Did you read the post at all? I was not talking about Mhz to performance ratio in different pentium and amd models. The point was that there was a _large_ speed increase in processing power since 1992 - that is all you should read into it. If you think there wasn't a _large_ speed increase in the last 14 years, then you don't live in the real world. It might not have been 100x it might have been 90.3x, that is not the point!
In _general_ on the same processor the faster the clock speed the greater computational throughput. If Mhz!=speed, then underclock you machines to 100Mhz, see if you'll notice a difference...
I can run 10Ghz over a copper wire, WHOA that is fast.
Good for you. Whatch the output power and keep it away from the brain and the reproductive organs.
I was just saying that in general, it is easier to find a hole in a huge bloated program than in a smaller program. Just because the Linux kernel is open source doesn't mean there won't be any root kits for it. In fact there have been!. How many time s did you go through the 100,000+ lines of code looking for bugs?
>This is about Windows. Windows is closed-source/proprietary, so "good hackers" won't be able to fix the system.
You are right, I just meant that eventually MS will be embarassed about the new hole in their application and tell some intern to fix.
>>If the hacker manages to exploit a hole in the display driver, the driver will not crash the system.
>How is this good? Without the display driver, how can you see what is going on?
I think it is very good if you are also running something else that cannot crash even if the display driver starts acting crazy. I was thinking also of the server environment. Also the separation kernel can perhaps detect that the display driver is malfunctioning and re-initialize it without restarting the system.
>>I think the present 4Ghz machines can hangle a %10 slowdown at the expense of say, %80, improved security.
>Microkernels do not guarantee security. And where do you get these numbers? I also have barely seen any 4 GHz machines.
By running the device drivers in user space security will be improved as the OS can still mentain control of the machine even during the failure of the device driver. A separation kernel is very small, maybe a couple thousand lines long. Compare that to the Linux kernel that is hundreds of thousands + of lines including drivers. I think it is quite obvious that it is much easier to mentain and debug a 1000 lines program than a 500,000 lines program. That is where the improved security comes it.
Many of the Green Hills' OS'es work on this principle, and whenever you board a Boeing airplane, it is probably the Green Hills software that is flying it. As much as I like Linux, I don't think I want the Almighty Penguin to fly my over the Atlantic.
The point of the post was that it is probably time to apply some of separation kernel guidelines to the desktop PCs.
The 4Ghz, 10% and 80% numbers are there just to illustrate the point I didn't do any benchmarks, it was just a guess to show that there will be a _small_ performance hit at a much _greater_ improvement in security.
The point of the post of was that in 1992 the top of the line might have been 60MHz machines, now you can overclock a pentium to 6Ghz, so we are talking about 100x speedup. The bus+memory didn't speed up as much, but still it many times faster. So in 1992 you couldn't do context switches fast enough, today you can and you might not notice much of a difference.
The problem is that people do want performance, but after their box gets rooted and their hard drive erased, they tend to change their priorities. I sure could have lived with a 10% slowdown and play Quacke at 90 fps instead of 100 fps if that meant that I had a more secure OS and my box did get rooted.
In 1992 for Linus, performance was very important. I don't think he anticipated the popularity of Linux and that one day people would be using it on their desktops and they will be connected to the net, and script kiddies will turn them into zombies.
But it is not 1992 anymore and I think it is time to reconsider. Security is becoming more and more important and slowdown can be compensated by just buying a faster CPU next year.
You are right, I was just thinking that a small (couple of thousand lines) open source microkernel will have much less exploits than a kernel that is a million lines together with drivers and everything else that can run in kernel space.
The lesson in this article should be also that there is something wrong with the Windows kernel if there can be written whole books about how to make rootkits for it. The same can go for the Linux kernel. (Yeah that's right, I bashed _the_ penguin on the head, mod me down!)
Kernels are so big and bloated that there is almost %100 chance of there being some exploitable whole in them. If the "good hackers" discover it, it will be patched, if the "bad hackers" discover it, they will make rookits.
A lot of the code that is not tested and buggy is in the drivers, and I don't understand why do current operating systems still have drivers that are run in the kernel instead of in the user space. The machines are fast enough to switch contexts between the display, mouse, sound, disk and communication with the ports. The kernel should be very small and only implement the security policies and handle communications between devices. If the hacker manages to exploit a hole in the display driver, the driver will not crash the system. These are called secure microkernels or separation kernels. I think the present 4Ghz machines can hangle a %10 slowdown at the expense of say, %80, improved security. In 18 months, the speed will double anyway;)
Check out this paper from NIST that talks about this. Also, more general info about it here
I doubt NASA is using one main computer program and they just plug in parameters. They probably have a lot of custom designed software as well as packages that go with commercial programs (Matlab, Mathematica etc.)
They are also calculations of the same orbit but for different purposes. If you just want to make a simplified simulation and use a simple 2 body problem for educational purposes, then you ignore many variables and just use college calculus to solve it. If you calculate the same orbit but for a possible distant future, then you account for more variables, use more complicated math, and probably a more complex program (algorithm). Then you have the calculations for an actual launch, where you would account for a lot more variables, these would be a lot more complicated and will be very complex.
Now that is just speculation, I have never worked at JPL. If I get close enough with my dark skin and thick accent I'll probably be shot on sight..., at least that is what my friend from Caltech told me who works there...
Nuclear might not be exaclty cheap but it is cheap enough for the French to build it and even export. Germany, where environmental freaks lobbied against nuclear power plants years ago, now import a lot of their power from France. I think storing nuclear waste in a mountain in Nevada is worth cutting down on the emissions and also on dependency on foreign oil, if according to many it slows down the melting of the ice caps - even better.
If they had hybrids that can store more electrical energy, and they can just be charged while they sit in the garage all night and be good for the next morning, I think that will be a 'good thing'.
At the same time, the warranty policy might be a rectroactive measure after an internal report is issued that the new drives have high failure rates and perhaps it leaks to the web. Then the company might increase the warranty to continue selling the drives. Most people that have warranties probably won't have time to go through the process of getting a replacement in case of failure. They'll probably mourn their lost data for a while then quickly run to their favorite computer store to get a _another_ brand. The warranty, I think, also gives the consumer that "warm and fuzzy" feeling of security and it generates tons of $$$ for places like Best Buy were they'll badger you until you get the "extended warranty".
Unless the drives were SCSI and were meant to be "hot swapped" you shouldn't do it. I "hot swapped" an Athlon CPU before, I was testing a motherboard and forgot to turn the power off, I removed the CPU and I think that killed it.
Good point. My 3 year old IBM 60GXP Deskstar ("Deathstar") just died taking with it lots of data, digitized movies, music and tons of installed applicatoin. Not all have been backed-up. When I bought the drive I thought that IBM makes reliable hardware and I would pay more for quality. As it turned out, the Deskstar series, have been having problems and are known to fail now. IBM dumped their hard drive unit faster than a hot potato. Not sure if the failure of the Deathstars is to blame.
Anyway my point is that Maxtor is not known to have drives of the highest quality. Their drives are often carried by Circuit City, Best Buy at lower price than WD, Seagate or others. The problem is that you don't know when you buy a new unit if it is plagued by quality issues, only a couple of years later you might find many others on the net complaining about sudden failures.
Also I am not sure if warranty is of much help. Sure, if I had a 5 year warranty, I would get a new drive from IBM. But hell, do I want another drive from them? Besides it is not the drive itself that is of much value (to me at least) but the data on it, and IBM can't to much about my lost project data.
If you have a RAID-1 array and a server that has to be on all the time. Then you remove the "bad" hard drive and insert a "good" one while the system is on. An internal drive can be hot-swapped if it is put in a special craddle with a handle and a lock. This has been the case with SCSI drives for a long time.
Well, you can tell them appart. When they waddle they swing their hips more and they flap their wingtips everytime they say "fabulous" (with a lisp of course).
No I have not watched TV, I was working on a project.
Yes, word of mouth, my mom called me and told me about the movie. After I saw it I told some of my friends who went and saw it and I think they liked it enough to recomment to thier friends and so on. So even though someone (let's say my parents in this case) saw an add on TV for it, I didn't and I don't know if all my friends did so that makes 'word of mouth' I think.
It is alien, yet people see themselves (or rather human ideals) in the penguins. Penguins overcome hardships to raise their children - that makes them like humans and humans identify with them. Penguins are monogamous (at least for the period of one mating season) - which is an ideal that many humans today don't live up to, again people identify with the penguins. Penguins are social creatures their behavior sometimes projects this "social justice" that any American likes, for example when one mother who lost her chick, tries to steal the chick from another penguin, all the females around came to protect the chick and to fight off the "thief".
A good fantasy or sci-fi, or any story about alien places and creatures is successful (=appeals to the audience) only if they emobody human ideals in them like justice, honesty, self-sacrifice, love, beauty, overcoming adversity and other such things. In other words if you had a movie about worms that live at the bottom of the ocean, or even some alien bacteria (or just mattrasses that sit around ) from Mars or say Titan, you couldn't entice the audience as much.
I agree. I saw it with my wife and we both liked it very much. I think there are a couple of components here that make it such a good movie.
[1] It is G rated. I guess some people are sick of car chases, boobies and scary monsters - go figure! The "naughtiest" part is penguins doing "it" to make more penguins. I actually saw a parent take their child out of the theatre for this one. Felt sorry for the poor kid...
[2] Penguins are somewhat similar to humans in the way they walk and behave. They walk upright but wobble and thus they look like "cute", "fat", "fuzzy" people. You couldn't pull this one off with snakes or, tigers. They are social creatures. In the movie ( I hope I don't spoil it for anyone;) a mother penguin who lost her egg, tries to steal the chick from another penguin. The other females in the group would not "approve" of such behavior and came to protect the chick and the mother from the "thief".
Also penguins are monogamous (emperor penguins are monogamous at least for duration of one year) - which often is not the case with many humans nowadays - not that there is anything wrong with it... So that also anthropomorphosizes them even more.
[3] The bravery and determination of the people who shot the movie is impressive. Very cold weather, very dangerous, all just to film the cute little birds. And, of course, as some post mentioned, some like Morgan Freeman.
[4] It spread mostly by word of mouth. This is similar to the "My Big Fat Greek Wedding". A small movie that made big $ because everyone told their friends to go see and how great it was. I heard about the movie from my parents then after I saw it, I recommended it to all my friends and they saw it. If everyone who see it does it - it is quite a few people..
Dear AC,
Your post is implying that stereotypes are bad. You list all the stereotypes then you say that there is nothing interesting there. But you never really prove or show that stereotypes are bad. It is as if I said "your post is full of adverbs, and that is why it should not be moderated too high".
You know I thought of that too. Back in 2000 I watched the commercials for Super Bowl (I didn't watch the whole super bowl per se) and there were all these ads for online companies and I thought, the chance of Joe Sixpack in 2000 wanting to go online and buy dog food from Pets.com is very low. Back then there weren't as many computers in homes yet, so it seemed silly to have so many ads for online companies.
But every investor had dollar signs flashing in their eyes and dreams of making billions and be the next Bill Gates.
I personally think many looked at Microsoft, and Intel and thought "Damn,if knew back then and bought the stock... well I learned my lession, I better go and invest now in all this technology"
Another warning sign was when I found out this guy with 2 years of technical college and almost no experience was getting $80/hour to make simple websites for companies. That couldn't last forever. A lot of those people are still out there crying that they are out of a job and that nobody wants them. They still don't understand that their skills are not worth that much. In no other field could you go to a technical school for 2 years then make $80/hour right way. It is amazing how so many were fooled.
Sorry for sounding bitter. Surprizingly, I am not an American, I came here from the former Soviet Union precisely to do science and made this my home. So if the Americans where that good at science I wouldn't be here;)
But now that I've lived here, I want my children to be able to study science and be proud of that and not have to always look up at some jocks who can't count and read or write yet he'll somehow become the manager of the of the future businesses.
What concerns me the most is the lack of balance. The problem as I see it, is that it is fundamentaly easier to study business, marketing, law and economics than to study science. In other words a computer geek, won't have a problem in a business class. Now take an average business student and try teaching him computer science, I think it will be much harder. So I think eventually after the countries like India and China will end up getting far enough ahead in terms of science, it won't take them much to figure out the business aspect too. (A side note, up until recently at P&G they were hiring scientists from Easern European countries like Romania and such, now they hire accountants and marketing specialists also, those positions are classical 'American' jobs). The same thing with Chinese, after getting a PhD at an American university, it won't take that much effort for them to get an MBA. Now try sending an american with an MBA to get a PhD in a science related field - no chance.
Also note the trends in outsourcing. Initially it was just basic, scripted customer support, then software development, then they exported the design of the software, now some companies export their accounting, marketing and other areas like that. Pretty soon the'll export everyone except the CEO. Eventually, the people overseas will get rid of the CEO too, and buy the company and run it. Then we'll become the Mexico and China. And then everything in China will be 'made in U.S.A' by people who can't compete they slacked off in school, played sports, went to band camp instead of learning how to read, write and count.
Well that was the point of my post. Neither Nextel (except for a new Motorola phone i850 that some coment mentioned) nor Sprint's 'walkie-talkies' work without a tower. So if you are in Montana in the woods, you still need to get real 'walkie-talkies' (also watch for the bears!).
But if you are close enough to the towers, why not just call the other person? With voice activated dialing, it is not much slower then 'push-to-talk'
At the same time in China kids are learning English and studying Calculus while here they are still watching cartoons, play football and video games. By the time American children graduate from highschool they will still need to take a couple of remedial university courses to finish learning to read and write, while the Chinese students are already studying biochemistry and quantum physics. Well, ok, maybe I exagerated, but you get the point.
Anyone who seems to be interested in science in this country was and still is a "nerd" and thus unpopular and a social outcast. Everyone wants to be friends with the athletic football jocks, the nerds and geeks are the ones who get picked on.
The only way kids are encouraged to be succesful (read=make tons of $$$) by the society (media, family, friends) is to go to college, join a fraternity, party 4 years while taking some business classes then join daddy's or uncle's company with a $80,000 starting salary. Well, that seems to be working so far but for how long?
So yeah, glamorizing science is a good step in the right direction, but I wonder if it too late.
Slashdot Mirror
Talk to this guy. He'll hook you up ;)
In _general_ on the same processor the faster the clock speed the greater computational throughput. If Mhz!=speed, then underclock you machines to 100Mhz, see if you'll notice a difference...
I can run 10Ghz over a copper wire, WHOA that is fast.
Good for you. Whatch the output power and keep it away from the brain and the reproductive organs.
I was just saying that in general, it is easier to find a hole in a huge bloated program than in a smaller program. Just because the Linux kernel is open source doesn't mean there won't be any root kits for it. In fact there have been!. How many time s did you go through the 100,000+ lines of code looking for bugs?
You are right, I just meant that eventually MS will be embarassed about the new hole in their application and tell some intern to fix.
>>If the hacker manages to exploit a hole in the display driver, the driver will not crash the system.
>How is this good? Without the display driver, how can you see what is going on?
I think it is very good if you are also running something else that cannot crash even if the display driver starts acting crazy. I was thinking also of the server environment. Also the separation kernel can perhaps detect that the display driver is malfunctioning and re-initialize it without restarting the system.
>>I think the present 4Ghz machines can hangle a %10 slowdown at the expense of say, %80, improved security.
>Microkernels do not guarantee security. And where do you get these numbers? I also have barely seen any 4 GHz machines.
By running the device drivers in user space security will be improved as the OS can still mentain control of the machine even during the failure of the device driver. A separation kernel is very small, maybe a couple thousand lines long. Compare that to the Linux kernel that is hundreds of thousands + of lines including drivers. I think it is quite obvious that it is much easier to mentain and debug a 1000 lines program than a 500,000 lines program. That is where the improved security comes it.
Many of the Green Hills' OS'es work on this principle, and whenever you board a Boeing airplane, it is probably the Green Hills software that is flying it. As much as I like Linux, I don't think I want the Almighty Penguin to fly my over the Atlantic.
The point of the post was that it is probably time to apply some of separation kernel guidelines to the desktop PCs.
The 4Ghz, 10% and 80% numbers are there just to illustrate the point I didn't do any benchmarks, it was just a guess to show that there will be a _small_ performance hit at a much _greater_ improvement in security.
The problem is that people do want performance, but after their box gets rooted and their hard drive erased, they tend to change their priorities. I sure could have lived with a 10% slowdown and play Quacke at 90 fps instead of 100 fps if that meant that I had a more secure OS and my box did get rooted.
In 1992 for Linus, performance was very important. I don't think he anticipated the popularity of Linux and that one day people would be using it on their desktops and they will be connected to the net, and script kiddies will turn them into zombies.
But it is not 1992 anymore and I think it is time to reconsider. Security is becoming more and more important and slowdown can be compensated by just buying a faster CPU next year.
You are right, I was just thinking that a small (couple of thousand lines) open source microkernel will have much less exploits than a kernel that is a million lines together with drivers and everything else that can run in kernel space.
Kernels are so big and bloated that there is almost %100 chance of there being some exploitable whole in them. If the "good hackers" discover it, it will be patched, if the "bad hackers" discover it, they will make rookits.
A lot of the code that is not tested and buggy is in the drivers, and I don't understand why do current operating systems still have drivers that are run in the kernel instead of in the user space. The machines are fast enough to switch contexts between the display, mouse, sound, disk and communication with the ports. The kernel should be very small and only implement the security policies and handle communications between devices. If the hacker manages to exploit a hole in the display driver, the driver will not crash the system. These are called secure microkernels or separation kernels. I think the present 4Ghz machines can hangle a %10 slowdown at the expense of say, %80, improved security. In 18 months, the speed will double anyway ;)
Check out this paper from NIST that talks about this. Also, more general info about it here
They are also calculations of the same orbit but for different purposes. If you just want to make a simplified simulation and use a simple 2 body problem for educational purposes, then you ignore many variables and just use college calculus to solve it. If you calculate the same orbit but for a possible distant future, then you account for more variables, use more complicated math, and probably a more complex program (algorithm). Then you have the calculations for an actual launch, where you would account for a lot more variables, these would be a lot more complicated and will be very complex.
Now that is just speculation, I have never worked at JPL. If I get close enough with my dark skin and thick accent I'll probably be shot on sight..., at least that is what my friend from Caltech told me who works there...
If they had hybrids that can store more electrical energy, and they can just be charged while they sit in the garage all night and be good for the next morning, I think that will be a 'good thing'.
At the same time, the warranty policy might be a rectroactive measure after an internal report is issued that the new drives have high failure rates and perhaps it leaks to the web. Then the company might increase the warranty to continue selling the drives. Most people that have warranties probably won't have time to go through the process of getting a replacement in case of failure. They'll probably mourn their lost data for a while then quickly run to their favorite computer store to get a _another_ brand. The warranty, I think, also gives the consumer that "warm and fuzzy" feeling of security and it generates tons of $$$ for places like Best Buy were they'll badger you until you get the "extended warranty".
Unless the drives were SCSI and were meant to be "hot swapped" you shouldn't do it. I "hot swapped" an Athlon CPU before, I was testing a motherboard and forgot to turn the power off, I removed the CPU and I think that killed it.
Anyway my point is that Maxtor is not known to have drives of the highest quality. Their drives are often carried by Circuit City, Best Buy at lower price than WD, Seagate or others. The problem is that you don't know when you buy a new unit if it is plagued by quality issues, only a couple of years later you might find many others on the net complaining about sudden failures.
Also I am not sure if warranty is of much help. Sure, if I had a 5 year warranty, I would get a new drive from IBM. But hell, do I want another drive from them? Besides it is not the drive itself that is of much value (to me at least) but the data on it, and IBM can't to much about my lost project data.
If you have a RAID-1 array and a server that has to be on all the time. Then you remove the "bad" hard drive and insert a "good" one while the system is on. An internal drive can be hot-swapped if it is put in a special craddle with a handle and a lock. This has been the case with SCSI drives for a long time.
Methane is an odorless gas. Need to go back and open your chemistry book...
Well, you can tell them appart. When they waddle they swing their hips more and they flap their wingtips everytime they say "fabulous" (with a lisp of course).
Yes, word of mouth, my mom called me and told me about the movie. After I saw it I told some of my friends who went and saw it and I think they liked it enough to recomment to thier friends and so on. So even though someone (let's say my parents in this case) saw an add on TV for it, I didn't and I don't know if all my friends did so that makes 'word of mouth' I think.
A good fantasy or sci-fi, or any story about alien places and creatures is successful (=appeals to the audience) only if they emobody human ideals in them like justice, honesty, self-sacrifice, love, beauty, overcoming adversity and other such things. In other words if you had a movie about worms that live at the bottom of the ocean, or even some alien bacteria (or just mattrasses that sit around ) from Mars or say Titan, you couldn't entice the audience as much.
[1] It is G rated. I guess some people are sick of car chases, boobies and scary monsters - go figure! The "naughtiest" part is penguins doing "it" to make more penguins. I actually saw a parent take their child out of the theatre for this one. Felt sorry for the poor kid...
[2] Penguins are somewhat similar to humans in the way they walk and behave. They walk upright but wobble and thus they look like "cute", "fat", "fuzzy" people. You couldn't pull this one off with snakes or, tigers. They are social creatures. In the movie ( I hope I don't spoil it for anyone ;) a mother penguin who lost her egg, tries to steal the chick from another penguin. The other females in the group would not "approve" of such behavior and came to protect the chick and the mother from the "thief".
Also penguins are monogamous (emperor penguins are monogamous at least for duration of one year) - which often is not the case with many humans nowadays - not that there is anything wrong with it... So that also anthropomorphosizes them even more.
[3] The bravery and determination of the people who shot the movie is impressive. Very cold weather, very dangerous, all just to film the cute little birds. And, of course, as some post mentioned, some like Morgan Freeman.
[4] It spread mostly by word of mouth. This is similar to the "My Big Fat Greek Wedding". A small movie that made big $ because everyone told their friends to go see and how great it was. I heard about the movie from my parents then after I saw it, I recommended it to all my friends and they saw it. If everyone who see it does it - it is quite a few people..
Dear AC, Your post is implying that stereotypes are bad. You list all the stereotypes then you say that there is nothing interesting there. But you never really prove or show that stereotypes are bad. It is as if I said "your post is full of adverbs, and that is why it should not be moderated too high".
But every investor had dollar signs flashing in their eyes and dreams of making billions and be the next Bill Gates.
I personally think many looked at Microsoft, and Intel and thought "Damn ,if knew back then and bought the stock... well I learned my lession, I better go and invest now in all this technology"
Another warning sign was when I found out this guy with 2 years of technical college and almost no experience was getting $80/hour to make simple websites for companies. That couldn't last forever. A lot of those people are still out there crying that they are out of a job and that nobody wants them. They still don't understand that their skills are not worth that much. In no other field could you go to a technical school for 2 years then make $80/hour right way. It is amazing how so many were fooled.
Oh, wow, I had about 4 of those under my bed for a while then threw them away. I wish I new they went for $50-$100 a pop.
You are an exception rather than the average. Keep up the good work, I wish there were more people like you!
But now that I've lived here, I want my children to be able to study science and be proud of that and not have to always look up at some jocks who can't count and read or write yet he'll somehow become the manager of the of the future businesses.
What concerns me the most is the lack of balance. The problem as I see it, is that it is fundamentaly easier to study business, marketing, law and economics than to study science. In other words a computer geek, won't have a problem in a business class. Now take an average business student and try teaching him computer science, I think it will be much harder. So I think eventually after the countries like India and China will end up getting far enough ahead in terms of science, it won't take them much to figure out the business aspect too. (A side note, up until recently at P&G they were hiring scientists from Easern European countries like Romania and such, now they hire accountants and marketing specialists also, those positions are classical 'American' jobs). The same thing with Chinese, after getting a PhD at an American university, it won't take that much effort for them to get an MBA. Now try sending an american with an MBA to get a PhD in a science related field - no chance.
Also note the trends in outsourcing. Initially it was just basic, scripted customer support, then software development, then they exported the design of the software, now some companies export their accounting, marketing and other areas like that. Pretty soon the'll export everyone except the CEO. Eventually, the people overseas will get rid of the CEO too, and buy the company and run it. Then we'll become the Mexico and China. And then everything in China will be 'made in U.S.A' by people who can't compete they slacked off in school, played sports, went to band camp instead of learning how to read, write and count.
But if you are close enough to the towers, why not just call the other person? With voice activated dialing, it is not much slower then 'push-to-talk'
Anyone who seems to be interested in science in this country was and still is a "nerd" and thus unpopular and a social outcast. Everyone wants to be friends with the athletic football jocks, the nerds and geeks are the ones who get picked on.
The only way kids are encouraged to be succesful (read=make tons of $$$) by the society (media, family, friends) is to go to college, join a fraternity, party 4 years while taking some business classes then join daddy's or uncle's company with a $80,000 starting salary. Well, that seems to be working so far but for how long?
So yeah, glamorizing science is a good step in the right direction, but I wonder if it too late.