I was talking in general. Sure there are a couple of good ones, that hire psychologists to match people better, that verify identities and so on, but most of them are not like that.
My parents have gone to one of those stupid "learn how to make money on the web" seminars. One of the "business models" presented was a dating site -- that is why there are so many of them out there. When someone hears about a "dating site" most likely they will think "scam".
I have heard of plenty of people who left their wives/husbands and kids for some "perfect soulmate" they met online. After a while that "someone" doesn't turn out to be that great and they are back on the net looking for the next "perfect soulmate".
I think people project their own idealized image of a soulmate unto the other individual they meet online. It is very easy to do when you only see their typing and a picture of them 10 years (and 50 lb.) ago. I am not saying people don't do it in person too, it's just that it is much easier online.
The free market is not a natural law. Unlike a natural law, the market will change depending on how we (people, economic agents) describe it. So by calling it a "natural law" _we_ let it function (and we act) as if it is a "natural law". One could probably say that it is like the effect of "observing" in the quantum world (the process of observing a system will change its state).
It is called "people wanting to be victims and wanting to be babied by the politicians."
A bunch of idiots got fooled by another bunch of (slightly smarter) idiots with a website, that they can find their "soul mates" and consequently a "happy married life" for a $5.99 membership fee. In the end they found out that the "hot local babes" are just pictures from pr0n sites and it was actually the employees of the website who replied back to them. This made them realize that their own little world where they are soooo good looking, hot and desirable by the opposite sex is just a silly fantasy, so they called their representative to do something about it (just like when they used to run to mommy when they got their knee bruised)...
Call me old fashioned, but to me "internet dating" sounds like an oxymoron anyway...
So who is the legislation supposed to protect? People from themselves?...Because they don't realize that "find your perfect soulmate for $39.99 a month" or "find a hot chick to bang tonight" are too good to be true. Let people live in their fantasy world where everyone has a soulmate and there are plenty of hot local babes" willing to please, just a few mouse click away and $5.99 registration fee.
The analogy was stupid in the first place. There is blurry line between "I am watching my neighbour" and "she is showing herself to me". For example if she walked around naked by her open window, in full light and plain view of my children while they are playing in the yard, she would probably be cited for exposure. Anyone stupid enough to not close the blinds/curtains when they are naked -- don't have much to complain.
So I agree with one of grand(grand?)parents that if the RF radiation is allowed to pass through my body, I should be able to decode it. Security should be based on encryption not on the hope that nobody will _probably_ listen in because we made the receivers illegal...
Broadcasting un-encrypted stuff is like shouting your private conversation from the top of a hill for the whole town to here. Now if you used a foreign language that nobody except you and intended recepients speak (think Navaho!) then you should be fine.
Yeah Ubuntu is smooth as butter. Have been using it since Hoary and never reinstalled or had problems with it. To upgrade I just always did "apt-get dist-upgrade"
Welcome to the RIAA/MPAA world domination extravaganza. It just goes to prove that those who have the most money and more lawyers will win after all, no mater what the law says...
Microsoft willing to give away free licences for the govt. computers in exchange for the piratebay sMicrosoft willing to give away free licences for the govt. computers in exchange for the piratebay shutdown would do it, for example...
With any legal system there are a million of loopholes, that his how the lawyers make their big bucks. It seems like one of those MPAA/RIAA/Microsoft/Adobe lawyers found a loophole in the Swedish law after all.
It seems the like the guys at the piratebay.org has fun with the legal threats, insulting all those idiots, I wouldn't be surprised that a good number of them took it personally, knowing how big and inflamed their egos are. Does it mean the bad guys win after all?
The the difference in my post between "mental" and "physical" was that "mental"=there is no actual skin pathalogy. i.e. no mites, bacteria, fungi or anything else, just a halluciantion or hypersensitivity or whatever else it might be. The treatment in this case would probably be antipsychotic drugs (thorazine and "friends"). The "physical"=there is something abnormal going on with the skin, some mites, fungi, bacteria etc. In this case the course of treatment would be antibiotics, ointments and so on.
Then I suggested that it could be both. Say someone does have an itchy skin or sensitive hair follicules or just bed bugs that bit them. Most people would be discomforted, they would call bug control, buy some lotion and they would probably be "ok". Some people though get severly traumatized by the idea of bugs crawling on/through their skin, to them there is nothing worse. Under such sever mental anguish it is possible that psychotic episodes would emerge, they would see "things" coming out of the skin, fibers of different colors, they pull their hair and dig their nails into the skin trying to get the bugs out. In the end they would end up with a really damaged skin which would even further affirm to them that there _is_ something wrong with their skin and then the whole thing repeats.
Some of people had the fibers that grow from their skin analyzed. They are cellulose but do not come from clothing. My hunch is that it is probably both mental and physical. There is something going on -- perhaps various species of scabies, skin mites (demodex foliculorum) and "friends", perhaps even some bacteria or fungi. Sometimes perhaps it has to be two of the factors at the same time.
.
As for the mental part of the disease, it seems that humans over the thousands and thousands of years have developed a basic disgust and revulsion to small crawling things on the skin -- spiders, lice, bed bugs, centipedes, worms, scorpions and so on. There is a good reason for that, those things are associated with disease, poison bites, and un-cleanliness. On average, people would probably be less afraid of a wolf than of one of those creepy-crawly things. That is why it is not surprise that a good percentage of these cases are mental.
Mind you doctors still don't know that much about the human skin. There is no cure for rosacea -- some think it is the demodex mite that causes it, some think it is a bacterial infection, some suspect it is just genetic. Some antibiotics have been shown to work, sometimes lasers help too, but nothing definite. A lot of guess work. Doctors are not gods, they only know what other doctors may have published in a journal or by doing research themselves (rarely happens). So just because they haven't been able to find anything doesn't mean it is not there.
Quite interesting. I wonder why Merk and friends did not already do a more extensive research into this?
The idea per se is not that revolutionary, phages have been known and used for sesearch purposes for a long time. I suspect there is a certain fear of injecting a virus (even one that supposedly hasn't been able to infect eukariotic cell) into a human patient. It is also concievable that bacteria could acquire resistance to phages. There are some bactera that cannot be infected with any known phages.
The Wiki Article you pointed to, mentions how this therapy was used by the Soviets for many years. This makes it sound exotic and new, but from a personal point of view, growing up in the former Soviet Union I can say that there wasn't much emphasis on human life and safety. Everything was owned and operated by the state and all the research was tied somehow with politics, which probably explains why the West takes Soviet-era research with a grain of salt. There could have been a lower success rate than reported or some results could have been fudged up more so than here in the West. As it happened with genetics, the Soviets often would persue anything other than the West is doing, just for sake of being different -- see Lysenko's "vernalisation" for example.
MRSA is already here and is bad enough, but there is already fear of the vancomycin resistant staph. Vancomycin, as it is clear from your story, is a last resort antibiotic, when all others have failed. There is evidence that there could be a super-super-bug that would be resistant to vancomycin as well. The common mechanism of action of these antibiotics is to provent the assembly of the cell wall. It so happens that only prokariotic cells (which staph. aureus is) have this external cell wall structure to prevent them from "exploding" due to high internal osmotic pressure. So this cell wall has been and is a good target for antiobiotics.
It is interesting how most of the antibiotics -- this new one and including the first one -- penicillin, are sythesized and produced by fungi. There is a constant battle for nutrition and space between the bacteria and the fungi -- some kind of an evolutionary yin and yang. One will always try to overtake the other and will develop new mechanisms for resistance or attack.
The way I see it, Ubuntu is Debian+. Yes it does pull most of its packages from Debian but there are also many developers who work on Ubuntu packages only, including the server components. (Besides the server components of Ubuntu and the desktop version components are not completely orthogonal to each other, and also when it comes down to it, they don't differ as much between the various distributions.)
I think Sun liked Ubuntu exactly because it has a good, clean, functional user interface. Sun desktop or CDE I think are years behind. So instead of fixing them, why not adopt an existing distro with a nice clean gui?
Overall, the benefit of Ubuntu is that there is an additional group of developers, users and fans who help test and fix stuff. Most of the fixes that Ubuntu group povides are minor annoyances kind of bugs, (handling of French keyboards, not enough comments in the samba conf file , an icon is missing etc.) but when all of them are added together they make a better, more integrated and polished OS. That is what made me switch to Ubuntu -- it just works. I have used RedHat, Fedora, Mandriva, even Gentoo but finally settled on Ubuntu for the last 1.5 years.
I call my parents in Europe quite often and always keep an eye out for a good deal on international calling. I have been looking at SkypeOut rates ever since it was introduced, it is still 2x more expensive than a good phone card.
But that is a management and HR problem. They need to screen for people that have the right skill level _and_ can work with others. That is probably true in any work environment, not just software.
Add drivers to your microkernel and you'll end up around the same number of lines if not more.
The argument wasn't that somehow a microkernel architecture will magically provide the same basic functionality with just 10K lines of code where Linux would need 6M. The point of the microkernel is that it compartmentalizes each driver and the kernel into their own address spaces with strictly enforced boundaries. In other words, if you have 6M lines of code and any of those lines could mess up the whole runtime address space of the other 6M lines -- it would be impossible to prove correctness (I am not arguing whether proving correctness is a good thing or not, it is done with mission critical systems so we'll assume it is necessary for now). The reason is because the possible state space that needs to be explored would increase exponentially with a number of lines of code. Just checking the correctness of a 8K lines of code microkernel could take months and months _BUT_ at the same time on a different computer you can check the correctness of the drivers in parallel it would still take about the same time if you have as many computers as you have drivers. This cannot be done when all the drivers share the same address space.
The point here is that I don't really buy Linus' arguments against microkernels. I do understand that he will never change his mind, that would mean pretty much re-writing the kernel from scratch. It is like Microsoft tomorrow deciding that open sourcing Windows and giving it away for free would be a good idea -- just not going to happen.
So why do the makers of real-time OSes spend months to get their products certified for covert channel analysis. How silly of them to do that, they should really just forget it because cosmic rays will come down and burn the transistors anyway, right?
Yes I did prove the sensor driver's correctness at the same time I proved the microkernel's correctness. It took about 4 months to do it, but now it is done. The reason I could do it is because the sensor's driver is in a separate address space than the kernel driver. There an exponential relationship between the increase in the number of lines of code in a module and the ability to prove its correctness [see the SAT problem].
Even if the sensor failed and its driver is un-usable, the microkernel _will_ keep the system in a pre-determined 'fail' state from which it start the alarm or activate the shutdown.
Now let's see what happens if your reactor is controlled by Linux 2.6.15.11 and there is an off-by-one error in some serial driver used to communicate with the sensor. Over time the driver will overwrite some data structure and your _whole_ system is frozen. You are in Soviet Russia and you reactor design is an RBMK with a positive void coeficient. The system has actually frozen during your attempted shutdown, your boron/graphite control rods are stuck 1/3 way through while your actuators and everything is frozen because of the stupid off-by-one error in the serial driver. Of course you wanted to prove that the Linux 2.6.15.11 would behave correctly with all of its drivers but would take a longer time than the age of the universe, so now I am sure the people living by _your_ reactor won't mind the radiation.
//sorry I could have said all this in just one sentence but was tempted by your example so I expanded on it;)
The important thing is that in the microkernel architecture the 10k microkernel lines of code will be in a separate address space. That is all that is needed. If those lines are mathematically provable correct, they they could take the system to a predetermined state in case of any of the drivers failing, it could reload the driver and try again or contact the admin and so on -- not freeze. When talking about the lines of code, we talk about the smaller chunk of code that can be verified that it works correctly. Yes, the most widely used microkernels today are around 10K lines. Even so it could take years to prove their correctness.
In case of the microkernel it would probably be a linear relationship. So, for example a mouse driver would have to do 30 context switches a second in a microkernel architecture and it won't have to do that in a regular Linux kernel. A context switch would linearly (not exponentially) increase the demand for resources. The bottom line is that microkernel s are already used in real-time OSes and other mission critical appliations, it is a work-able architecture and not just a lab toy. Obviously there is not horrible exponential increase in time or space requirements.
Yes, most of those 6M lines are drivers but they share the same address space as the whole kernel. So one cannot prove that the 10K lines of the kernel are correct and be done with it (as you would in a microkernel) because any one of those 6M lines could write or read anything in the kernel space during run-time. So a little known and tested mouse driver could freeze the system. You can only hope the system that just froze wasn't doing any mission critical tasks that would be responsible for human lives and such.
I was talking about viruses and Quake to exemplify the extreme tendency of home desktop consumer to completely overlook the security aspect. But whatever kernel linux would be running it would be running on a box that games would be played at home but it would also supposedly might be used in mission critical situations. As the grandparent pointed out, there is a tendency for servers to become more "mission critical". So then the question becomes, who should Linus try to satifsy? Most would say -- the home users of course, but I don't agree. The speed argument, which was the initial argument against microkernels back in the day, was valid. A 33MHz machine couldn't not cope with a mouse driver that would need at least 30 context switches per second in microkernel architecture, but today a 3GHz Pentium IV might cope with it fine. I think if people were properly educated and would be explained clearly that they would be getting a more secure system, they would be willing to way another second for the browser to open. Next year, they'll upgrade and they'll be running faster anyway.
The "hard to code" argument is partially valid, but I think the developers are not "stupid" and could cope with it. Linus goes a long way to show that microkernel are not easier to _develop_ -- I don't think that was ever the case though. Micro- doesn't mean simple.
Slashdot Mirror
My parents have gone to one of those stupid "learn how to make money on the web" seminars. One of the "business models" presented was a dating site -- that is why there are so many of them out there. When someone hears about a "dating site" most likely they will think "scam".
I don't go to any dating site, that's the point. The $5.99 was just a number, it might as well have been $41.43...
I think people project their own idealized image of a soulmate unto the other individual they meet online. It is very easy to do when you only see their typing and a picture of them 10 years (and 50 lb.) ago. I am not saying people don't do it in person too, it's just that it is much easier online.
The free market is not a natural law. Unlike a natural law, the market will change depending on how we (people, economic agents) describe it. So by calling it a "natural law" _we_ let it function (and we act) as if it is a "natural law". One could probably say that it is like the effect of "observing" in the quantum world (the process of observing a system will change its state).
common sense is quite a valuable tool, it is free and available to all, just have to reach for it and use it - simple as that!
A bunch of idiots got fooled by another bunch of (slightly smarter) idiots with a website, that they can find their "soul mates" and consequently a "happy married life" for a $5.99 membership fee. In the end they found out that the "hot local babes" are just pictures from pr0n sites and it was actually the employees of the website who replied back to them. This made them realize that their own little world where they are soooo good looking, hot and desirable by the opposite sex is just a silly fantasy, so they called their representative to do something about it (just like when they used to run to mommy when they got their knee bruised)...
So who is the legislation supposed to protect? People from themselves? ...Because they don't realize that "find your perfect soulmate for $39.99 a month" or "find a hot chick to bang tonight" are too good to be true. Let people live in their fantasy world where everyone has a soulmate and there are plenty of hot local babes" willing to please, just a few mouse click away and $5.99 registration fee.
So I agree with one of grand(grand?)parents that if the RF radiation is allowed to pass through my body, I should be able to decode it. Security should be based on encryption not on the hope that nobody will _probably_ listen in because we made the receivers illegal...
Broadcasting un-encrypted stuff is like shouting your private conversation from the top of a hill for the whole town to here. Now if you used a foreign language that nobody except you and intended recepients speak (think Navaho!) then you should be fine.
Yeah Ubuntu is smooth as butter. Have been using it since Hoary and never reinstalled or had problems with it. To upgrade I just always did "apt-get dist-upgrade"
Welcome to the RIAA/MPAA world domination extravaganza. It just goes to prove that those who have the most money and more lawyers will win after all, no mater what the law says...
With any legal system there are a million of loopholes, that his how the lawyers make their big bucks. It seems like one of those MPAA/RIAA/Microsoft/Adobe lawyers found a loophole in the Swedish law after all.
It seems the like the guys at the piratebay.org has fun with the legal threats, insulting all those idiots, I wouldn't be surprised that a good number of them took it personally, knowing how big and inflamed their egos are. Does it mean the bad guys win after all?
Then I suggested that it could be both. Say someone does have an itchy skin or sensitive hair follicules or just bed bugs that bit them. Most people would be discomforted, they would call bug control, buy some lotion and they would probably be "ok". Some people though get severly traumatized by the idea of bugs crawling on/through their skin, to them there is nothing worse. Under such sever mental anguish it is possible that psychotic episodes would emerge, they would see "things" coming out of the skin, fibers of different colors, they pull their hair and dig their nails into the skin trying to get the bugs out. In the end they would end up with a really damaged skin which would even further affirm to them that there _is_ something wrong with their skin and then the whole thing repeats.
. As for the mental part of the disease, it seems that humans over the thousands and thousands of years have developed a basic disgust and revulsion to small crawling things on the skin -- spiders, lice, bed bugs, centipedes, worms, scorpions and so on. There is a good reason for that, those things are associated with disease, poison bites, and un-cleanliness. On average, people would probably be less afraid of a wolf than of one of those creepy-crawly things. That is why it is not surprise that a good percentage of these cases are mental.
Mind you doctors still don't know that much about the human skin. There is no cure for rosacea -- some think it is the demodex mite that causes it, some think it is a bacterial infection, some suspect it is just genetic. Some antibiotics have been shown to work, sometimes lasers help too, but nothing definite. A lot of guess work. Doctors are not gods, they only know what other doctors may have published in a journal or by doing research themselves (rarely happens). So just because they haven't been able to find anything doesn't mean it is not there.
The idea per se is not that revolutionary, phages have been known and used for sesearch purposes for a long time. I suspect there is a certain fear of injecting a virus (even one that supposedly hasn't been able to infect eukariotic cell) into a human patient. It is also concievable that bacteria could acquire resistance to phages. There are some bactera that cannot be infected with any known phages.
The Wiki Article you pointed to, mentions how this therapy was used by the Soviets for many years. This makes it sound exotic and new, but from a personal point of view, growing up in the former Soviet Union I can say that there wasn't much emphasis on human life and safety. Everything was owned and operated by the state and all the research was tied somehow with politics, which probably explains why the West takes Soviet-era research with a grain of salt. There could have been a lower success rate than reported or some results could have been fudged up more so than here in the West. As it happened with genetics, the Soviets often would persue anything other than the West is doing, just for sake of being different -- see Lysenko's "vernalisation" for example.
It is interesting how most of the antibiotics -- this new one and including the first one -- penicillin, are sythesized and produced by fungi. There is a constant battle for nutrition and space between the bacteria and the fungi -- some kind of an evolutionary yin and yang. One will always try to overtake the other and will develop new mechanisms for resistance or attack.
I think Sun liked Ubuntu exactly because it has a good, clean, functional user interface. Sun desktop or CDE I think are years behind. So instead of fixing them, why not adopt an existing distro with a nice clean gui?
Overall, the benefit of Ubuntu is that there is an additional group of developers, users and fans who help test and fix stuff. Most of the fixes that Ubuntu group povides are minor annoyances kind of bugs, (handling of French keyboards, not enough comments in the samba conf file , an icon is missing etc.) but when all of them are added together they make a better, more integrated and polished OS. That is what made me switch to Ubuntu -- it just works. I have used RedHat, Fedora, Mandriva, even Gentoo but finally settled on Ubuntu for the last 1.5 years.
I call my parents in Europe quite often and always keep an eye out for a good deal on international calling. I have been looking at SkypeOut rates ever since it was introduced, it is still 2x more expensive than a good phone card.
But that is a management and HR problem. They need to screen for people that have the right skill level _and_ can work with others. That is probably true in any work environment, not just software.
The argument wasn't that somehow a microkernel architecture will magically provide the same basic functionality with just 10K lines of code where Linux would need 6M. The point of the microkernel is that it compartmentalizes each driver and the kernel into their own address spaces with strictly enforced boundaries. In other words, if you have 6M lines of code and any of those lines could mess up the whole runtime address space of the other 6M lines -- it would be impossible to prove correctness (I am not arguing whether proving correctness is a good thing or not, it is done with mission critical systems so we'll assume it is necessary for now). The reason is because the possible state space that needs to be explored would increase exponentially with a number of lines of code. Just checking the correctness of a 8K lines of code microkernel could take months and months _BUT_ at the same time on a different computer you can check the correctness of the drivers in parallel it would still take about the same time if you have as many computers as you have drivers. This cannot be done when all the drivers share the same address space.
The point here is that I don't really buy Linus' arguments against microkernels. I do understand that he will never change his mind, that would mean pretty much re-writing the kernel from scratch. It is like Microsoft tomorrow deciding that open sourcing Windows and giving it away for free would be a good idea -- just not going to happen.
So why do the makers of real-time OSes spend months to get their products certified for covert channel analysis. How silly of them to do that, they should really just forget it because cosmic rays will come down and burn the transistors anyway, right?
Even if the sensor failed and its driver is un-usable, the microkernel _will_ keep the system in a pre-determined 'fail' state from which it start the alarm or activate the shutdown.
Now let's see what happens if your reactor is controlled by Linux 2.6.15.11 and there is an off-by-one error in some serial driver used to communicate with the sensor. Over time the driver will overwrite some data structure and your _whole_ system is frozen. You are in Soviet Russia and you reactor design is an RBMK with a positive void coeficient. The system has actually frozen during your attempted shutdown, your boron/graphite control rods are stuck 1/3 way through while your actuators and everything is frozen because of the stupid off-by-one error in the serial driver. Of course you wanted to prove that the Linux 2.6.15.11 would behave correctly with all of its drivers but would take a longer time than the age of the universe, so now I am sure the people living by _your_ reactor won't mind the radiation.
The important thing is that in the microkernel architecture the 10k microkernel lines of code will be in a separate address space. That is all that is needed. If those lines are mathematically provable correct, they they could take the system to a predetermined state in case of any of the drivers failing, it could reload the driver and try again or contact the admin and so on -- not freeze. When talking about the lines of code, we talk about the smaller chunk of code that can be verified that it works correctly. Yes, the most widely used microkernels today are around 10K lines. Even so it could take years to prove their correctness.
In case of the microkernel it would probably be a linear relationship. So, for example a mouse driver would have to do 30 context switches a second in a microkernel architecture and it won't have to do that in a regular Linux kernel. A context switch would linearly (not exponentially) increase the demand for resources. The bottom line is that microkernel s are already used in real-time OSes and other mission critical appliations, it is a work-able architecture and not just a lab toy. Obviously there is not horrible exponential increase in time or space requirements.
Yes, most of those 6M lines are drivers but they share the same address space as the whole kernel. So one cannot prove that the 10K lines of the kernel are correct and be done with it (as you would in a microkernel) because any one of those 6M lines could write or read anything in the kernel space during run-time. So a little known and tested mouse driver could freeze the system. You can only hope the system that just froze wasn't doing any mission critical tasks that would be responsible for human lives and such.
The "hard to code" argument is partially valid, but I think the developers are not "stupid" and could cope with it. Linus goes a long way to show that microkernel are not easier to _develop_ -- I don't think that was ever the case though. Micro- doesn't mean simple.