The most interesting discussion on the comments list was the material on removing export. I had always thought export was a good idea and the comments from EDG completely changed my mind.
The key is that you can look much more appealing to your boss when someone else is chasing you. Your market value is the most someone will pay for you today, so you sometimes need to look around energetically to determine your market value. Once you have established this empirically, you can receive what you're worth (by definition).
Last year my friend had given up on getting an adequate raise at his current employer (asked many times, received little). He looked around and got a firm job offer from another outfit, went to his boss and asked whether the boss wanted to make a counter offer, and the boss gave a raise on the spot more than twice the raise my friend had imagined he could get. He turned down the other offer, stayed with his old outfit, and is much more happily (and gainfully) employed, and has a better relationship with his boss.
The key to this is that he would have been very happy to take the other job too, so this was not a bluff, nor a negotiation in bad faith.
Another friend received the following advice from his boss:
We don't have enough money to pay everyone what they're worth, so we pay them the minimum it takes to keep them. Most people grumble but stay put. Substantial raises are reserved for people who do great work AND have other options. If you have a firm competing offer and I can't live without your contributions, I can get authorization to give you a good raise, but otherwise you're pretty much stuck with COLA.
I see two problems with this. It's not that there is some absolute right vs. wrong way to program, but that this does not fit my style.
The problems I see are:
It only addresses a subset of the possible errors of this sort. It catches assignment instead of equality check only when one argument is a constant, but lulls the coder into a false sense of security regarding the same error when both arguments are variables:
if (my_pointer_1 = my_pointer_2) {... }
It makes it harder to read the code. The natural flow of mathematical or logical thought in English is left-to-right: "is x a pentagon?" not "is a pentagon x?". As a believer in literate programming, I perfer to write code that's closer to native language and use tools together with close human reading of the code to catch errors. For the same reason, I tend to write more verbose code that avoids side effects:
for(;*px != 0; ++px, ++py)
{
*px = *py;
}
instead of
while (*py++ = *px++);
Following this idiom, I just tell my compiler to flag any conditional expression whose argument has side effects as an error and I get a compile-time error for every conditional expression with a side effect. MSVC lets me selectively tell the compiler, via #pragmas, to promote certain warnings to errors. I presume that gcc has a similar capability.
About it being valid C, of course you're right (that's why it's a warning, not an error), but as Alan Holub said about C in the title of one of his books, it gives you Enough rope to shoot yourself in the foot. If such expressions were illegal, we wouldn't need lint and relatives as much as we do.
I have no idea how prawns are meant to be cooked, but barbequeueing em just made black, crispy, carcinogenous prawns that tasted like arse.
Buy a copy of The Thrill of the Grill by Chris Schlesinger. Lotsa good ways to grill shrimp (as we call them on the correct side of the equator) without scorching them.
Schelinger will also explain to you why "barbecue" and "grill" are two completely different concepts. Barbecueing shrimp (cooking them for 12 hours with smokey indirect heat) would indeed be sacrilege against the gods who created these tasty protein lumps, but grilling them (cooking them briefly directly over coals) will produce delicious results.
Think about tan(89.99) versus tan(89.991) (which is very ill-conditioned around 90). Both numbers are not terribly truncated by floating point, but the results are different by about 1,000. Try it and you'll see floating point error isn't as dangerous as things like cancellation, ill-conditioning and the like.
Can you back this up with facts? I would be interested in how many millionaires are savvy businessmen and how many are stupid kids who inherited their wealth without doing squat.
Neither tend to be famous, of course, which is your larger point, but I don't see where you get off making such an assertion without justifying it.
Scaling becomes particularly problematic in DNA computing. The propensity of PCR to exponentially amplifying any contaminant DNA is particularly dangerous for problems that require a large number of PCR iterations.
As I understand them, the fundamentals of DNA computing largely lie in the ability of DNA to act as a quasi-nondeterministic finite automaton via controlled chemical reactions that explore the space of potential solutions with massive parallelism. When a candidate solution is found, it is verified with polynomial complexity. This reduces NP problems to P, but means that you have to be able to operate a very efficient physical/chemical sieve to select the correct answer from among a large number of candidate solutions. It also means that you must scrupulously avoid any contaminants in the pool of potential solutions (see above).
The Bush administration and conservatives are terrified of the word "nuclear" as well--when it applies to any device not under US control
So far as I know, the President has said little or nothing about anything "nuclear." "Nucular," on the other hand, scares the pretzels out of him.
This may explain why, after fulminating on national TV about the danger of "nucular" weapons in Iraq, he held his troops back for several weeks while looters plundered the largest nuclear materials depot in Iraq, which contained over one and a half tons of enriched uranium.
This myth is completely without substance. Volcanoes introduce very little chlorine into the stratosphere. Stratospheric chlorine has been measured to be more than 80% due to CFCs, whereas the largest volcanic injection ever observed (El Chichon, which you mention) increased the stratospheric chlorine content by only 2%. When people look at the chlorine content of the stratosphere, they find it to be dominated by CFCs, so if you want to blame nature, you must find a natural source of CFCs.
This study by NASA explains why volcanic plumes, which contain tremendous amounts of chlorine, don't leave much chlorine in the strtatosphere.
The paper in Science emphasizes that the uptake of tropospheric H2 by soil is unknown and could possibly completely compensate for an increased anthropogenic hydrogen burden.
Another key question is how the residence time of H2 in the stratosphere compares to the residence time of CO2 in the troposphere. If H2 has a significantly shorter residence time than the 120 year residence time of anthropogenic CO2, then it would be a good choice to switch to H2 today and then replace H2 with another alternative at a future date, since the H2 would drop back to its natural level faster than CO2 would. If H2 has a longer residence time in the stratosphere, then the best choice might be to stick with CO2 emissions.
2) then it doesn't matter if non-employees have access.
I am also stating that this part is correct.
If it doesn't matter whether non-employees have access, then why do you advocate implementing a company policy about rogue APs and going around once a month scanning for them? That seems like a lot of effort to throw at something that "doesn't matter."
The Times article suggested that the vaporware point would be very hard to prove. It is pitched as the weak link in the case, not as a groundbreaking strategy.
Also, just for reference, even if the prosecution succeeds, it just means that you can't lie to shareholders about vaporware. There's still nothing wrong with sowing FUD against competitors by making vaporware claims in your advertising so long as you keep them out of your stock prospectus and annual report.
No, you need to find a high-paying occupation that by its nature cannot be outsourced to foreign countries.
This is why so many intelligent Americans end up being lawyers.
At the height of the bubble, Paul Krugman wrote that software development and other high-tech jobs were so easy to outsource that their prestige pay scales were doomed. He also believes that much lawyering will be outsourced or automated (a la Nolo Press and write-your-own-will software).
Paul Krugman has one answer. He argues that as the computer revolution matures, the highly-skilled elite of professionals and programmers which has done well during the past two decades will itself become obsolete. "The time may come when most tax lawyers are replaced by expert systems software, but human beings are still needed - and well-paid - for such truly difficult occupations as gardening, house cleaning, and the thousands of other services that will receive an ever-growing share of our expenditure as mere consumer goods become steadily cheaper," he writes. The computer-adept professionals of today will be like the cottage weavers of the 19th century who cashed in on the earlier technical revolution in spinning until the development of the power loom made them redundant in their turn.
Re:Human behavior more predictable than weather?
on
Crime Prediction
·
· Score: 1
Humans - fear, greed, envy and lust about sums it up
Which one of those made your mother and father read bedtime stories to you when you were young?
Joint and several liability makes it hard to sever the employer's exposure in court. Strict liability means that what makes sense to you or me is not necessarily what happens in court. Consider this:
In the case of
Bigbee v. Pacific Telephone, an intoxicated driver lost control of her car, veered off the road, jumped a curb, crossed the sidewalk, went into a parking lot, and hit a man standing in a telephone booth fifteen feet from the road. It is not surprising that the man in the phone booth filed a lawsuit. What is surprising, though, is that he sued the company responsible for the design and installation of the phone booth.
Although the lower court tossed out the case, liberal California Supreme Court Justice Rose Bird ruled against the phone company. She ruled that the risk that someone might veer off the road and crash into the telephone booth was foreseeable, therefore a jury could hold the company liable. Furthermore, she found of no consequence that the harm to the plaintiff came about because the driver was intoxicated.
In a related case, the California Supreme Court ruled that when a police officer pulls over a car for doing 85 in a 55 zone and the occupants of the care include several children who were not wearing seatbelts, and a truck on the other side of a median divider veers off the road, crosses the median, and hits the car, injuring the girls, the policeman may be sued for having made the traffic stop.
Human behavior more predictable than weather?
on
Crime Prediction
·
· Score: 1
I have a hard time believing that they can predict the liklihood that there will be a mugging on a given block with more accuracy than the liklihood that it will rain on the same block.
I had a professor who literally drank 24 cups of coffee per day. University workdays are 12-16 hours rather than 8, so this would make 1.5-2 cups per hour. Every time he walked through the lab he had a fresh cup of coffee in his hand, so I can believe it.
Conversations tended to go at the pace of cattle auctions, which was kind of fun, but when he got to age 65 or so he had to cut down to 8 cups per day because his caffeine jitters got in the way of aligning the molecular beam systems.
At least she has a degree from somewhere. A couple of years ago, the Associate Director for Lasers at Lawrence Livermore National Laboratory had his security clearance revoked and resigned his position when it came out that he had lied about having a Ph.D. in Engineering from Princeton.
This gives you a lot of faith in the kind of screening they do at the national weapons labs and at Homeland Security!
Leave it to CNet to revive Bill Joy's spurious fear of gray goo taking over the world. Get over it, Bill. To paraphrase Stephen Jay Gould, It's here, it's bacteria, get used to it!
I sure would pick rare over numerous to depict risk.
Damn straight! You are correct to do so. The total number is completely irrelevant to an individual considering an exam. The right number is the probability, which you correctly focus on. The problem you're dealing with is that the patient is likely to read about a rare adverse event in the newspaper and focus on that, just as with shark attacks at beaches.
In contrast, according to the Institute of Medicine, any time you check into a hospital in the United States, you face about a 1.5% chance of experiencing serious injury or death due to a preventable medical error, for a total of somewhere between 44,000 and 98,000 preventable deaths each year in the US.
A large fraction of these are from medication errors (about 7% of all hospital admissions have at least one medication error, but most of these do not cause serious injury or death). Compared to this, MRI risks are not worth losing sleep over.
I won't quibble with your experience. I would only point out that since so many MRIs are performed, burns can be both numerous (raw count) and rare (small fraction of total scans performed).
If the burn rate were 0.001% (This number is just pulled out of my hat to illustrate. I have no reason to believe that it has any relation to the actual frequency of burns), any given clinic could expect to perform tens of thousands of scans without seeing a single burn, but given 40 million scans performed each year there would still be hundreds of burns per year.
Slashdot Mirror
The most interesting discussion on the comments list was the material on removing export. I had always thought export was a good idea and the comments from EDG completely changed my mind.
Last year my friend had given up on getting an adequate raise at his current employer (asked many times, received little). He looked around and got a firm job offer from another outfit, went to his boss and asked whether the boss wanted to make a counter offer, and the boss gave a raise on the spot more than twice the raise my friend had imagined he could get. He turned down the other offer, stayed with his old outfit, and is much more happily (and gainfully) employed, and has a better relationship with his boss.
The key to this is that he would have been very happy to take the other job too, so this was not a bluff, nor a negotiation in bad faith.
Another friend received the following advice from his boss:
The problems I see are:
About it being valid C, of course you're right (that's why it's a warning, not an error), but as Alan Holub said about C in the title of one of his books, it gives you Enough rope to shoot yourself in the foot . If such expressions were illegal, we wouldn't need lint and relatives as much as we do.
MY compiler (Microsoft C++) does catch this
and issues a warning. Doesn't gcc?I'm glad somebody noticed! Sometimes I wonder...
Buy a copy of The Thrill of the Grill by Chris Schlesinger. Lotsa good ways to grill shrimp (as we call them on the correct side of the equator) without scorching them.
Schelinger will also explain to you why "barbecue" and "grill" are two completely different concepts. Barbecueing shrimp (cooking them for 12 hours with smokey indirect heat) would indeed be sacrilege against the gods who created these tasty protein lumps, but grilling them (cooking them briefly directly over coals) will produce delicious results.
tan(89.990) = -2.0460
tan(89.991) = -2.0408
perhaps you're thinking of
tan(1.571) = -4909.8
and
tan(1.578) = -138.8
Can you back this up with facts? I would be interested in how many millionaires are savvy businessmen and how many are stupid kids who inherited their wealth without doing squat.
Neither tend to be famous, of course, which is your larger point, but I don't see where you get off making such an assertion without justifying it.
As I understand them, the fundamentals of DNA computing largely lie in the ability of DNA to act as a quasi-nondeterministic finite automaton via controlled chemical reactions that explore the space of potential solutions with massive parallelism. When a candidate solution is found, it is verified with polynomial complexity. This reduces NP problems to P, but means that you have to be able to operate a very efficient physical/chemical sieve to select the correct answer from among a large number of candidate solutions. It also means that you must scrupulously avoid any contaminants in the pool of potential solutions (see above).
So far as I know, the President has said little or nothing about anything "nuclear." "Nucular," on the other hand, scares the pretzels out of him.
This may explain why, after fulminating on national TV about the danger of "nucular" weapons in Iraq, he held his troops back for several weeks while looters plundered the largest nuclear materials depot in Iraq, which contained over one and a half tons of enriched uranium.
This study by NASA explains why volcanic plumes, which contain tremendous amounts of chlorine, don't leave much chlorine in the strtatosphere.
Another key question is how the residence time of H2 in the stratosphere compares to the residence time of CO2 in the troposphere. If H2 has a significantly shorter residence time than the 120 year residence time of anthropogenic CO2, then it would be a good choice to switch to H2 today and then replace H2 with another alternative at a future date, since the H2 would drop back to its natural level faster than CO2 would. If H2 has a longer residence time in the stratosphere, then the best choice might be to stick with CO2 emissions.
I am also stating that this part is correct.
If it doesn't matter whether non-employees have access, then why do you advocate implementing a company policy about rogue APs and going around once a month scanning for them? That seems like a lot of effort to throw at something that "doesn't matter."
Also, just for reference, even if the prosecution succeeds, it just means that you can't lie to shareholders about vaporware. There's still nothing wrong with sowing FUD against competitors by making vaporware claims in your advertising so long as you keep them out of your stock prospectus and annual report.
This is why so many intelligent Americans end up being lawyers.
At the height of the bubble, Paul Krugman wrote that software development and other high-tech jobs were so easy to outsource that their prestige pay scales were doomed. He also believes that much lawyering will be outsourced or automated (a la Nolo Press and write-your-own-will software).
Here is a paraphrase of his argument from Proceedings of the Sixth European Assembly on Telework and New Ways of Working - Telework '99 , Aarhuss, Denmark, 22-24 September 1999:
Which one of those made your mother and father read bedtime stories to you when you were young?
What do you mean? Linux is my sex life!
I have a hard time believing that they can predict the liklihood that there will be a mugging on a given block with more accuracy than the liklihood that it will rain on the same block.
I had a professor who literally drank 24 cups of coffee per day. University workdays are 12-16 hours rather than 8, so this would make 1.5-2 cups per hour. Every time he walked through the lab he had a fresh cup of coffee in his hand, so I can believe it.
Conversations tended to go at the pace of cattle auctions, which was kind of fun, but when he got to age 65 or so he had to cut down to 8 cups per day because his caffeine jitters got in the way of aligning the molecular beam systems.
This gives you a lot of faith in the kind of screening they do at the national weapons labs and at Homeland Security!
Leave it to CNet to revive Bill Joy's spurious fear of gray goo taking over the world. Get over it, Bill. To paraphrase Stephen Jay Gould, It's here, it's bacteria, get used to it!
Damn straight! You are correct to do so. The total number is completely irrelevant to an individual considering an exam. The right number is the probability, which you correctly focus on. The problem you're dealing with is that the patient is likely to read about a rare adverse event in the newspaper and focus on that, just as with shark attacks at beaches.
In contrast, according to the Institute of Medicine, any time you check into a hospital in the United States, you face about a 1.5% chance of experiencing serious injury or death due to a preventable medical error, for a total of somewhere between 44,000 and 98,000 preventable deaths each year in the US.
A large fraction of these are from medication errors (about 7% of all hospital admissions have at least one medication error, but most of these do not cause serious injury or death). Compared to this, MRI risks are not worth losing sleep over.
If the burn rate were 0.001% (This number is just pulled out of my hat to illustrate. I have no reason to believe that it has any relation to the actual frequency of burns), any given clinic could expect to perform tens of thousands of scans without seeing a single burn, but given 40 million scans performed each year there would still be hundreds of burns per year.