I think data backup is very important. What is even more important is making sure you can do recoveries from the backups.
Absolutely. If you haven't tested to make sure you can actually recover your data when you need it, you might discover you've been wasting your time, and at the worst possible moment. And, of course, that testing of your recovery procedure needs to include integrity testing so you have confidence the data you're recovering is what it's supposed to be.
The big one I've seen with my current employer is that they've failed to expand their IT staff as the organization as a whole has expanded. The predictable result is that nothing but the most urgent requests gets handled promptly, and minor problems fester indefinitely.
Another important point is that even if you assume internet addiction is a thing and they have good tests for it, they still haven't proven the direction of causality. It's also possible that the addicts' brains were different to start with and those differences made they susceptible to addiction. It's a classic cum hoc, ergo propter hoc fallacy.
Parody is actually a form of fair use; it's legally considered a form of criticism, which is one of the things fair use is intended to protect. Fair use is actually a very complicated legal issue that has to be decided on the particulars of each case. It depends on a balance of four different factors (by statute):
1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;
2) the nature of the copyrighted work;
3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and
4) the effect of the use upon the potential market for or value of the copyrighted work.
So commercial use definitely weighs against a ruling of fair use, but isn't an absolute factor. Courts have worked with these principles and like to talk about "transformative" nature of the work that's doing the copying. If the new work is something radically different from the original, it's far more likely to be considered fair use.
My impression is that Google won its case by arguing that they were only showing selected snippets of the works they had scanned (helps on point 3) and that it should actually help the potential market for the copyrighted works they were copying (helps on point 4). The idea is that by showing only short snippets, they were making people familiar with the original work while not providing them with enough of it to be really useful. That should make people more, not less, likely to buy the original.
Your pessimism is misplaced. The whole reason Allergan pulled this maneuver is to get around a new review process that Congress passed quite recently. That seems like very solid evidence Congress is not completely in thrall to patent holders.
So they stuck them in a unregulated bank from some guy no one has heard of and the bank was robbed. Blame the currency!
They aren't blaming it on the currency; they're blaming it on the complete lack of any kind of regulation surrounding the currency. The people who advocate Bitcoin (and similar cryptocurrency) wanted to create a whole new economy without all that pesky government regulation. Now they're learning the hard way how important all that regulation is to having a functioning economy.
It was a vaguely reasonable concept. The very best way of making vegetable juice is with a two stage process. Stage one is grinding the vegetables, and stage two is pressing them to get the juice out. Done properly, it gets out more juice than centrifugal or masticating juicers, and the juice is supposed to be of higher quality, with more enzyme activity and higher vitamin and mineral content. You can effectively get out all the nutrients and leave the fiber behind*.
The problem is that two stage juicing is way more involved than centrifugal or masticating juicers. Grinding the vegetables tends to be messy, pressing involves bags or cloths to keep the pulp from going out with the juice, and there's a ton of cleanup. Juicero promised to give the better quality juice you get with two-stage juicers without the hassle and mess. For a wealthy but busy person, the price of the juicer and juice packs could be justified by the ease of use.
*The people who are really serious about this understand that fiber is a necessary part of the diet, but there's so much fiber in most vegetables, especially greens, that it limits how much you can eat. Vegetable juice is supposed to serve as a supplement to eating whole vegetables. That way you can get the fiber together with several times as much nutrition as you could get if you limited yourself to whole vegetables.
This is incorrect. There's a rule in chess that the game is a draw if the same position is reached three times. Since there are a finite number of possible positions and a you're only allowed to visit each of those positions a finite number of times in a game, there must also be a finite number of games.
I think the ideal application is something like the regular routes here in the Los Angeles area. There are a huge number of truckers who go back and forth between the port complex and major logistics centers in the Inland Empire. The round-trip is well under the 200-300 mile range of these trucks, and it involves a lot of travel through highly congested areas where electric vehicles have an inherent efficiency advantage. Perhaps more important, the ports have a long-standing problem with diesel exhaust pollution, and they're talking about phasing out diesel trucks for electrics to help deal with it. This is going to make that practical rather than pie-in-the-sky.
I think you're missing the importance of this point:
(and other amphipathic, semi-volatile compounds)
The authors of the study looked at guaiacol specifically, but their intention is to use it as a typical example of a large class of similar compounds that are important in the flavor of more kinds of whiskey than just Scotch. This kind of thing- studying one compound as an example of an entire class- is common when it just isn't practical to study every member of a class. They clearly can't study every flavor molecule in a mixture as complex as whiskey, so they pick one that's typical and extrapolate the results to similar molecules. Studying an intractably complex system by considering a simpler model is an essential part of how real science works.
It's an example of the principle that possession is 9/10 of the law. There was an easement, but once he physically blocked it the case had to work its way through the courts to force him to open it up again. An asshole willing to spend a lot of money on legal fees can keep a case like that tied up in the courts for a very long time.
From reading the description, the difference between "discouraged" and "marginally attached" workers is pretty thin. "Discouraged" workers are ones who say specifically that they don't think they can find work as a reason for not looking, while "other marginally attacked" workers are other people who would like work but aren't looking for some other reason. I guess that might include seasonal workers during the off-season; they could look for a different job but have decided not to bother. It looks to me as if the difference between U4 and U5 is pretty small in practice, and U5 is not massively higher than U3. The big gap right now is between U5 and U6, and that's a gap that varies a lot over the course of a business cycle.
The big question is what you want to measure. The unemployment rate (as conventionally defined) is trying to measure what fraction of people who want jobs can't find them. The author of the cited article is correct that the official unemployment rate is leaving out some people who probably ought to be counted, like people who have given up looking for work. This can be really important, because bad data may cause economists to recommend bad policy. In the USA during the 1990s, for instance, sustained low official unemployment wound up encouraging "hard core" unemployed people who were left out of the official statistics to start looking for work. That meant low unemployment didn't cause inflation to take off the way economists predicted. A different measure of unemployment that made fewer assumptions about who was employable might have prevented them from making that mistake.
That said, there are problems with the author's proposal of including everyone between 16 and 64 as the pool of potential workers. The economy has changed over time in ways that systematically change who is likely to look for work. Higher education is far more important than it used to be, so that college age people probably shouldn't be looking for full-time work, and 16 to 18 year olds certainly shouldn't be. At the same time, though, there are fewer stay-at-home parents, which increases the expected size of the workforce. That means using the entire 16-64 year old population as the potential workforce will make comparisons to historical data much less useful, which also undermines the value of the data for policy decisions.
Probably the best solution is to give up on the idea of capturing the state of employment in a single number. The US government, for instance, calculates no fewer than 6 versions of the unemployment rate and a "labor participation rate" that is closer to the kind of calculation the original author wants. One of those rates is the official unemployment rate, but it can be compared against other rates to see if they're changing in sync. A common comparison is between U3 (the official rate) and U6 (which counts part time workers who would like to work full time as unemployed and includes people who have given up looking for work as part of the potential workforce). U3 is what has traditionally been used to measure unemployment, but U6 probably gives a better idea of how much real slack there is in the labor force.
It's not clear that it will ever be necessary to force states to adopt self-driving vehicles, either. If their concerns are actually addressed (this is a "union", right?) then it should be possible to get them on board.
States rights should not apply to rules for things like car manufacturing, and certainly not to long-haul trucking. The US Constitution specifically grants Congress the power to regulate interstate commerce, and vehicles that are capable of, and expressly designed to, travel from state to state are exactly the kind of thing the framers had in mind when they gave the federal government that power. Having uniform rules so manufacturers don't have to make 50 different models of car- and so people driving from state to state don't have problems because their car is illegal in the next state over- is the whole point of having a federal government with the power to regulate interstate commerce.
Too many hours and you don't produce quality of work. Studies have shown extra vacation and time away from the office INCREASE productivity.
That's true, but it's also true that kind of study necessarily focuses on average people. There are probably some people who can work longer house and remain productive, It's possible, maybe even likely, that part of the secret to people like Musk and Zuckerberg succeeding is that they have an exceptional ability to stay productive while working long hours. IOW, those guys may be successful in part because they put in incredibly long hours, but it doesn't mean other people will be able to get the same results by doing the same thing, even ignoring the importance of luck, being in the right place at the right time, etc.
It seems to me that the confusion is because dictionaries have both functions. Yes, they've always been intended to be descriptive, providing a guide to the language as people actually use it. But they've also always had some tendency to be prescriptive, telling people what the accepted correct usage is. A good example is with word pairs like imply/infer or compose/comprise, where there's a "standard" usage that gives each word a distinct and non-overlapping meaning and a common usage where they can be used interchangeably for some meanings. A good dictionary will list the overlapping meaning but provide a note saying that it's non-standard usage.
I disagree. In both cases, the problem is not the framework (or standard), it's the blind trust in it and the misconception that it's going to make you deliver higher quality.
The big problem with adopting quality frameworks* is that people adopt them to check a checkbox without understanding how they are supposed to work. Lousy but reproducible work is the result of doing the bare minimum to get certification. Unfortunately, that bare minimum is still a lot of effort because you have to document all your processes and keep records of your work. The real value comes from analyzing those painstakingly kept records to figure out where your problems are and updating your procedures to try to fix them.
I think this kind of checkbox compliance is why so many people hate quality frameworks. They go through a lot of trouble to get that checkbox, but because they only do the minimum the checkbox is all they get, and it's not a good return on their effort. It's only by moving on to continuous process improvement that the effort really pays off in improved quality.
*My experience is with cGMP for regulated drug manufacturing, but AFAIK most quality frameworks have the same general approach and outlook.
Maybe systemd has won the day, but that's no reason to stop people from working on a systemd-free system if that's what they want to do. Maybe systemd will turn out to be the disaster the naysayers were predicting and we'll all be happy they didn't give up. More likely, it will remain a hobby project for a handful of people who are resisting change for the sake of resisting change.
Ultimately, though, that's their choice. When systemd really started taking over, one of the regular comments was that people who didn't like it were free to fork their own distributions that didn't use it. Nobody who said that back then should complain because somebody took them seriously. As long as they aren't actively interfering with anyone else, they should be free to pursue their interests. Real freedom of choice includes the freedom to make unpopular choices.
The whole point of the lawsuit is about one company stealing proprietary information from another. We undo the whole purpose of the suit if the court turns around and allows that information to leak out by making absolutely everything about the case public.
Their objective measurement is only as good as the algorithm they're using and the data they're feeding in. The data quality can vary dramatically- I've seen houses where I know the data is good and others where it's woefully inaccurate. It's also important to realize that inaccurate data for other people's houses can throw off the valuation for your house with accurate data, since it corrupts their whole database. And there's a big worry that their algorithms are pretty lousy.
The reality is that the entire market is comprised of numbers pulled out of the air.
Not really. Sales prices are based on actual negotiations between real people who have real money on the line, very often for by far the largest purchase of their lifetime. Those sales prices have a strong influence on sellers' asking prices. In places where there's still substantial new home construction, they're also constrained by actual land and construction prices. There's still some arbitrariness to the pricing, but not particularly more than any other market.
Retaliation for various kinds of employee whistleblowing, including complaints about sexual harassment, is illegal. If they really were electronically stalking her as retaliation for her sexual harassment complaint, that would be grounds for a lawsuit even if the same actions would otherwise be perfectly legal.
Slashdot Mirror
Absolutely. If you haven't tested to make sure you can actually recover your data when you need it, you might discover you've been wasting your time, and at the worst possible moment. And, of course, that testing of your recovery procedure needs to include integrity testing so you have confidence the data you're recovering is what it's supposed to be.
The big one I've seen with my current employer is that they've failed to expand their IT staff as the organization as a whole has expanded. The predictable result is that nothing but the most urgent requests gets handled promptly, and minor problems fester indefinitely.
Another important point is that even if you assume internet addiction is a thing and they have good tests for it, they still haven't proven the direction of causality. It's also possible that the addicts' brains were different to start with and those differences made they susceptible to addiction. It's a classic cum hoc, ergo propter hoc fallacy.
Parody is actually a form of fair use; it's legally considered a form of criticism, which is one of the things fair use is intended to protect. Fair use is actually a very complicated legal issue that has to be decided on the particulars of each case. It depends on a balance of four different factors (by statute):
So commercial use definitely weighs against a ruling of fair use, but isn't an absolute factor. Courts have worked with these principles and like to talk about "transformative" nature of the work that's doing the copying. If the new work is something radically different from the original, it's far more likely to be considered fair use.
My impression is that Google won its case by arguing that they were only showing selected snippets of the works they had scanned (helps on point 3) and that it should actually help the potential market for the copyrighted works they were copying (helps on point 4). The idea is that by showing only short snippets, they were making people familiar with the original work while not providing them with enough of it to be really useful. That should make people more, not less, likely to buy the original.
Your pessimism is misplaced. The whole reason Allergan pulled this maneuver is to get around a new review process that Congress passed quite recently. That seems like very solid evidence Congress is not completely in thrall to patent holders.
The aren't really crazy; they're just willfully ignorant.
They aren't blaming it on the currency; they're blaming it on the complete lack of any kind of regulation surrounding the currency. The people who advocate Bitcoin (and similar cryptocurrency) wanted to create a whole new economy without all that pesky government regulation. Now they're learning the hard way how important all that regulation is to having a functioning economy.
Artificial intelligence won't produce Slashdot stories. For that we need Artificial Stupidity.
It was a vaguely reasonable concept. The very best way of making vegetable juice is with a two stage process. Stage one is grinding the vegetables, and stage two is pressing them to get the juice out. Done properly, it gets out more juice than centrifugal or masticating juicers, and the juice is supposed to be of higher quality, with more enzyme activity and higher vitamin and mineral content. You can effectively get out all the nutrients and leave the fiber behind*.
The problem is that two stage juicing is way more involved than centrifugal or masticating juicers. Grinding the vegetables tends to be messy, pressing involves bags or cloths to keep the pulp from going out with the juice, and there's a ton of cleanup. Juicero promised to give the better quality juice you get with two-stage juicers without the hassle and mess. For a wealthy but busy person, the price of the juicer and juice packs could be justified by the ease of use.
*The people who are really serious about this understand that fiber is a necessary part of the diet, but there's so much fiber in most vegetables, especially greens, that it limits how much you can eat. Vegetable juice is supposed to serve as a supplement to eating whole vegetables. That way you can get the fiber together with several times as much nutrition as you could get if you limited yourself to whole vegetables.
This is incorrect. There's a rule in chess that the game is a draw if the same position is reached three times. Since there are a finite number of possible positions and a you're only allowed to visit each of those positions a finite number of times in a game, there must also be a finite number of games.
I think the ideal application is something like the regular routes here in the Los Angeles area. There are a huge number of truckers who go back and forth between the port complex and major logistics centers in the Inland Empire. The round-trip is well under the 200-300 mile range of these trucks, and it involves a lot of travel through highly congested areas where electric vehicles have an inherent efficiency advantage. Perhaps more important, the ports have a long-standing problem with diesel exhaust pollution, and they're talking about phasing out diesel trucks for electrics to help deal with it. This is going to make that practical rather than pie-in-the-sky.
I think you're missing the importance of this point:
The authors of the study looked at guaiacol specifically, but their intention is to use it as a typical example of a large class of similar compounds that are important in the flavor of more kinds of whiskey than just Scotch. This kind of thing- studying one compound as an example of an entire class- is common when it just isn't practical to study every member of a class. They clearly can't study every flavor molecule in a mixture as complex as whiskey, so they pick one that's typical and extrapolate the results to similar molecules. Studying an intractably complex system by considering a simpler model is an essential part of how real science works.
It's an example of the principle that possession is 9/10 of the law. There was an easement, but once he physically blocked it the case had to work its way through the courts to force him to open it up again. An asshole willing to spend a lot of money on legal fees can keep a case like that tied up in the courts for a very long time.
From reading the description, the difference between "discouraged" and "marginally attached" workers is pretty thin. "Discouraged" workers are ones who say specifically that they don't think they can find work as a reason for not looking, while "other marginally attacked" workers are other people who would like work but aren't looking for some other reason. I guess that might include seasonal workers during the off-season; they could look for a different job but have decided not to bother. It looks to me as if the difference between U4 and U5 is pretty small in practice, and U5 is not massively higher than U3. The big gap right now is between U5 and U6, and that's a gap that varies a lot over the course of a business cycle.
The big question is what you want to measure. The unemployment rate (as conventionally defined) is trying to measure what fraction of people who want jobs can't find them. The author of the cited article is correct that the official unemployment rate is leaving out some people who probably ought to be counted, like people who have given up looking for work. This can be really important, because bad data may cause economists to recommend bad policy. In the USA during the 1990s, for instance, sustained low official unemployment wound up encouraging "hard core" unemployed people who were left out of the official statistics to start looking for work. That meant low unemployment didn't cause inflation to take off the way economists predicted. A different measure of unemployment that made fewer assumptions about who was employable might have prevented them from making that mistake.
That said, there are problems with the author's proposal of including everyone between 16 and 64 as the pool of potential workers. The economy has changed over time in ways that systematically change who is likely to look for work. Higher education is far more important than it used to be, so that college age people probably shouldn't be looking for full-time work, and 16 to 18 year olds certainly shouldn't be. At the same time, though, there are fewer stay-at-home parents, which increases the expected size of the workforce. That means using the entire 16-64 year old population as the potential workforce will make comparisons to historical data much less useful, which also undermines the value of the data for policy decisions.
Probably the best solution is to give up on the idea of capturing the state of employment in a single number. The US government, for instance, calculates no fewer than 6 versions of the unemployment rate and a "labor participation rate" that is closer to the kind of calculation the original author wants. One of those rates is the official unemployment rate, but it can be compared against other rates to see if they're changing in sync. A common comparison is between U3 (the official rate) and U6 (which counts part time workers who would like to work full time as unemployed and includes people who have given up looking for work as part of the potential workforce). U3 is what has traditionally been used to measure unemployment, but U6 probably gives a better idea of how much real slack there is in the labor force.
States rights should not apply to rules for things like car manufacturing, and certainly not to long-haul trucking. The US Constitution specifically grants Congress the power to regulate interstate commerce, and vehicles that are capable of, and expressly designed to, travel from state to state are exactly the kind of thing the framers had in mind when they gave the federal government that power. Having uniform rules so manufacturers don't have to make 50 different models of car- and so people driving from state to state don't have problems because their car is illegal in the next state over- is the whole point of having a federal government with the power to regulate interstate commerce.
That's true, but it's also true that kind of study necessarily focuses on average people. There are probably some people who can work longer house and remain productive, It's possible, maybe even likely, that part of the secret to people like Musk and Zuckerberg succeeding is that they have an exceptional ability to stay productive while working long hours. IOW, those guys may be successful in part because they put in incredibly long hours, but it doesn't mean other people will be able to get the same results by doing the same thing, even ignoring the importance of luck, being in the right place at the right time, etc.
It seems to me that the confusion is because dictionaries have both functions. Yes, they've always been intended to be descriptive, providing a guide to the language as people actually use it. But they've also always had some tendency to be prescriptive, telling people what the accepted correct usage is. A good example is with word pairs like imply/infer or compose/comprise, where there's a "standard" usage that gives each word a distinct and non-overlapping meaning and a common usage where they can be used interchangeably for some meanings. A good dictionary will list the overlapping meaning but provide a note saying that it's non-standard usage.
The big problem with adopting quality frameworks* is that people adopt them to check a checkbox without understanding how they are supposed to work. Lousy but reproducible work is the result of doing the bare minimum to get certification. Unfortunately, that bare minimum is still a lot of effort because you have to document all your processes and keep records of your work. The real value comes from analyzing those painstakingly kept records to figure out where your problems are and updating your procedures to try to fix them.
I think this kind of checkbox compliance is why so many people hate quality frameworks. They go through a lot of trouble to get that checkbox, but because they only do the minimum the checkbox is all they get, and it's not a good return on their effort. It's only by moving on to continuous process improvement that the effort really pays off in improved quality.
*My experience is with cGMP for regulated drug manufacturing, but AFAIK most quality frameworks have the same general approach and outlook.
Maybe systemd has won the day, but that's no reason to stop people from working on a systemd-free system if that's what they want to do. Maybe systemd will turn out to be the disaster the naysayers were predicting and we'll all be happy they didn't give up. More likely, it will remain a hobby project for a handful of people who are resisting change for the sake of resisting change.
Ultimately, though, that's their choice. When systemd really started taking over, one of the regular comments was that people who didn't like it were free to fork their own distributions that didn't use it. Nobody who said that back then should complain because somebody took them seriously. As long as they aren't actively interfering with anyone else, they should be free to pursue their interests. Real freedom of choice includes the freedom to make unpopular choices.
The whole point of the lawsuit is about one company stealing proprietary information from another. We undo the whole purpose of the suit if the court turns around and allows that information to leak out by making absolutely everything about the case public.
Their objective measurement is only as good as the algorithm they're using and the data they're feeding in. The data quality can vary dramatically- I've seen houses where I know the data is good and others where it's woefully inaccurate. It's also important to realize that inaccurate data for other people's houses can throw off the valuation for your house with accurate data, since it corrupts their whole database. And there's a big worry that their algorithms are pretty lousy.
Not really. Sales prices are based on actual negotiations between real people who have real money on the line, very often for by far the largest purchase of their lifetime. Those sales prices have a strong influence on sellers' asking prices. In places where there's still substantial new home construction, they're also constrained by actual land and construction prices. There's still some arbitrariness to the pricing, but not particularly more than any other market.
Retaliation for various kinds of employee whistleblowing, including complaints about sexual harassment, is illegal. If they really were electronically stalking her as retaliation for her sexual harassment complaint, that would be grounds for a lawsuit even if the same actions would otherwise be perfectly legal.
Zero minutes in the dryer will leave many dogs pissing on the floor, eating shoes, etc.