently the Fedora kernel maintainer.)
You know, the author mentions three claims (paraphrased by me):
1. Global warming is happening (significantly so)
2. CO2 level is increasing
3. CO2 contributes to future warming.
He actually agrees with all of them.
The only thing he disagrees with is whether this is caused by civilization.
Now, in the last IPCC report (THE source for the scientific consensus, at least of the one attacked by Lindzen in this article), it is said that human activities are "likely" the cause of global warming. Not "certainly", but "likely". Even on an earth that is getting hot, I can't see how this is suppressing the view that humans may not be the cause of global warming.
In fact, the only thing he is attacking with any substance are the casual, frequently overheard claims that the recent increase in storms, tornados etc. is caused by global warming. Well, that's a straw-man as far as I am concerned. I have never heard this a scientific claim, just as an informal "could well be" answer e.g. by meteorologist in reply to question by journalists.
Some facts from the article not mentioned in the summary:
This decision isn't final yet, the parents will most likely appeal.
The crucial argument in the court's decision seems to have been that the personal rights of the parents were not violated, since they could not be identified by their last name. This is actually disputable, their name is pretty unique in Germany. (A search in the phone directory didn't turn up anyone with the name.)
The court did not consider the mentioning of the name a violation of Tron's own personal rights.
("Personal rights" is my translation by me of "Persoenlichkeitsrechte", which is technical term in German law speak. Maybe "Right to personal privacy" would be a better translation.)
...to remove the name from article, IMHO. It was important to defend the right to give the name. But the name does not add anything to the article, and if it hurts the feeling of those closely involved, there is a good case for not mentioning it.
Just because you CAN mention the name, this doesn't mean you have to.
The slashdot summary does not quite get the proportions right.
Wow, you have worries. If once in a while the summary is both correct and does have something to do with tfa, I rather feel like it's time to celebrate...
The waste material isn't actually that much of a problem. It's dangerous stuff, and you can't really "dispose" of it, I.E. leave it somewhere and forget about it. You've gotta live with it. Hundred of thousands of tonnes. But actually, it's not that much. Almost all of France's waste for the past 40 years sits in a place the size of a large warehouse.
Well, the problem is that you have to store it for some 10,000 years. That's 2500 warehouses of pretty dangerous stuff, that you have to protect for a very long time. Protect it from criminals, terrorists, natural disasters. Again for 10,000 years!
And that's only the dangers we think of at the moment. Are you really so sure we will have a stable enough government for 10,000 years to come to guarantee just the basic protection of the waste storage sites?
It is beyond me to estimate the dangers of running a nuclear power plant, whether it is worth the risk. But the nuclear waste problem is what makes me want to get rid of nuclear power.
(But then, I am from Germany, probably the country most critical of nuclear power all over Europe.)
...I can actually say that this sounds a lot like RMS' opinion, and in my opinion this opinion causes a lot of practical harm.
His opinion has a lot of annoying consequences, for example we are not allowed to provide Windows binaries for download on our project home page. (Our project is very portable, and we maintain MSVC project files.) Instead, people have to google for a Windows binary of our program instead, or look around on some other websites for a useful link. I would so much love to make it really easy for Windows Users to play around with our program; in its niche it could actually get a lot of people in touch with open source.
In fact, several of our current developers trying things out in a purely windows environment, and have then switched to Linux when they realized how much better the development environment in any Linux distribution is, compared to anything you can get for free on Windows. One of them had not even known about open source before he got started.
Big parts of this entire discussion is completely off-topic. Linus is not talking about Specs that define what your software is supposed to do, when you have a customer, you of course have to agree on one and then implement it. He is talking about relying on hardware specs when you want your software to interoperate with actual devices.
How do you think a BIOS is typically developed:
Reading the ACPI spec, implementing it, followed by extensive unit testing verifying that the spec was implemented correctly.
Browsing the ACPI spec, implementing it, then trying to boot windows until it works.
If you write the Linus ACPI code according to the spec, you are betting that the first point above is reality. Not the kind of bet you would like to make.
Of course, this example works as well with s/ACPI/HTML/; s/boot windows/load the page in IE/ etc. etc.
People here in Germany are constantly worried that the research in Germany universities might not be able to compete with the top places in the world anymore. This finally proves them all sooo wrong!
What if he gets hit by a bus? What would happen then?
Nobody knows but I am not worried. The Linux kernel community has always been great in adopting to new circumstances. Alan Cox decided to drop kernel work for a year to do a MBA? No problem, his role got taken over by a couple of people. Linus has to decide to drop BK? No big problem, Linus start writing GIT, which is quickly taken over by other people, and after 2 weeks development continues almost as if nothing happened. Dave Jones gets overworked pushing bug fixes from 2.4 to 2.6? No problem, maintainers step in and push the stuff themselves.
I have a lot of trust in the robustness of the LK development.
Of course, development would suffer without Andrew Morton. His skill in managing the process is just impressive. But it would not break the process.
I'm fairly ignorant about the kernel development process, so I ask: could automated testing play a greater role in the quality assurance of the project?
Short answer: no.
Long answer: Of course, automatic testing has its role, and the OSDL and LTP guys have been doing this regularly. It is great to monitor for any kind of regressions, including performance-wise.
But most kernel bugs are new, and it is pretty tough to set up a kernel test that intelligently discovers new bugs.
Maybe even more importantly, most kernel bugs are pretty unique. They happen when you start a weird program foobar on a weird system with a fooznoo gadget. No way you can test for this with an automatic setup. The usage scenarios of a Linux kernel in the real world have such a huge variety on such a diversity of hardware that nothing can beat the testing by us suckers who download the latest kernel and, guess what, use it.
LWN (Linux Weekly News) has just done a review of personal finance managers for Linux: http://lwn.net/Articles/149383/ (It will become freely available this thursday.)
I am a high school mathematics teacher and I train students for mathematics competitions.
Funny that you mention this, I also do training for the IMO team in my country.
Dr. Wildberger has an idea that he thinks will make trigonometry more intuitive and I hope he is really onto something here.
I think a more appropriate characterization would be the following: Wildberger is showing how some computations can avoid angles entirely (in particular, when you don't have to add angles!), and are more intuitive when you avoid this and thus trigonometry. He puts them into somewhat nice framework. I agree it may sometimes lead to more elegant (but essential equivalent) solution to cute little geometry problems.
However, this doesn't go well with the spin he is giving this himself: "For the past two thousand years we have relied on the false assumptions that distance is the best way to measure the separation of two points, and that angle is the best way to measure the separation of two lines."
Sorry, but distance is the best way to measure separation of points.
Or further down: "So teachers have resigned themselves to teaching students about circles and pi and complicated trigonometric functions that relate circular arc lengths to x and y projections - all in order to analyse triangles." What a non-sense. Teacher don't teach students trigonometry in order to analyse triangles. They teach them trigonometry because they are some of the most fundamental functions in mathematics, popping up all over in physics and engineering (and because they cannot teach them the complex exponential function, which does the same more elegantly). And analysing triangles is a cute way of doing something with them.
Sorry to spoil the fun, but while his approach is another way of presenting trigonomic geometry that some people might find cute (I don't care for it), this buzz about "establishing new foundations" of geometry is absolute non-sense.
Here is what he does: He replaces the distances by its square, and calls it a squandrance. He replaces the angle by the square of its sine and calls it a spread.
Ok, the relations between the squandrances and spreads of a rectangular triangle are simpler than those between its sides and its angles -- they are just simply obtained from pythagoras' theorem.
However, two way more fundamental relations suddenly become horribly difficult: Say going from town A to C I go from A to B then from B to C. A to B is 5 squandrance, B to C is 3 squandrance, how much squandrance is it from A to C? No, not 8, but 5 + 3 + 2*sqrt(8). The simple addition of distances becomes a square-root function...
Also, say I turn left by 30 degree, and then do that again. Guess what, I turned left by 2*30 = 60 degrees. However, if you were doing this in spreads, you don't want me to tell you the answer. I think there is a reason why sailors have been using angles and not spreads...
As for engineers: Why do they have to learn trigonometry? Not for geometry, you can just do that with coordinates. But because it turns up in waves everywhere, which in turn turn up everywhere. (Aside: the mathematical reason is that sine and cosine are the solutions of one of the most fundamental differential equations.) For example, the voltage of an AC current will depend proportionally on a sine fucntion. The angle you substitute in sine will be proportional to time. The spread you would have to substitute in a rational geometry function instead is NOT proportional to time, but to the square of the sine of time. Too bad my stopwatch measures time in seconds and not in the square of its sine...
Will his approach lead to faster computations? Of course not. Whereever it would help, people of course already knew last year already (and maybe even, gosh, in 2003) to use some trigonometric substitution....
Now I finally understand why cell phones are banned in the immigration area at least at Detroit Metropolitan Airport, it has never made sense to me before...
Of course, there are now many "I told you so"'s coming around, telling the Linux kernel developers that they never should have agreed to an odd, revokable license. (And I mostly agree.) But one of Linus' points was always "When Larry gets evil, we can just export the data and switch to another SCM." Which is exactly what is happening now. He told us so.
And in fact I have a lot of respect for Linus for dropping BK now right on the spot, when the problems it created were becoming too big -- after he had been unimpressed by 3 years of anti-BK flaming...
This has been quoted so many times, I still think it is a silly exaggeration. A couple of things changed when Linus switched to BK:
1. He wrote scripts so that he didn't have to jump between applying patches and reading e-mail, instead he is now reading a batch of patches, queuing them, and then starts a script to apply them.
2. Developers have instant access to Linus' tree. Any source control system would have provided this.
3. The comments to the patches in the e-mails sent to Linus now actually make it to the public. Just about any GNU project does this via ChangeLog under any revision control system.
4. A script was written to automatically extract release notes from the changelog comments.
5. Merging with subsystems maintainer is easier if they pile up the patches in bitkeeper repositories.
Maybe all of the above together yielded a factor of two. But only with respect to 5. is BK at all relevant. And even there -- by a HUGE amount the largest merge point is Andrew Morton, who uses quilt instead of BK to manage his tree with some hundred patches per week throughput to Linus. And I haven't read any complaints from Linus that he isn't using BK.
While a few developers liked the idea, most preferred to get more serious about the 2.6.x.y idea. And since Greg "the USB-guy" KH and Chris Wright volunteered to maintain it, it seems like this will be taken seriously this time.
This sounds like the best solution in the best interest of actual users.
Well, perhaps Linux is maturing enough where it could be 2.6 forever.
There is always room for improvement, new ideas, new architectures, hardware, etc that open up new pathways to more flexible or secure operating system organization.
What you are missing that the linux kernel development process has matured quite a lot. Now there is a steady stream of new features into 2.6. There is no backlog of huge patches that introduce new features and are available only in vendor kernels.
I wouldn't bet a lot of money on it, but I wouldn't be surprised if the current kernel in 5 years was 2.6.xx (which still would look completely different to current 2.6.11).
Still, this means that the security community has to always play catch up. Wouldn't it be much better if Linus dropped an e-mail to vendor-sec 2 days before he applies a security patch to the kernel, so that vendors can release a new kernel quickly, and users of a mainline kernel will be informed with an announcement immediately once the patch gets public?
the only real problem with last years election is that for most of/.ers, the wrong guy won. the us civil rights commission did two exhaustive studies of florida. guess what? nothing. no fraud, no intimidation, no disenfranchisement.
Maybe I shouldn't get involved in this inner-american debate, but I just think stronlgy this is non-sense, and I will try to explain why.
Fraud, intimidation etc. are only the worst of irregularities that can occur. I think you should set your standard much higher than that. I find any sign of non-uniformity of the process highly disturbing. Having to queue for 4 hours to vote in some areas, 1 hours somewhere else, and 5 min at other places is just unacceptable. If you start accepting this, how will you stop an election official from purposely assigning few resources to neighborhoods traditionally voting for the other side, thus inducing longer lines there that will certainly turn some people away?
Similarly, some 100 lost votes due to incorrect voting machine usage -- never mind these votes would not have changed the outcome, this is a just a disaster. I am sure noone has done this on purpose, but just the possibility of this allegation is a loss for the credibility of democracy.
In the game of bridge, there is a nice law (yeah, bridge has lots of rules, and they call them laws) about organizing tournaments: when pre-dealt hands are used (which makes sense for many reasons), there is the danger that anyone seeing the hands in advance could easily win the tournament. That law says: "The pre-dealing of hands should be organized in such a manner that any allegation of fraud can be convincingly refuted." I think it's time democracy tries for the same standards as bridge tournaments. Not accepting non-uniformity would be a good start.
Slashdot Mirror
(Davej is a long time kernel hacker and currently the Fedora kernel maintainer.)
He actually agrees with all of them.
The only thing he disagrees with is whether this is caused by civilization.
Now, in the last IPCC report (THE source for the scientific consensus, at least of the one attacked by Lindzen in this article), it is said that human activities are "likely" the cause of global warming. Not "certainly", but "likely". Even on an earth that is getting hot, I can't see how this is suppressing the view that humans may not be the cause of global warming.
In fact, the only thing he is attacking with any substance are the casual, frequently overheard claims that the recent increase in storms, tornados etc. is caused by global warming. Well, that's a straw-man as far as I am concerned. I have never heard this a scientific claim, just as an informal "could well be" answer e.g. by meteorologist in reply to question by journalists.
This decision isn't final yet, the parents will most likely appeal.
The crucial argument in the court's decision seems to have been that the personal rights of the parents were not violated, since they could not be identified by their last name. This is actually disputable, their name is pretty unique in Germany. (A search in the phone directory didn't turn up anyone with the name.)
The court did not consider the mentioning of the name a violation of Tron's own personal rights.
("Personal rights" is my translation by me of "Persoenlichkeitsrechte", which is technical term in German law speak. Maybe "Right to personal privacy" would be a better translation.)
Just because you CAN mention the name, this doesn't mean you have to.
Wow, you have worries. If once in a while the summary is both correct and does have something to do with tfa, I rather feel like it's time to celebrate...
Well, the problem is that you have to store it for some 10,000 years. That's 2500 warehouses of pretty dangerous stuff, that you have to protect for a very long time. Protect it from criminals, terrorists, natural disasters. Again for 10,000 years!
And that's only the dangers we think of at the moment. Are you really so sure we will have a stable enough government for 10,000 years to come to guarantee just the basic protection of the waste storage sites?
It is beyond me to estimate the dangers of running a nuclear power plant, whether it is worth the risk. But the nuclear waste problem is what makes me want to get rid of nuclear power.
(But then, I am from Germany, probably the country most critical of nuclear power all over Europe.)
...as funny, because I can't take the thought that he might be serious.
Can't be true. Even Mountain Dew doesn't have as high a caffeine concentration as a programmer after 15 years of work.
What the heck, I for sure don't trust wikipedia much, but still a lot more than some popular news sites.
His opinion has a lot of annoying consequences, for example we are not allowed to provide Windows binaries for download on our project home page. (Our project is very portable, and we maintain MSVC project files.) Instead, people have to google for a Windows binary of our program instead, or look around on some other websites for a useful link. I would so much love to make it really easy for Windows Users to play around with our program; in its niche it could actually get a lot of people in touch with open source.
In fact, several of our current developers trying things out in a purely windows environment, and have then switched to Linux when they realized how much better the development environment in any Linux distribution is, compared to anything you can get for free on Windows. One of them had not even known about open source before he got started.
How do you think a BIOS is typically developed:
- Reading the ACPI spec, implementing it, followed by extensive unit testing verifying that the spec was implemented correctly.
- Browsing the ACPI spec, implementing it, then trying to boot windows until it works.
If you write the Linus ACPI code according to the spec, you are betting that the first point above is reality. Not the kind of bet you would like to make.Of course, this example works as well with s/ACPI/HTML/; s/boot windows/load the page in IE/ etc. etc.
People here in Germany are constantly worried that the research in Germany universities might not be able to compete with the top places in the world anymore. This finally proves them all sooo wrong!
Nobody knows but I am not worried. The Linux kernel community has always been great in adopting to new circumstances. Alan Cox decided to drop kernel work for a year to do a MBA? No problem, his role got taken over by a couple of people. Linus has to decide to drop BK? No big problem, Linus start writing GIT, which is quickly taken over by other people, and after 2 weeks development continues almost as if nothing happened. Dave Jones gets overworked pushing bug fixes from 2.4 to 2.6? No problem, maintainers step in and push the stuff themselves.
I have a lot of trust in the robustness of the LK development.
Of course, development would suffer without Andrew Morton. His skill in managing the process is just impressive. But it would not break the process.
Short answer: no.
Long answer: Of course, automatic testing has its role, and the OSDL and LTP guys have been doing this regularly. It is great to monitor for any kind of regressions, including performance-wise.
But most kernel bugs are new, and it is pretty tough to set up a kernel test that intelligently discovers new bugs. Maybe even more importantly, most kernel bugs are pretty unique. They happen when you start a weird program foobar on a weird system with a fooznoo gadget. No way you can test for this with an automatic setup. The usage scenarios of a Linux kernel in the real world have such a huge variety on such a diversity of hardware that nothing can beat the testing by us suckers who download the latest kernel and, guess what, use it.
The main alternatives are GnuCash and KMyMoney.
Funny that you mention this, I also do training for the IMO team in my country.
Dr. Wildberger has an idea that he thinks will make trigonometry more intuitive and I hope he is really onto something here.
I think a more appropriate characterization would be the following: Wildberger is showing how some computations can avoid angles entirely (in particular, when you don't have to add angles!), and are more intuitive when you avoid this and thus trigonometry. He puts them into somewhat nice framework. I agree it may sometimes lead to more elegant (but essential equivalent) solution to cute little geometry problems.
However, this doesn't go well with the spin he is giving this himself: "For the past two thousand years we have relied on the false assumptions that distance is the best way to measure the separation of two points, and that angle is the best way to measure the separation of two lines." Sorry, but distance is the best way to measure separation of points.
Or further down: "So teachers have resigned themselves to teaching students about circles and pi and complicated trigonometric functions that relate circular arc lengths to x and y projections - all in order to analyse triangles." What a non-sense. Teacher don't teach students trigonometry in order to analyse triangles. They teach them trigonometry because they are some of the most fundamental functions in mathematics, popping up all over in physics and engineering (and because they cannot teach them the complex exponential function, which does the same more elegantly). And analysing triangles is a cute way of doing something with them.
Here is what he does: He replaces the distances by its square, and calls it a squandrance. He replaces the angle by the square of its sine and calls it a spread.
Ok, the relations between the squandrances and spreads of a rectangular triangle are simpler than those between its sides and its angles -- they are just simply obtained from pythagoras' theorem.
However, two way more fundamental relations suddenly become horribly difficult: Say going from town A to C I go from A to B then from B to C. A to B is 5 squandrance, B to C is 3 squandrance, how much squandrance is it from A to C? No, not 8, but 5 + 3 + 2*sqrt(8). The simple addition of distances becomes a square-root function...
Also, say I turn left by 30 degree, and then do that again. Guess what, I turned left by 2*30 = 60 degrees. However, if you were doing this in spreads, you don't want me to tell you the answer. I think there is a reason why sailors have been using angles and not spreads...
As for engineers: Why do they have to learn trigonometry? Not for geometry, you can just do that with coordinates. But because it turns up in waves everywhere, which in turn turn up everywhere. (Aside: the mathematical reason is that sine and cosine are the solutions of one of the most fundamental differential equations.) For example, the voltage of an AC current will depend proportionally on a sine fucntion. The angle you substitute in sine will be proportional to time. The spread you would have to substitute in a rational geometry function instead is NOT proportional to time, but to the square of the sine of time. Too bad my stopwatch measures time in seconds and not in the square of its sine...
Will his approach lead to faster computations? Of course not. Whereever it would help, people of course already knew last year already (and maybe even, gosh, in 2003) to use some trigonometric substitution....
Now I finally understand why cell phones are banned in the immigration area at least at Detroit Metropolitan Airport, it has never made sense to me before...
And in fact I have a lot of respect for Linus for dropping BK now right on the spot, when the problems it created were becoming too big -- after he had been unimpressed by 3 years of anti-BK flaming...
Are there any PC vendors planning to ship Windows N (with their own media player instead)?
1. He wrote scripts so that he didn't have to jump between applying patches and reading e-mail, instead he is now reading a batch of patches, queuing them, and then starts a script to apply them.
2. Developers have instant access to Linus' tree. Any source control system would have provided this.
3. The comments to the patches in the e-mails sent to Linus now actually make it to the public. Just about any GNU project does this via ChangeLog under any revision control system.
4. A script was written to automatically extract release notes from the changelog comments.
5. Merging with subsystems maintainer is easier if they pile up the patches in bitkeeper repositories.
Maybe all of the above together yielded a factor of two. But only with respect to 5. is BK at all relevant. And even there -- by a HUGE amount the largest merge point is Andrew Morton, who uses quilt instead of BK to manage his tree with some hundred patches per week throughput to Linus. And I haven't read any complaints from Linus that he isn't using BK.
This sounds like the best solution in the best interest of actual users.
There is always room for improvement, new ideas, new architectures, hardware, etc that open up new pathways to more flexible or secure operating system organization.
What you are missing that the linux kernel development process has matured quite a lot. Now there is a steady stream of new features into 2.6. There is no backlog of huge patches that introduce new features and are available only in vendor kernels.
I wouldn't bet a lot of money on it, but I wouldn't be surprised if the current kernel in 5 years was 2.6.xx (which still would look completely different to current 2.6.11).
Still, this means that the security community has to always play catch up. Wouldn't it be much better if Linus dropped an e-mail to vendor-sec 2 days before he applies a security patch to the kernel, so that vendors can release a new kernel quickly, and users of a mainline kernel will be informed with an announcement immediately once the patch gets public?
Maybe I shouldn't get involved in this inner-american debate, but I just think stronlgy this is non-sense, and I will try to explain why.
Fraud, intimidation etc. are only the worst of irregularities that can occur. I think you should set your standard much higher than that. I find any sign of non-uniformity of the process highly disturbing. Having to queue for 4 hours to vote in some areas, 1 hours somewhere else, and 5 min at other places is just unacceptable. If you start accepting this, how will you stop an election official from purposely assigning few resources to neighborhoods traditionally voting for the other side, thus inducing longer lines there that will certainly turn some people away?
Similarly, some 100 lost votes due to incorrect voting machine usage -- never mind these votes would not have changed the outcome, this is a just a disaster. I am sure noone has done this on purpose, but just the possibility of this allegation is a loss for the credibility of democracy.
In the game of bridge, there is a nice law (yeah, bridge has lots of rules, and they call them laws) about organizing tournaments: when pre-dealt hands are used (which makes sense for many reasons), there is the danger that anyone seeing the hands in advance could easily win the tournament. That law says: "The pre-dealing of hands should be organized in such a manner that any allegation of fraud can be convincingly refuted." I think it's time democracy tries for the same standards as bridge tournaments. Not accepting non-uniformity would be a good start.