Well, it's not always "optional" when you're talking about a networked device, which is pretty much everything these days.
The truth is, though, that on the Android side your milage may vary depending on the carrier and model of phone. It's a fact of life when you don't really have a choice to install whatever OS you want without going through unreasonable rigamarole -- and when I say "rigamarole" I say this as someone who downloaded Debian 0.93R5 over a 2400 baud modem onto a stack of floppies then figured out how to hand configure the X Window System.
Honesty to others, and especially honesty to self. An engineer has to be a realist in a world of wishful thinkers. He's got to work well with others, but be able to stand up to them as well. He also ought to be bold, but conscientious; sometimes taking risks but never unnecessary or sloppy ones.
Yes, that's a very good point, although calling lower skilled labor "worthless" is going a bit too far. It's not as simple as one man can do the work of ten so now I'm going to let go 90% of my work force. It's been 30 years since I took Economics, but if I recall what what you should do is dependent upon price elasticity for the product you're selling. If you can't sell any more widgets by lowering the price, then of course you'll let 90% of your workers go. But if you can sell lots more widgets with a small price reduction, productivity improvements might lead you to hire more workers.
I don't think that's likely in this case. What I think is more likely in the short term at least is replacing kitchen staff with robots will mean you replace a modest number of very cheap workers with a small number of much more highly paid workers, and whether that's a good idea depends on details we probably don't know.
That's a bit simplistic. If you expand a system, it will automatically attract more riders if the cost to the rider remains the same. And therein lay the rub. The rational basis of the decision to expand the system is a cost/benefit one, even if the system is publicly subsidized. The rationale for public subsidies is based on there being market externalities, but once you factor in those externalities the benefits have to outweigh the costs.
That's where politics comes in. Service expansion is popular, but rising fares and taxes are not. Fortunately the public has a short memory, so what you do is expand service and "pay" for it by deferring maintenance. You get the credit as a genius, and your successors get the blame. It's what economists call an "agency cost".
I had a boss once who hired a consultant and was angry when the consultant told him to do the obvious thing.
"You know what a consultant is?" he groused. "Someone who borrows your watch to tell you the time."
I thought about this for a moment. "Yeah, but what's he supposed to do if you're standing there with the watch on your wrist and you don't know what time it is?"
Some of this no doubt has to do with the real estate market in Silicon Valley. I live in a pretty high-regulation state (although we don't have to post signs saying our building may contain carcinogens), but we don't have to go through such rigamarole to get workable business premises. That's because the market is still competitive for landlords.
In an ideal world which meets the assumptions of the models we studied in Econ 101 you're unassailably right. However in the real world empirical evidence that a minimum wage results in less labor demanded is mixed. Why would that be? Clearly it must mean that wages are unnaturally lower than what the model regards as optimal.
Nor is a moderate minimum wage as undesirable as neoclassical purists suggest. Unlike those in textbooks, real labour markets are not perfectly competitive.
If the real world doesn't behave as your model predicts, it's not the real world that is wrong (unless you're Austrian School). Since in this case the model's predictive results are mixed, it makes the most sense to regard it as useful but too simplistic to be absolutely reliable.
Actually I know the answer to this one. The formal etiquette forms of address technically apply only to communications addressed to the person in question. You follow them if your are speaking or writing to that person. If you are writing about them you follow the rules prescribed by the publication's style guide, which are chosen for both clarity and to establish a consistent in-house feel.
For example, if you look at Associated Press articles on the current US president, the first reference in the article will use the president's full name ("President Barack Obama addressed the UN..."). Subsequent references to the president will simply use the president's last name ("Obama said...). The submitter uses the house style of the New York Times (and many other papers): First reference is "President Barack Obama" (or sometimes "President Obama") and subsequent references are to "Mr. Obama". This style only applies to news at the Times; opinion pieces sometimes affect different styles to show different levels of deference and formality.
They can tell a lot about you by routine purchases, especially if they look for patterns of change. I've read about the Target case; it isn't just obvious things like buying prenatal vitamins and maternity clothes; a sudden switch in preference for unscented products is common with the hormonal changes pregnant women experience.
The prediction doesn't have to be perfect to be uncannily accurate.
is that they're biased. They're written on both sides as if nothing the other side wants has any validity whatsoever. That's fundamentally dishonest, but the court's supposed to take two dishonest arguments and pick the one that is least untrue.
Let's look at this like engineers for a moment, as opposed to debating like lawyers. Let's look at the facts and implications in a balanced way.
The phone in question almost certainly contains no useful information. What the government is looking for is a precedent saying that vendors have to provide them with a way into their products that is convenient enough to be used speculatively. This is the reason for the overheated rhetoric from the FBI; they're reluctant to admit exactly what it is they're looking for, because it has obvious and enormous potential for abuse.
But while Apple is right to focus on these potential abuses, I think they've overstated their case in several respects. Mainly, I think they've overstated the technical difficulty of providing the FBI with what it wants. Does anyone really think it would take man-months to change the number of PIN tries or the timeout between them? And I think they've conflated the security of the device with the security of the user's information -- which they have no problems handing over to the FBI if it's on the user's iCloud and the FBI has a court order.
Now here's what this all suggests to my mind a reasonable compromise. Apple could furnish the FBI with the data from the device, but not the tool used to extract the data. They can maintain a secure facility on their campus where Apple personnel extract the data; they would ten furnish the data to the FBI, along with the PIN and the device, restored to the software it had when they received it. The FBI would pay all expenses incurred along with a reasonable overhead fee. Now it's true the government could steal this version of iOS. But by the same token they could steal the iOS source code and Apple's signing key. So short of a black bag job, this would provide the government with data it was legally entitled to, and only such data.
And here's what I think the FBI would make of that proposal: they wouldn't like it. Because it'd be a hell of a lot more convenient to be able to break into any phone they wanted without having to pay anything (as we all know government employee time is free).
The poor cybersecurity stance of US firms puts information that is proprietary to their Chinese trading partners as risk, and thus affects the security of the Chinese state. But what can the Chinese government do about that? Call up the US government and say, "Make those clowns get their act together!"? The US government is paralyzed by even bigger clowns.
So what you do is pick out some of the worst offenders and shake them down. Not for so much money that they go out of business -- they are after al your trading partners -- but for enough that they decide to start running their businesses like grown-ups.
This has to be one of the most enlightened uses of realpolitik in modern history.
Well, different measures tell you different things. The total installations being done is actually much more interesting to you if you're a solar manufacturer, because it tells you the size of the market - a market with uniform standards and commercial laws and language. The fraction that is of all the new energy generation being installed is also useful because it gives you some idea of how much headroom there is to expand the market.
Let's say you've got a solar technology that allows you undercut most competitors. A situation where there's enough solar being sold to make it credible to buyers and room to expand the market is very happy situation for you.
The per capita number solar installations on the other hand tells you pretty much nothing of interest. It's more of a curiosity.
The history of Federal support for research, especially into agriculture and medicine, goes well back into the 1800s, back to the 1860s establishment of land grant universities (including MIT and Cornell). The earliest research project I know of that was funded by the Federal Government was the Lewis and Clark expedition. Even if you don't count geography per se as "science" (a modern viewpoint), the expedition had significant natural history goals. Lewis and Clarke received extensive mission preparation by the American Philosophical (i.e., scientific) Society
The reason this is a good rule of thumb is that when news sources have any proof of something, they don't pose it as a question. Suppose a headline asks, "Is Donald Trump considering gender reassignment surgery?" If they had any proof that the Donald was having a sex change operation, they'd write it this way: "Donald Trump Getting Sex Change Operation!!!" The fact that they can't phrase it that way means they can't prove this and makes "no" a safe (non-refutable) answer.
Let's call this the "weak" version of Bettridge's Law: when a question is posed in a headline, "no" is a safe answer. In THIS case, however, "no" is probably a correct answer, because the question itself entails a sloppy generalization. It's equivalent to asking the question, "Are there NO situations in which a paper might reasonably be non-free?" All it takes is finding a single example to say "no", e.g., an entirely privately funded research paper intended for that organization's internal consumption.
If you qualified the scope of the question, the answer might plausibly be "yes", e.g. "Should All Publicly Funded Research Papers Not Involving Classified Materials be Free?" That, however, presumes more familiarity with logic than most editors evidently have.
Well, it's not so simple as that. What you're really interested in is the interaction of radioactive materials with a complex biological entity -- your body. The rate and type of decay is only one parameter in that interaction; the chemistry and physical form also matter, along with the chemistry and stability of the daughter products.
For example strontium is in the same periodic table column as calcium -- which obviously is a major component of our bodies. Therefore strontium is digested and metabolized the same way calcium is. This make Strontium-90 a much higher concern than other isotopes with similar half lives (roughly 29 years). An intact block of metallic strontium is moderately hazardous. An aerosol suspension of colloidal Sr-90 particles is extremely hazardous.
The Radium 226 in old watch pigments has a half-life of 1600 years, and is perfectly safe to wear inside a sealed case on your wrist. But you don't want to ingest it. It's probably best to avoid working on old radium watches because the pigment breaks down into a very fine powder. Would I panic if I had a single exposure to an opened radium watch? No. I just wouldn't make a habit of it.
The "duck-and-cover" era advice about avoiding atomic fallout tries to balance survival priorities against each other. You're supposed to stay in your shelter for several weeks, which allows the levels of the most radioactive isotopes to fall. But the reason you come out after several weeks is not that it's perfectly safe to do so; it's that you can't live for years or even decades in a shelter. So the compromise is to stay in the shelter long only enough to avoid dying quickly of acute radiation sickness. After two or three weeks the levels of highly radioactive Sr-91 and Sr-92 are negligible; the levels of Sr-90 are hazardous and will remain so for decades.
Depending on the degree, form, and nature of the contamination of these ships, it could prove a serious handicap to their ongoing operation. Not because the sailors will come down with acute radiation sickness, but because of the laborious precautions needed to avoid chronically exposing sailors. TFA doesn't say much about specifics, but I do know Fukushima released a great deal of Cesium-137, which has a half-life of 30 years and is in the same periodic table column as potassium, which is obviously biologically very active. It's also readily water soluble and can enter the body that way. Less Sr-90 was released, but depending on exactly where and when the ships were contaminated that could also pose an operational handicap because calcium remains in the body much longer than potassium.
As for Hiroshima and Nagasaki, it's important to realize that they were high altitude detonations. That both reduced the levels of contamination and spread the contamination widely throughout the region. It'd serve little purpose to abandon the city centers. Compared to, say, Chernobyl with its burning radioactive graphite, the detonations were relatively clean radiological disasters.
Oh for Pete's sake, get some Pepto-Bismol. If you're worried about the millions of dollars that went into this, look at the pictures in TFA. It's a student project. An Indian student project.
It's a mild steel bike frame with bare circuit boards glued to pieces of cardboard attacked to the frame with twist ties. Which to my mind makes this much cooler: it's a hack. The "balancing mechanism"? Training wheels. Which is good engineering; don't waste time and money on desirable but complicated stuff until you've ironed to the most important bits.
And the thing is the real hotheads among his supporters crave the catharsis of a confrontation. Being declared "war" against suits them right down to the ground.
Unless you subscribe to the theory that by making all firearms expensive and difficult to obtain or own (unless you are rich or politically connected) is worthwhile because it also happens to deny access for "crazies", then that is not Daniel Webster's goal.
Slashdot Mirror
Well, more like taking chunks of other countries' colonies...
Well, it's not always "optional" when you're talking about a networked device, which is pretty much everything these days.
The truth is, though, that on the Android side your milage may vary depending on the carrier and model of phone. It's a fact of life when you don't really have a choice to install whatever OS you want without going through unreasonable rigamarole -- and when I say "rigamarole" I say this as someone who downloaded Debian 0.93R5 over a 2400 baud modem onto a stack of floppies then figured out how to hand configure the X Window System.
I'm assuming people don't want to hire sociopaths.
Honesty to others, and especially honesty to self. An engineer has to be a realist in a world of wishful thinkers. He's got to work well with others, but be able to stand up to them as well. He also ought to be bold, but conscientious; sometimes taking risks but never unnecessary or sloppy ones.
Yes, that's a very good point, although calling lower skilled labor "worthless" is going a bit too far. It's not as simple as one man can do the work of ten so now I'm going to let go 90% of my work force. It's been 30 years since I took Economics, but if I recall what what you should do is dependent upon price elasticity for the product you're selling. If you can't sell any more widgets by lowering the price, then of course you'll let 90% of your workers go. But if you can sell lots more widgets with a small price reduction, productivity improvements might lead you to hire more workers.
I don't think that's likely in this case. What I think is more likely in the short term at least is replacing kitchen staff with robots will mean you replace a modest number of very cheap workers with a small number of much more highly paid workers, and whether that's a good idea depends on details we probably don't know.
That's a bit simplistic. If you expand a system, it will automatically attract more riders if the cost to the rider remains the same. And therein lay the rub. The rational basis of the decision to expand the system is a cost/benefit one, even if the system is publicly subsidized. The rationale for public subsidies is based on there being market externalities, but once you factor in those externalities the benefits have to outweigh the costs.
That's where politics comes in. Service expansion is popular, but rising fares and taxes are not. Fortunately the public has a short memory, so what you do is expand service and "pay" for it by deferring maintenance. You get the credit as a genius, and your successors get the blame. It's what economists call an "agency cost".
I had a boss once who hired a consultant and was angry when the consultant told him to do the obvious thing.
"You know what a consultant is?" he groused. "Someone who borrows your watch to tell you the time."
I thought about this for a moment. "Yeah, but what's he supposed to do if you're standing there with the watch on your wrist and you don't know what time it is?"
Some of this no doubt has to do with the real estate market in Silicon Valley. I live in a pretty high-regulation state (although we don't have to post signs saying our building may contain carcinogens), but we don't have to go through such rigamarole to get workable business premises. That's because the market is still competitive for landlords.
In an ideal world which meets the assumptions of the models we studied in Econ 101 you're unassailably right. However in the real world empirical evidence that a minimum wage results in less labor demanded is mixed. Why would that be? Clearly it must mean that wages are unnaturally lower than what the model regards as optimal.
As the Economist notes:
If the real world doesn't behave as your model predicts, it's not the real world that is wrong (unless you're Austrian School). Since in this case the model's predictive results are mixed, it makes the most sense to regard it as useful but too simplistic to be absolutely reliable.
Exactly right. Automation increases the marginal productivity of labor, and therefore the value of labor.
Or vice versa.
That is true. But Target actually hired a mathematician to design the program and measure its impact. They have the data to show it works quite well.
Actually I know the answer to this one. The formal etiquette forms of address technically apply only to communications addressed to the person in question. You follow them if your are speaking or writing to that person. If you are writing about them you follow the rules prescribed by the publication's style guide, which are chosen for both clarity and to establish a consistent in-house feel.
For example, if you look at Associated Press articles on the current US president, the first reference in the article will use the president's full name ("President Barack Obama addressed the UN..."). Subsequent references to the president will simply use the president's last name ("Obama said...). The submitter uses the house style of the New York Times (and many other papers): First reference is "President Barack Obama" (or sometimes "President Obama") and subsequent references are to "Mr. Obama". This style only applies to news at the Times; opinion pieces sometimes affect different styles to show different levels of deference and formality.
They can tell a lot about you by routine purchases, especially if they look for patterns of change. I've read about the Target case; it isn't just obvious things like buying prenatal vitamins and maternity clothes; a sudden switch in preference for unscented products is common with the hormonal changes pregnant women experience.
The prediction doesn't have to be perfect to be uncannily accurate.
is that they're biased. They're written on both sides as if nothing the other side wants has any validity whatsoever. That's fundamentally dishonest, but the court's supposed to take two dishonest arguments and pick the one that is least untrue.
Let's look at this like engineers for a moment, as opposed to debating like lawyers. Let's look at the facts and implications in a balanced way.
The phone in question almost certainly contains no useful information. What the government is looking for is a precedent saying that vendors have to provide them with a way into their products that is convenient enough to be used speculatively. This is the reason for the overheated rhetoric from the FBI; they're reluctant to admit exactly what it is they're looking for, because it has obvious and enormous potential for abuse.
But while Apple is right to focus on these potential abuses, I think they've overstated their case in several respects. Mainly, I think they've overstated the technical difficulty of providing the FBI with what it wants. Does anyone really think it would take man-months to change the number of PIN tries or the timeout between them? And I think they've conflated the security of the device with the security of the user's information -- which they have no problems handing over to the FBI if it's on the user's iCloud and the FBI has a court order.
Now here's what this all suggests to my mind a reasonable compromise. Apple could furnish the FBI with the data from the device, but not the tool used to extract the data. They can maintain a secure facility on their campus where Apple personnel extract the data; they would ten furnish the data to the FBI, along with the PIN and the device, restored to the software it had when they received it. The FBI would pay all expenses incurred along with a reasonable overhead fee. Now it's true the government could steal this version of iOS. But by the same token they could steal the iOS source code and Apple's signing key. So short of a black bag job, this would provide the government with data it was legally entitled to, and only such data.
And here's what I think the FBI would make of that proposal: they wouldn't like it. Because it'd be a hell of a lot more convenient to be able to break into any phone they wanted without having to pay anything (as we all know government employee time is free).
Oh I get it,
evidently not.
The poor cybersecurity stance of US firms puts information that is proprietary to their Chinese trading partners as risk, and thus affects the security of the Chinese state. But what can the Chinese government do about that? Call up the US government and say, "Make those clowns get their act together!"? The US government is paralyzed by even bigger clowns.
So what you do is pick out some of the worst offenders and shake them down. Not for so much money that they go out of business -- they are after al your trading partners -- but for enough that they decide to start running their businesses like grown-ups.
This has to be one of the most enlightened uses of realpolitik in modern history.
Well, different measures tell you different things. The total installations being done is actually much more interesting to you if you're a solar manufacturer, because it tells you the size of the market - a market with uniform standards and commercial laws and language. The fraction that is of all the new energy generation being installed is also useful because it gives you some idea of how much headroom there is to expand the market.
Let's say you've got a solar technology that allows you undercut most competitors. A situation where there's enough solar being sold to make it credible to buyers and room to expand the market is very happy situation for you.
The per capita number solar installations on the other hand tells you pretty much nothing of interest. It's more of a curiosity.
The history of Federal support for research, especially into agriculture and medicine, goes well back into the 1800s, back to the 1860s establishment of land grant universities (including MIT and Cornell). The earliest research project I know of that was funded by the Federal Government was the Lewis and Clark expedition. Even if you don't count geography per se as "science" (a modern viewpoint), the expedition had significant natural history goals. Lewis and Clarke received extensive mission preparation by the American Philosophical (i.e., scientific) Society
The reason this is a good rule of thumb is that when news sources have any proof of something, they don't pose it as a question. Suppose a headline asks, "Is Donald Trump considering gender reassignment surgery?" If they had any proof that the Donald was having a sex change operation, they'd write it this way: "Donald Trump Getting Sex Change Operation!!!" The fact that they can't phrase it that way means they can't prove this and makes "no" a safe (non-refutable) answer.
Let's call this the "weak" version of Bettridge's Law: when a question is posed in a headline, "no" is a safe answer. In THIS case, however, "no" is probably a correct answer, because the question itself entails a sloppy generalization. It's equivalent to asking the question, "Are there NO situations in which a paper might reasonably be non-free?" All it takes is finding a single example to say "no", e.g., an entirely privately funded research paper intended for that organization's internal consumption.
If you qualified the scope of the question, the answer might plausibly be "yes", e.g. "Should All Publicly Funded Research Papers Not Involving Classified Materials be Free?" That, however, presumes more familiarity with logic than most editors evidently have.
Well, it's not so simple as that. What you're really interested in is the interaction of radioactive materials with a complex biological entity -- your body. The rate and type of decay is only one parameter in that interaction; the chemistry and physical form also matter, along with the chemistry and stability of the daughter products.
For example strontium is in the same periodic table column as calcium -- which obviously is a major component of our bodies. Therefore strontium is digested and metabolized the same way calcium is. This make Strontium-90 a much higher concern than other isotopes with similar half lives (roughly 29 years). An intact block of metallic strontium is moderately hazardous. An aerosol suspension of colloidal Sr-90 particles is extremely hazardous.
The Radium 226 in old watch pigments has a half-life of 1600 years, and is perfectly safe to wear inside a sealed case on your wrist. But you don't want to ingest it. It's probably best to avoid working on old radium watches because the pigment breaks down into a very fine powder. Would I panic if I had a single exposure to an opened radium watch? No. I just wouldn't make a habit of it.
The "duck-and-cover" era advice about avoiding atomic fallout tries to balance survival priorities against each other. You're supposed to stay in your shelter for several weeks, which allows the levels of the most radioactive isotopes to fall. But the reason you come out after several weeks is not that it's perfectly safe to do so; it's that you can't live for years or even decades in a shelter. So the compromise is to stay in the shelter long only enough to avoid dying quickly of acute radiation sickness. After two or three weeks the levels of highly radioactive Sr-91 and Sr-92 are negligible; the levels of Sr-90 are hazardous and will remain so for decades.
Depending on the degree, form, and nature of the contamination of these ships, it could prove a serious handicap to their ongoing operation. Not because the sailors will come down with acute radiation sickness, but because of the laborious precautions needed to avoid chronically exposing sailors. TFA doesn't say much about specifics, but I do know Fukushima released a great deal of Cesium-137, which has a half-life of 30 years and is in the same periodic table column as potassium, which is obviously biologically very active. It's also readily water soluble and can enter the body that way. Less Sr-90 was released, but depending on exactly where and when the ships were contaminated that could also pose an operational handicap because calcium remains in the body much longer than potassium.
As for Hiroshima and Nagasaki, it's important to realize that they were high altitude detonations. That both reduced the levels of contamination and spread the contamination widely throughout the region. It'd serve little purpose to abandon the city centers. Compared to, say, Chernobyl with its burning radioactive graphite, the detonations were relatively clean radiological disasters.
Oh for Pete's sake, get some Pepto-Bismol. If you're worried about the millions of dollars that went into this, look at the pictures in TFA. It's a student project. An Indian student project.
It's a mild steel bike frame with bare circuit boards glued to pieces of cardboard attacked to the frame with twist ties. Which to my mind makes this much cooler: it's a hack. The "balancing mechanism"? Training wheels. Which is good engineering; don't waste time and money on desirable but complicated stuff until you've ironed to the most important bits.
And the thing is the real hotheads among his supporters crave the catharsis of a confrontation. Being declared "war" against suits them right down to the ground.
Give me a lever, and a place to stand...
Unless you subscribe to the theory that by making all firearms expensive and difficult to obtain or own (unless you are rich or politically connected) is worthwhile because it also happens to deny access for "crazies", then that is not Daniel Webster's goal.
Textbook straw man argument.