Who said anything about system compromise? I was thinking along the lines of "the mail was sent 1.23 seconds after user A got a DHCP lease for the sending address, which was 0.58 seconds after user B gave up that address". "Forensic" has a meaning outside of computer security circles, you know.
Its called a cell phone.
Good for you. I prefer a nicely analog Swiss model, myself. My timepiece of choice isn't very external-clock compatible.
We'll assume that non-relativistic simultaneity is a reasonable model for the current Internet for the sake of conversation. It's more polite for the non-physicists among us.
if I lose my internet connection for a month I'm still within a second of the atomic clock.
ISC ntpd does the same thing (see "FREQUENCY DISCIPLINE" in ntpd(1) on a Unix machine near you). OpenNTPD depends on the kernel to Do The Right Thing, which the BSDs, at least, seem to do. It has the additional advantage of not being under Dan's Crackpot License so people who actually want to develop it have the legal right to do so, which makes it a much more attractive target for people who want to add features, fix bugs, etc.
NTP is based on UTC and doesn't handle leap seconds as gracefully as TAI.
That may or may not be true - I don't know enough about that exact topic to comment - but POSIX is also based on UTC and not TAI. DJB's tools may perfectly implement the standard that noone else uses, but I'm not ready to consider that a benefit.
Of course, the interesting part is that packets don't travel instantaneously. It's not enough to ask what the time is - you also need to know how long it takes the responder to answer. Furthermore, travel times aren't symmetric and you have to estimate how much of the total elapsed time between sending the request and receiving the answer occurs during each leg of the trip. Remember, the other guy faces the same constraints that you do - you can't just ask him for the answer.
Time synchronization is a deceptively simple concept. I'm very grateful that some very smart people have thought this through to a level of detail that I probably never would.
I usually use good ol' ISC ntpd, but I've been looking at OpenNTPD for leaf nodes. Do you have any links to criticisms handy? I remember one that included things like "it's mainly developed on OpenBSD" that seemed to be pretty well debunked.
Don't be so needlessly antisocial. Pick a nice public stratum-2 server and leave the big guys alone. It reduces load (thus latency, thus inaccuracy) at the top and probably gives you better accuracy, assuming you're not in the same building as tycho.
I'd much rather sync against my ISP's GPS-based NTP server than a better source far away. It's better in every way, and it won't make the stratum-1 guys want to punch you.
By the way, clockspeed hasn't been updated since October 1998. OpenNTPD is a light, modern client that you might wish to consider.
Yes, if you're into the meta-philosophy of timekeeping. If not, then the first part of this book obviously isn't for you.
All I know is that I have the two Domain Controllers sync to 6 outside machines (reliable servers)
Do you know enough about NTP to be able to evaluate whether those are good sources? If not, then the second part of this book may be useful.
All the computers in the network are within a second of eachother - and thats good nuff for me.
...you hope. As the reviewer said, precise synchronization is absolutely critical for forensic purposes. Can you conclusively demonstrate that Event A on machine Foo occurred before Event B on machine Bar? Would you be willing to testify to that fact if a conviction or monetary judgment were on the line?
Even if you don't think you need it, it's so trivially easy to get millisecond accuracy on all but the most unstable tickers that I never have understood why people don't sync their computers. If nothing else, it's nice to be able to set your watch to a known-accurate time source, then use that to set the rest of the clocks in your house. Am I the only one that hates seeing different times on the combination oven and microwave that has two separate displays within two feet?
If you take hash algo A at 32 bits, and algo B at 32 bits, but B is weaker than A, then hash collision calculation would be less than the complexity of A squared. (Since B is weaker than A)
Yes, but if you later discover that A reduces to O(1), while B stays at 32 bits, then your combined algorithm degrades to 32 bits instead of none. Wouldn't that be a benefit?
I mean, it would seem dumb to JIT-compile to X86, which in turn is translated to VLIW
...unless the CPU is better at runtime profiling than the compiler is at compiletime profiling, in which case it might actually be a very big win to let it do its magic.
Yeah? How far? I'm in Norfolk, about 60 miles south of Yankton.
I don't usually see these types of cars 'pimped' out.
I hate to say this, but the local chicken plant's parking lot is a pretty large showroom for such things.
I think people need to realize that it is ok to drive a car, especially a cheap one that is cheap to maintain, and have an older (carburated) pikcup that they use seldom. This type of combination is great.
The first vehicle I ever drove was a rusted-out '70 Chevy Custom in my dad's field (we had five acres on the edge of town, so "field" may be an overstatement). My mom still uses it to haul things around, and just can't bear to get rid of it. On the other hand, driving that same truck loaded with topsoil a few years ago was one of the more exciting events of my life, if "realizing that ancient drum brakes don't stop a load of dirt on a dime" meets your definition of exciting.
I'll certainly agree that the Series II was a better engine, but my previous car was an inherited '87 Olds 98 with a 3800, and I was always happy with its performance.
The valve issue is something to take into consideration though.
Just so long as you're aware of it. I wonder how many of the hopped-up Sentras you see will still be running 5 years from now.:-)
Though hardly anyone puts after-market exhausts on the mid-sized GM cars.
Sez you. I live in a small town in Nebraska where my wife's Toyota minivan gets strange looks. You are highly encouraged to buy American around here, even if I personally don't consider that a factor (what's the difference between a Mexican Chevy and a Tennessee-made Toyota?). You'd be amazed at the number of weird aftermarket things you see around here.
Note that the only difference between pavelvp's article summary and the first two paragraphs of the actual article are that he removed the words "on Tuesday" from the the last sentence. Basically, if you read the summary, then you've been tricked into reading 1/3 of the article itself. That's a cruel prank to play on the Slashdot crowd.
I guess it's small potatoes in the scheme of things, though.
If Solaris is available under a Free Software license, then who really cares which one "wins"? If I find myself using a Solaris kernel that incorporates the good stuff from Linux, I lack the imagination to see how I'd be worse off. If Solaris isn't available under a GPL-compatible license, then I can't see enough people migrating to it to make a huge dent in Linux usage. Once again, I'd be no worse off.
I guess this just seems like a non-issue. Linux Killer? No way. Linux's Friendly Competitor? Welcome to the club!
I own a 3.8L V6 1991 Oldsmobile. It doesn't have much for power, or in fact much of anything.
I have a 3.8L V6 1998 Oldsmobile and it can pass anything but a gas station. You may have other problems; the 3800 was never known as a wimpy engine.
Put a bigger exhaust on it to move the air out as well
Note that if you decrease the exhaust's backpressure out of the designed range, you can dramatically increase your emissions while simultaneously burning out your valves. Bigger can be a little better, but don't get carried away.
The government would look at the total value of the patent, adjust it for the time value of the money, and pay out a whopping big check.
If that were the intent (rather than simply nationalizing the product), then what benefit does the government receive from their actions? The right to distribute the drug by their own criteria instead of the economy's? I just don't see what the supposed advantage of that move would be.
As you said in your original post, distribution is the key problem preventing drugs from reaching the poorest. I see no reason to believe that government control would be any more effective than industry efforts at solving that problem, and every reason to believe the opposite.
How about saving millions of childrens lives? Isn't that fucking good enough anymore? Last I checked, glory and triumph for humankind was worth a lot more than paper. Is money all that anybody cares about these days?
I am Mr. Evil Big Drugs. I just spent $5 billion on AIDS research. The government takes my work from me and gives it away without my consent. What does the public get today? A cure for AIDS. What does the public not get tomorrow? A cure for any other disease I was currently researching, because I'm pulling all funding so that I can bail out with whatever capital I can still scavenge from my dying company.
And who should be the one entitled to "Get rich" from this cure? The scientist who read some charts and took some samples? Or the people who built his lab; built, delivered, installed, and repair his equipment; the power company, the water company, et. al.
What is it exactly that you think scientists do? Stand around until the answer to their problems magically appears? Assuming that the lab owner paid the utility bills and its employees according to their contracts, those parties have zero right to any additional profits - they were already paid for their contribution. That's a strawman, and if you're smart enough to know it, then you should be ashamed. If you're not, then this conversation is pointless.
Your "solution" feels good, as if the government would finally be Doing Something. Too bad it kills millions as Mr. Evil Big Drugs and all of his associates that aren't too stupid to read the writing on the wall jump ship and switch to a different industry.
It's no fun aimlessly ranting on Slashdot if the experts involved are going to come along and correct us. Please move along and allow us to go back to explaining why you're wrong, thanks.
I have always hoped that should a real cure for AIDs be developed that the United States government would sieze the intellectual property and put it into the public domain.
Of course, any siezure of property has to be (1) in the public interest, and (2) fairly compensated. I know I'd happily support a politician who advocated such an action, regardless of how much money it might cost.
Congratulations: that's the stupidest thing I've read on Slashdot today. Why do you want to kill millions by demonstrating to HIV and cancer drug researchers that you're going to steal their work?
Your conditions are even worse:
1) It's almost never in the public interest to prove that you're willing to remove the incentive for creating new things.
2) Fairly compensating someone for what you stole from them is inherently impossible. Either you pay full market value (which gives you a net gain of zero for the "public interest"), or you pay them less than it's worth (which is hardly fair, is it?).
I am not a huge Ayn Rand fan, but you owe it to yourself to read "Atlas Shrugged" if for no other reason than to see what motivates industry. Hint: it isn't the idea of having government confiscate the work on which you've spent billions of dollars.
Slashdot Mirror
Bad time to find out that the changeset you just sent, compressed, to your boss has the same hash as goatse.cx.
Its called a cell phone.
Good for you. I prefer a nicely analog Swiss model, myself. My timepiece of choice isn't very external-clock compatible.
We'll assume that non-relativistic simultaneity is a reasonable model for the current Internet for the sake of conversation. It's more polite for the non-physicists among us.
ISC ntpd does the same thing (see "FREQUENCY DISCIPLINE" in ntpd(1) on a Unix machine near you). OpenNTPD depends on the kernel to Do The Right Thing, which the BSDs, at least, seem to do. It has the additional advantage of not being under Dan's Crackpot License so people who actually want to develop it have the legal right to do so, which makes it a much more attractive target for people who want to add features, fix bugs, etc.
NTP is based on UTC and doesn't handle leap seconds as gracefully as TAI.
That may or may not be true - I don't know enough about that exact topic to comment - but POSIX is also based on UTC and not TAI. DJB's tools may perfectly implement the standard that noone else uses, but I'm not ready to consider that a benefit.
Of course, the interesting part is that packets don't travel instantaneously. It's not enough to ask what the time is - you also need to know how long it takes the responder to answer. Furthermore, travel times aren't symmetric and you have to estimate how much of the total elapsed time between sending the request and receiving the answer occurs during each leg of the trip. Remember, the other guy faces the same constraints that you do - you can't just ask him for the answer.
Time synchronization is a deceptively simple concept. I'm very grateful that some very smart people have thought this through to a level of detail that I probably never would.
I usually use good ol' ISC ntpd, but I've been looking at OpenNTPD for leaf nodes. Do you have any links to criticisms handy? I remember one that included things like "it's mainly developed on OpenBSD" that seemed to be pretty well debunked.
Only if you want to understand what it's actually doing and intelligently discuss it. Otherwise, nope.
Don't be so needlessly antisocial. Pick a nice public stratum-2 server and leave the big guys alone. It reduces load (thus latency, thus inaccuracy) at the top and probably gives you better accuracy, assuming you're not in the same building as tycho.
I'd much rather sync against my ISP's GPS-based NTP server than a better source far away. It's better in every way, and it won't make the stratum-1 guys want to punch you.
By the way, clockspeed hasn't been updated since October 1998. OpenNTPD is a light, modern client that you might wish to consider.
Yes, if you're into the meta-philosophy of timekeeping. If not, then the first part of this book obviously isn't for you.
All I know is that I have the two Domain Controllers sync to 6 outside machines (reliable servers)
Do you know enough about NTP to be able to evaluate whether those are good sources? If not, then the second part of this book may be useful.
All the computers in the network are within a second of eachother - and thats good nuff for me.
...you hope. As the reviewer said, precise synchronization is absolutely critical for forensic purposes. Can you conclusively demonstrate that Event A on machine Foo occurred before Event B on machine Bar? Would you be willing to testify to that fact if a conviction or monetary judgment were on the line?
Even if you don't think you need it, it's so trivially easy to get millisecond accuracy on all but the most unstable tickers that I never have understood why people don't sync their computers. If nothing else, it's nice to be able to set your watch to a known-accurate time source, then use that to set the rest of the clocks in your house. Am I the only one that hates seeing different times on the combination oven and microwave that has two separate displays within two feet?
We already have your MP3s, and your pr0n isn't that good. I wouldn't bother.
Yes, but if you later discover that A reduces to O(1), while B stays at 32 bits, then your combined algorithm degrades to 32 bits instead of none. Wouldn't that be a benefit?
...unless the CPU is better at runtime profiling than the compiler is at compiletime profiling, in which case it might actually be a very big win to let it do its magic.
You just made that up to see if we're paying attention, didn't you?
<dotcom>Is $85,000 acceptable to start, sir?</dotcom>
Yeah? How far? I'm in Norfolk, about 60 miles south of Yankton.
I don't usually see these types of cars 'pimped' out.
I hate to say this, but the local chicken plant's parking lot is a pretty large showroom for such things.
I think people need to realize that it is ok to drive a car, especially a cheap one that is cheap to maintain, and have an older (carburated) pikcup that they use seldom. This type of combination is great.
The first vehicle I ever drove was a rusted-out '70 Chevy Custom in my dad's field (we had five acres on the edge of town, so "field" may be an overstatement). My mom still uses it to haul things around, and just can't bear to get rid of it. On the other hand, driving that same truck loaded with topsoil a few years ago was one of the more exciting events of my life, if "realizing that ancient drum brakes don't stop a load of dirt on a dime" meets your definition of exciting.
The valve issue is something to take into consideration though.
Just so long as you're aware of it. I wonder how many of the hopped-up Sentras you see will still be running 5 years from now. :-)
Though hardly anyone puts after-market exhausts on the mid-sized GM cars.
Sez you. I live in a small town in Nebraska where my wife's Toyota minivan gets strange looks. You are highly encouraged to buy American around here, even if I personally don't consider that a factor (what's the difference between a Mexican Chevy and a Tennessee-made Toyota?). You'd be amazed at the number of weird aftermarket things you see around here.
I guess it's small potatoes in the scheme of things, though.
Drink.
TIMEFMT='%E'; time /home > /tmp/du.out
Adjust to suit if using an older shell. You're welcome.
I guess this just seems like a non-issue. Linux Killer? No way. Linux's Friendly Competitor? Welcome to the club!
I have a 3.8L V6 1998 Oldsmobile and it can pass anything but a gas station. You may have other problems; the 3800 was never known as a wimpy engine.
Put a bigger exhaust on it to move the air out as well
Note that if you decrease the exhaust's backpressure out of the designed range, you can dramatically increase your emissions while simultaneously burning out your valves. Bigger can be a little better, but don't get carried away.
Source: Car Talk
Nobody knows. Liquid lens.
If that were the intent (rather than simply nationalizing the product), then what benefit does the government receive from their actions? The right to distribute the drug by their own criteria instead of the economy's? I just don't see what the supposed advantage of that move would be.
As you said in your original post, distribution is the key problem preventing drugs from reaching the poorest. I see no reason to believe that government control would be any more effective than industry efforts at solving that problem, and every reason to believe the opposite.
I am Mr. Evil Big Drugs. I just spent $5 billion on AIDS research. The government takes my work from me and gives it away without my consent. What does the public get today? A cure for AIDS. What does the public not get tomorrow? A cure for any other disease I was currently researching, because I'm pulling all funding so that I can bail out with whatever capital I can still scavenge from my dying company.
And who should be the one entitled to "Get rich" from this cure? The scientist who read some charts and took some samples? Or the people who built his lab; built, delivered, installed, and repair his equipment; the power company, the water company, et. al.
What is it exactly that you think scientists do? Stand around until the answer to their problems magically appears? Assuming that the lab owner paid the utility bills and its employees according to their contracts, those parties have zero right to any additional profits - they were already paid for their contribution. That's a strawman, and if you're smart enough to know it, then you should be ashamed. If you're not, then this conversation is pointless.
Your "solution" feels good, as if the government would finally be Doing Something. Too bad it kills millions as Mr. Evil Big Drugs and all of his associates that aren't too stupid to read the writing on the wall jump ship and switch to a different industry.
Economy 101: you fail it.
It's no fun aimlessly ranting on Slashdot if the experts involved are going to come along and correct us. Please move along and allow us to go back to explaining why you're wrong, thanks.
Of course, any siezure of property has to be (1) in the public interest, and (2) fairly compensated. I know I'd happily support a politician who advocated such an action, regardless of how much money it might cost.
Congratulations: that's the stupidest thing I've read on Slashdot today. Why do you want to kill millions by demonstrating to HIV and cancer drug researchers that you're going to steal their work?
Your conditions are even worse:
1) It's almost never in the public interest to prove that you're willing to remove the incentive for creating new things.
2) Fairly compensating someone for what you stole from them is inherently impossible. Either you pay full market value (which gives you a net gain of zero for the "public interest"), or you pay them less than it's worth (which is hardly fair, is it?).
I am not a huge Ayn Rand fan, but you owe it to yourself to read "Atlas Shrugged" if for no other reason than to see what motivates industry. Hint: it isn't the idea of having government confiscate the work on which you've spent billions of dollars.