I heard this story first (or possibly second) hand in Sweden; specifically Goteborg.
So, this is for admissions for dental school, about 5 years ago. Some bug causes the students with the _lowest_ test scores to be admitted.
The error is discovered but then the admission decisions have already been sent out. The school finds it inhumane to retract the offers those who have been admitted in error. However, they also find it unfair for the most qualified students: it is decided to admit both groups.
The funniest part is that rumor has it that there was no significant difference in performance between the two groups.
But then, I suppose no one actually got hurt there.
Yeah, and when I quickly googled your article I found that the stolen uranium was U238, depleted uranium; used for example as a _shield_ against more radioactive materials (because of its great density). Anyway, this sort of illustrates how differently radioactivity is evaluated from everything else.
If a new Chemical University opened I doubt you would hear 20 year old stories about Chem majors who brew alcohol which could have contained methanol (but didn't) or the chemical experiment which almost started a fire (but didn't); yet I am sure there are plenty of examples even involving deaths.
A bunch of drunk college frat boys with nuclear waste. Nothing can go wrong with that at all.
Many American and European universities already have small reactors and dangerous radioactive material. I have never heard of an incident where anybody got hurt.
While solar, wind and tidal power look very attractive, they suffer from the problem of being at the mercy of nature.
I think the biggest problem with these technologies is that they take up very large areas. This is fundamental limitation; if we want to get large amounts of energy from solar cells and wind power we have to give up large areas for these purposes.
As a friend of the environment, I would much rather use such areas for wildlife/ national parks and take the energy from nuclear plants.
It is too bad the environmental movement is so dogmatic; they get these ideas that certain things are Bad, and at that point no science or rational comparisons can make them change their mind. It does not matter if that there has been tremendous development of nuclear technology in terms of efficiency/ security/ waste.
It seems like GM is facing the same issues. Instead of discussing intresting tradeoffs suchs as herbicedes/ GM/ larger areas for cultivation the enviroment lobby is completely fanatic.
It too bad, because the issues they argue about are really important.
Mars is all of 1% closer this year than it has ever been in the last 600 years. This is an almost insignificant amount
Yes and no. Mars brightness varies very much over a cycle of around 2 years. Right now is the peak, which makes a great time for Mars observations. In addition, this particular peak is slightly stronger than it has been for a long time, but as you say that effect is quite insignificant.
But why so negative about the buzz? While it makes little difference for observations and travel, isn't it great that people from all walks of life can feel excitement and wonder over our red neighbor. This is the type of public interest needed to make travel there possible.
Mass automation is a huge opportunity and also a huge risk for billions of people. It has to be managed, not left to the whims of the market, which will be increasingly controlled by fewer and fewer extremely wealthy people.
The great amounts of capital are not increasingly controlled by fewer and fewer extremely wealthy people. It increasingly owned by normal people, specifically through their insurances and retirement funds.
These discussions are interesting, but the more extreme predictions are worth taking with a grain of salt. The imminent arrival of an automated society where everyone is unemployed has been predicted for 200 years; yet unemployement has fluctuated around a few percent or a little more in the down cycle.
Do a full backup once a year and a 20 month lifespan for the media doesn't matter...
Maybe not. The 20 months is just an abstraction for when the media has reached a far stage of aging. Of course, in reality, it ages continiously.
Making a copy after 10 months (of a half-aged) file does not solve the problem unless you have a scheme for recovering everything that has been lost during the first 10 months.
There are such schemes; essentially if you build in redundancy in the data and restore it when you copy. I don't know if these are commonly used by standard file formats and copy programs though (Anybody knows?).
Just-in-time = Minimize your inventory of subcomponents by ordering as late and as little as possible from suppliers.
Build-to-order = Build a car if and only if ordered by a customer
Now, are you sure that Mini Cooper are built to order ? If you can go to a dealer and drive one home then clearly it is not.(I don't doubt that they practice just-in-time inventory, it is industry standard)
Tor
Re:Huh? That's how it's here in the Netherlands
on
Build-to-Order Cars?
·
· Score: 4, Interesting
You can do that in the USA, too. Virtually any auto maker will build you a car to your specs (options, color, etc.). Since the 1980s the options have been more-or-less bundled, and it's a bitch to order them separated (I want a sunroof but I don't want to pay for fog lights to get it!), but you can do it.
People seem to have missed the point completely. The point isn't that build-to-order enables a bunch of new varieties. Sure, you can still order your favorite color and CD player. The difference is how it is achieved: presently they build a bunch of common varieties. If you want an unusual combination, they will make sure that it is included in the next shipment (or possibly even build it for you).
The difference with build-to-order is that cars are only made if they are ordered. This has the potential of lower car costs, because you minimize the losses of the varieites that stand unsold, and the ones that have to be shipped from one area to another.
I don't see how this company is going to overcome the public's desire for instant gratification. Pick a car, drive it home today.
This is a good point. I think the answer is money. If I could get a car at say $500 or $1000 cheaper (which may be achievable with the new business model), I would probably be willing to wait with the gratification for a week or two.
The current auto manufactureres are already the "Dell of automobiles." They give you a selection of options (color, radio, interior, etc.) and you can choose what you want. If you want something for which they don't have a configuration then you don't get it
The difference is that if you don't want something that they do have, then the unwanted car will be standing, and potentially sold at a discount (whereas the unwanted Dell is never built). The losses from the unwanted cars are passed on to the consumer, in the price of the popular varieties.
Tor
Re:Su-30 series or Quality/Quantity
on
In-Flight Reboot?
·
· Score: 1
In pure features the Su-27 is an amazing plane. Anyone who has ever seen the Su-27 do the cobra manouver or the thrust vectored Su-30MKI or Su-35 do the 360 degree Kulbit manouver can attest to what these planes can do in close air combat. These are extreme manouvers that western planes cannot do for the simple reason that the engines in western planes receive no air at such high angles of attack and therefore often flame-out or stall. Not only this but the newer radars on the Su-30s and missiles are longer ranging than just about anything the west has with the exception of the F-14's AIM-54 Phoenix. As for stealth, newer Su-30's are coated with radar absorbant paint which reduce the advantages that a dedicated stealth fighter such as the F-22 would have in BVR combat.
In the hands of a good pilot I very much doubt that the Su-30 would automatically lose in combat. That however is the crux of the matter: Pilot training.
I think there is another very important issue as well. The SUs can indeed do amazing manouvers which make them formidable in close combat. It is just that air combat is rarely close anymore. The winner is typically the guy with the best informations systems and sensors and longest range missiles. I think the US has done the right strategic decision in focusing on this rather than on extreme manouverability.
Look at how unpredictably the stock markets behave already
Before we bash this system too much, it is probably worth thinking about alternatives. Everybody should agree that asking middle eastern experts about events in the middle east is a good idea; if you have many experts you can esitmate a probability by taking the average of their estimated probabilities. This is certainly imperfect, but what solution is likely to do better?
The market idea has the additional benefit of letting the experts place bets; i.e., they can put in more money if they are more confident in their assessment and less money if they are less confident.
Right now, apparently, the market is only open for a select group of experts. But say that even anonynmous_loser was let into this market (and that the problem of terrorists betting on their own acts were solved). Why would he palce a bet unless he thought he had a valuable insight, not accurately captured already? Similarly, an expert who finds that he always loses money on this in encouraged to quit, but a good expert gets encouraged to bet more.
I would challenge people who don't believe in this to come up with a more efficient prediction system. Given a set of middle eastern experts with varying (but uncertain) capability and who make different predictions; the desired output is accurate probabilities of events.
But that has nothing to do with OS stability. The OS does not have to determine if the program will end, or even shutdown properly. Since the OS is the arbiter of resources, it can make the decision to disallow a program from executing any further, without consulting the program beforehand. It is also the protector of programs, keeping one from trouncing another.
Hum, my post was in response to someone saying that an OS shouldn't allow any app to crash. This is how I reasoned.
The OS can either shut down suspicious apps after some finite time (a), or just let them run as long as they want, (b).
As we both seem to agree, the OS cannot always be expected to know if an app will terminate properly.
If it does a), then some programs which work fine but take a bit of time will be killed by the OS - which could be considered crashing.
But if it follows b) then some bad apps which ought to be killed will be roaming free, which perhaps also could be considered a crash.
Thus it would be impossible to create an OS which is fool-proof from bad apps.
But frankly, I know more about theoretical computer science than of practical OS considerations and crashing apps. Perhaps the reasoning above is not consistent with the typical definitions of a "crash".
During the 70s and 80s the mantra was for companies to get bigger and bigger, until they become large conglomerates.
These ideas build on relatively dubious arguments. It is true that bigger usually means economies of scale, but this is not really true when one company acquired another that did something completely different (as was often the case).
During the 90s these ideas were replaced by the opposite (and, I believe economically more sound) thinking. Rather than getting as big as possible, companies should identify the activities that they are really good at, and do only that. They could sell and consult on those things, but should outsource everything else.
Most companies would probably not list as their core competencies IT maintenance/ development, payroll processing and telephone support. It follows that it may be a good idea to outsource this to somebody specializing in doing only that.
As for the outsourcing to the US/ abroad it is simply a question of whether the lower foreign salaries outweight the costs of dealing with somebody 12 timezones away who speaks poor English.
Personally I am getting help developing a piece of commercial software right now. While I wouldn't have any moral issues with hiring such help from India, I prefer hiring somebody locally so that we can meet face to face and discuss issues as they come up.
Similarly, I think your best bet is to make sure that you are competitive on the basis of good, simple communication and on a good understanding of the customer and the end user.
A nice idea, but imagine what it would be like in practice? Britney, Christina, and friends would all have amazing karma and artists like Brian Eno would languish at the bottom of the Hellmouth because mainstream people wouldn't get it.
I think Amazon has been quite successful in avoiding this. You search on specific key words and then look at ratings and reviews. They also have tips such as "people who bought this also liked that". This could work for music also.
1 Cardiac? Running up and down the stairs a few times isn't going to do it. You have to spend at the very least 20 minutes with elevated pulse. And why do you want to do it during your 10 hours of work? Use a machine or take a jog before or after.
2 Build some muscle. Essentially the same as 1), it would make more sense to go the gym or such before or after work.
3 Trying to stop blood clots and other nasty effects of sitting still for too long. Well, then, you don't need any push-ups, just take a little walk and do a little stretch. Maybe do like Wally and find somebody to chat with in the other end of the office.
There was a time when companies would make long-term investments in a program, knowing it'd be many years before they'd earn it back, but those days are long gone
Counter-examples: X-Box, CPU devlopment, pharmacy research, movie production.
The reason that private companies won't build a space shuttle is that they haven't been contracted to do so. Of course they could try to build one anyway and sell it afterwards but that is extremely risky (as opposed to just delayed payments).
I just got this crazy mental picture of a crew of burly blonde men trying to heft an oil drum sized silencer onto a 155mm howitzer.
Forgive me, but where on Earth would you use silencer in the Swedish artillery?
No, we actually never used silencers, and I don't know much about them. I should have read your post more carefully, I thought you cracked down on my mentioning of bullet velocities and the speed of sound - but those are things that every nerd knows...
And btw, we actually did use 155mm howitzers, and there were a bunch of burly blonde men (and a few women). Good stuff. ; )
What on earth makes you think that supersonic projectile velocity is necessary for a sniper rifle?
Well, sniping is often done at long ranges. If you are firing at 600 or 1000 m that is a 2 or 3 seconds that pass before the bullet reaches the target.
...that low velocity rounds can be highly useful in a number of applications.
I don't doubt that at all, but now the issue was their benefit in sniper rifles, in particular those made out of Gauss guns. And I think my point still holds, that subsonic is not your typical method of choice for sniper rifles (and if you really were prepared to go subsonic, then as your examples show you wouldn't need a Gauss gun).
the fuckers head will already be exploded by the time he hears it!!
Yes, this is the case for all sniper rifles. But the as the original poster was aware it can be pretty useful to snipe silently anyway, because the fucker may have friends (with weapons).
Slashdot Mirror
I heard this story first (or possibly second) hand in Sweden; specifically Goteborg.
So, this is for admissions for dental school, about 5 years ago. Some bug causes the students with the _lowest_ test scores to be admitted.
The error is discovered but then the admission decisions have already been sent out. The school finds it inhumane to retract the offers those who have been admitted in error. However, they also find it unfair for the most qualified students: it is decided to admit both groups.
The funniest part is that rumor has it that there was no significant difference in performance between the two groups.
Tor
This was an interesting story.
But then, I suppose no one actually got hurt there.
Yeah, and when I quickly googled your article I found that the stolen uranium was U238, depleted uranium; used for example as a _shield_ against more radioactive materials (because of its great density). Anyway, this sort of illustrates how differently radioactivity is evaluated from everything else.
If a new Chemical University opened I doubt you would hear 20 year old stories about Chem majors who brew alcohol which could have contained methanol (but didn't) or the chemical experiment which almost started a fire (but didn't); yet I am sure there are plenty of examples even involving deaths.
Tor
A bunch of drunk college frat boys with nuclear waste. Nothing can go wrong with that at all.
Many American and European universities already have small reactors and dangerous radioactive material. I have never heard of an incident where anybody got hurt.
Tor
While solar, wind and tidal power look very attractive, they suffer from the problem of being at the mercy of nature.
I think the biggest problem with these technologies is that they take up very large areas. This is fundamental limitation; if we want to get large amounts of energy from solar cells and wind power we have to give up large areas for these purposes.
As a friend of the environment, I would much rather use such areas for wildlife/ national parks and take the energy from nuclear plants.
It is too bad the environmental movement is so dogmatic; they get these ideas that certain things are Bad, and at that point no science or rational comparisons can make them change their mind. It does not matter if that there has been tremendous development of nuclear technology in terms of efficiency/ security/ waste.
It seems like GM is facing the same issues. Instead of discussing intresting tradeoffs suchs as herbicedes/ GM/ larger areas for cultivation the enviroment lobby is completely fanatic.
It too bad, because the issues they argue about are really important.
Tor
Mars is all of 1% closer this year than it has ever been in the last 600 years. This is an almost insignificant amount
Yes and no. Mars brightness varies very much over a cycle of around 2 years. Right now is the peak, which makes a great time for Mars observations. In addition, this particular peak is slightly stronger than it has been for a long time, but as you say that effect is quite insignificant.
But why so negative about the buzz? While it makes little difference for observations and travel, isn't it great that people from all walks of life can feel excitement and wonder over our red neighbor. This is the type of public interest needed to make travel there possible.
Tor
Mass automation is a huge opportunity and also a huge risk for billions of people. It has to be managed, not left to the whims of the market, which will be increasingly controlled by fewer and fewer extremely wealthy people.
The great amounts of capital are not increasingly controlled by fewer and fewer extremely wealthy people. It increasingly owned by normal people, specifically through their insurances and retirement funds.
These discussions are interesting, but the more extreme predictions are worth taking with a grain of salt. The imminent arrival of an automated society where everyone is unemployed has been predicted for 200 years; yet unemployement has fluctuated around a few percent or a little more in the down cycle.
Tor
Do a full backup once a year and a 20 month lifespan for the media doesn't matter...
Maybe not. The 20 months is just an abstraction for when the media has reached a far stage of aging. Of course, in reality, it ages continiously.
Making a copy after 10 months (of a half-aged) file does not solve the problem unless you have a scheme for recovering everything that has been lost during the first 10 months.
There are such schemes; essentially if you build in redundancy in the data and restore it when you copy. I don't know if these are commonly used by standard file formats and copy programs though (Anybody knows?).
Tor
on a tight budget, no less
Hire an expensive professinal, that will solve the problem and be more than a tight budget.
Tor
Just-in-time = Minimize your inventory of subcomponents by ordering as late and as little as possible from suppliers.
Build-to-order = Build a car if and only if ordered by a customer
Now, are you sure that Mini Cooper are built to order ? If you can go to a dealer and drive one home then clearly it is not.(I don't doubt that they practice just-in-time inventory, it is industry standard)
Tor
You can do that in the USA, too. Virtually any auto maker will build you a car to your specs (options, color, etc.). Since the 1980s the options have been more-or-less bundled, and it's a bitch to order them separated (I want a sunroof but I don't want to pay for fog lights to get it!), but you can do it.
People seem to have missed the point completely. The point isn't that build-to-order enables a bunch of new varieties. Sure, you can still order your favorite color and CD player. The difference is how it is achieved: presently they build a bunch of common varieties. If you want an unusual combination, they will make sure that it is included in the next shipment (or possibly even build it for you).
The difference with build-to-order is that cars are only made if they are ordered. This has the potential of lower car costs, because you minimize the losses of the varieites that stand unsold, and the ones that have to be shipped from one area to another.
I don't see how this company is going to overcome the public's desire for instant gratification. Pick a car, drive it home today.
This is a good point. I think the answer is money. If I could get a car at say $500 or $1000 cheaper (which may be achievable with the new business model), I would probably be willing to wait with the gratification for a week or two.
Tor
The current auto manufactureres are already the "Dell of automobiles." They give you a selection of options (color, radio, interior, etc.) and you can choose what you want. If you want something for which they don't have a configuration then you don't get it
The difference is that if you don't want something that they do have, then the unwanted car will be standing, and potentially sold at a discount (whereas the unwanted Dell is never built). The losses from the unwanted cars are passed on to the consumer, in the price of the popular varieties.
Tor
In pure features the Su-27 is an amazing plane. Anyone who has ever seen the Su-27 do the cobra manouver or the thrust vectored Su-30MKI or Su-35 do the 360 degree Kulbit manouver can attest to what these planes can do in close air combat. These are extreme manouvers that western planes cannot do for the simple reason that the engines in western planes receive no air at such high angles of attack and therefore often flame-out or stall. Not only this but the newer radars on the Su-30s and missiles are longer ranging than just about anything the west has with the exception of the F-14's AIM-54 Phoenix. As for stealth, newer Su-30's are coated with radar absorbant paint which reduce the advantages that a dedicated stealth fighter such as the F-22 would have in BVR combat.
In the hands of a good pilot I very much doubt that the Su-30 would automatically lose in combat. That however is the crux of the matter: Pilot training.
I think there is another very important issue as well. The SUs can indeed do amazing manouvers which make them formidable in close combat. It is just that air combat is rarely close anymore. The winner is typically the guy with the best informations systems and sensors and longest range missiles. I think the US has done the right strategic decision in focusing on this rather than on extreme manouverability.
Tor
Look at how unpredictably the stock markets behave already
Before we bash this system too much, it is probably worth thinking about alternatives. Everybody should agree that asking middle eastern experts about events in the middle east is a good idea; if you have many experts you can esitmate a probability by taking the average of their estimated probabilities. This is certainly imperfect, but what solution is likely to do better?
The market idea has the additional benefit of letting the experts place bets; i.e., they can put in more money if they are more confident in their assessment and less money if they are less confident.
Right now, apparently, the market is only open for a select group of experts. But say that even anonynmous_loser was let into this market (and that the problem of terrorists betting on their own acts were solved). Why would he palce a bet unless he thought he had a valuable insight, not accurately captured already? Similarly, an expert who finds that he always loses money on this in encouraged to quit, but a good expert gets encouraged to bet more.
I would challenge people who don't believe in this to come up with a more efficient prediction system. Given a set of middle eastern experts with varying (but uncertain) capability and who make different predictions; the desired output is accurate probabilities of events.
Tor
But that has nothing to do with OS stability. The OS does not have to determine if the program will end, or even shutdown properly. Since the OS is the arbiter of resources, it can make the decision to disallow a program from executing any further, without consulting the program beforehand. It is also the protector of programs, keeping one from trouncing another.
Hum, my post was in response to someone saying that an OS shouldn't allow any app to crash. This is how I reasoned.
The OS can either shut down suspicious apps after some finite time (a), or just let them run as long as they want, (b).
As we both seem to agree, the OS cannot always be expected to know if an app will terminate properly.
If it does a), then some programs which work fine but take a bit of time will be killed by the OS - which could be considered crashing.
But if it follows b) then some bad apps which ought to be killed will be roaming free, which perhaps also could be considered a crash.
Thus it would be impossible to create an OS which is fool-proof from bad apps.
But frankly, I know more about theoretical computer science than of practical OS considerations and crashing apps. Perhaps the reasoning above is not consistent with the typical definitions of a "crash".
Tor
It doesn't matter what "causes" the crash. The OS should be essentially crashproof
But it is theoretically impossible to for an observer (i.e., the OS) to determine whether another program (i.e., the app) will shut down properly.
In computer science, this is known as the halting problem, and it can be mathematically proven.
Tor
Is it really as simple as money?
During the 70s and 80s the mantra was for companies to get bigger and bigger, until they become large conglomerates.
These ideas build on relatively dubious arguments. It is true that bigger usually means economies of scale, but this is not really true when one company acquired another that did something completely different (as was often the case).
During the 90s these ideas were replaced by the opposite (and, I believe economically more sound) thinking. Rather than getting as big as possible, companies should identify the activities that they are really good at, and do only that. They could sell and consult on those things, but should outsource everything else.
Most companies would probably not list as their core competencies IT maintenance/ development, payroll processing and telephone support. It follows that it may be a good idea to outsource this to somebody specializing in doing only that.
As for the outsourcing to the US/ abroad it is simply a question of whether the lower foreign salaries outweight the costs of dealing with somebody 12 timezones away who speaks poor English.
Personally I am getting help developing a piece of commercial software right now. While I wouldn't have any moral issues with hiring such help from India, I prefer hiring somebody locally so that we can meet face to face and discuss issues as they come up.
Similarly, I think your best bet is to make sure that you are competitive on the basis of good, simple communication and on a good understanding of the customer and the end user.
Tor
Maybe a moderation system is in order?
A nice idea, but imagine what it would be like in practice? Britney, Christina, and friends would all have amazing karma and artists like Brian Eno would languish at the bottom of the Hellmouth because mainstream people wouldn't get it.
I think Amazon has been quite successful in avoiding this. You search on specific key words and then look at ratings and reviews. They also have tips such as "people who bought this also liked that". This could work for music also.
Tor
1 Cardiac? Running up and down the stairs a few times isn't going to do it. You have to spend at the very least 20 minutes with elevated pulse. And why do you want to do it during your 10 hours of work? Use a machine or take a jog before or after.
2 Build some muscle. Essentially the same as 1), it would make more sense to go the gym or such before or after work.
3 Trying to stop blood clots and other nasty effects of sitting still for too long. Well, then, you don't need any push-ups, just take a little walk and do a little stretch. Maybe do like Wally and find somebody to chat with in the other end of the office.
Tor
Summary: Functionally clumsy, but it looks cool!!
That watch is my polar opposite, it seems. (probably true of most geeks).
Now you can have it all... buy the watch.
Tor
As it it wasn't hard enough to get a date, now they want me to strap a PDA to my wrist. Yeah, that'll reel 'em in...
Translation: Psiren is a woman at Caltech. ; )
Tor
than an entire mini-PDA would be a watch linked to your regular PDA or computer.
It should be able to display reminders etc but would n't really need a full user interface.
Tor
There was a time when companies would make long-term investments in a program, knowing it'd be many years before they'd earn it back, but those days are long gone
Counter-examples: X-Box, CPU devlopment, pharmacy research, movie production.
The reason that private companies won't build a space shuttle is that they haven't been contracted to do so. Of course they could try to build one anyway and sell it afterwards but that is extremely risky (as opposed to just delayed payments).
Tor
I just got this crazy mental picture of a crew of burly blonde men trying to heft an oil drum sized silencer onto a 155mm howitzer. Forgive me, but where on Earth would you use silencer in the Swedish artillery?
No, we actually never used silencers, and I don't know much about them. I should have read your post more carefully, I thought you cracked down on my mentioning of bullet velocities and the speed of sound - but those are things that every nerd knows...
And btw, we actually did use 155mm howitzers, and there were a bunch of burly blonde men (and a few women). Good stuff. ; )
Tor
What on earth makes you think that supersonic projectile velocity is necessary for a sniper rifle?
...that low velocity rounds can be highly useful in a number of applications.
Well, sniping is often done at long ranges. If you are firing at 600 or 1000 m that is a 2 or 3 seconds that pass before the bullet reaches the target.
I don't doubt that at all, but now the issue was their benefit in sniper rifles, in particular those made out of Gauss guns. And I think my point still holds, that subsonic is not your typical method of choice for sniper rifles (and if you really were prepared to go subsonic, then as your examples show you wouldn't need a Gauss gun).
Tor
the fuckers head will already be exploded by the time he hears it!!
Yes, this is the case for all sniper rifles. But the as the original poster was aware it can be pretty useful to snipe silently anyway, because the fucker may have friends (with weapons).
Tor