I hope if that man ends up on American soil that the citizens of this country (US) riot and raise fucking hell. What our government plans to do is wrong, it's illegal, and they know it. So does every citizen and every member of the press.
How many cases can you list where any significant number of US citizens raised such a fuss in defense of someone?
I'm not saying it couldn't happen; I'd just be interested in the precedent(s) that would give you such hope.
Ahh. So your assertion is that if it is "given" to him, as opposed to his asking for it, that's not espionage?
Hey, how 'bout you post your address here, so that a few of us readers can "give" you things that will make you guilty of assorted crimes. I don't actually have anything like that in my posession right now (that I know of), but I'm sure I could make something up given a bit of time.
Hmmm... I think you mean you can run a Windows Mobile phone about a week at idle before it gets cut off by a 500MB cap. If you actually use it for what it's advertised for, you'll hit that cap in a much shorter time. If you visit any site that sends a video ad, you might hit the cap in a single day.
They have managed to turn the free airways into the most expensive form of communication ever.
You should google the phrase "whatever the market will bear". It's the primary explanation of most corporate behavior. If people will pay a price, that's the price that will be asked. If people will accept a decrease in service with no price change, then that's what will happen. As people keep reminding us here, the primary (and often only) intent of any corporation is maximizing income while decreasing expenses. Supplying a service is an expense, so you'd expect them to supply the minimum service that doesn't result in a major loss of paying customers. And you should expect them to constantly probe the market to see if they can charge higher prices for the same service.
There's nothing at all amazing about this behavior. It's the expected behavior.
... but I could not help noticing that it was the physicists, not the voodoo priests, who had made possible the life-promoting technology we enjoy today.
A hypothesis has been proposed that the important thing explaining the success of "science" isn't its logic or its experimental method. Rather, its real innovation was open publication.
Various historians have pointed out that the "scientific method" (or, more accurately, methods) have been discovered by people throughout history. Thus, lots of "primitive" societies have had locally developed medicine that is effective to various degrees, and such medicine is discovered by the expected observation and experimentation. But most of the groups that developed such methods had one serious limitation: They have generally been controlled by "guilds" that kept their information as a closely-held secret. This is especially clear with medicine, which has existed thoughout history, and is almost always a specialty that requires admittance to the exclusive club. That club might discover various useful medicines and medical techniques, but they rarely shared them with their neighbors, who were treated as competitors.
This attitude has pervaded our history. But, the hypothesis explains, in Europe during the 15th and 16th centuries, an unusual practice developed in Europe. Some of the scholarly types there, including the astrologers, doctors, engineers, and others, developed a practice of publishing their results and distributing the information to anyone who could afford a subscription. Most of the publications ended up in libraries, where they were permanently available to anyone who was literate. The result was that, for the first time in history, European scholars could easily build on each others' work, and nothing was forgotten because it was only known to a tiny group in some remote village.
Our modern science and technology, the hypothesis suggests, is merely a result of several centuries of the "standing on the shoulders of giants" approach, which is merely a rephrasing of the ethic of open publication.
An implication of this hypothesis, if it is true, is that the Western world's current attempt to privatize all knowledge by increasingly more repressive copyright and patent law should lead to a loss of the lead in scientific development. These legal restrictions amount to blocking the use of published information, ending the usefulness of the open-publication model of development. We've already seen what could be the start of this, with the reported uses of patent and copyright law to prevent competitors from building on the "IP" owner's knowledge. For example, medical researchers have been prevented from doing studies comparing a commercial drug with potential replacement drugs, by requiring a license from the owner of a patented drug to use it in scientific studies.
If this hypothesis is accurate, we'd expect scientific development to slowly migrate to countries that don't impose such restrictions on the practice of building on others' patented results. It might be interesting to see if there is evidence to support such predictions...
Eat anything preprocessed? insects are in them, ground up with the rest of it.
Some years back, there was an interesting Q/A in Consumer Reports on this. A letter replied to a recent article on the limits to allowed insect fragments in food, and asked what was dangerous about eating insects. The letters editor referred the question to one of their medical experts, who replied: Actually, there's nothing dangerous about eating most insects. In fact, they're quite nutritious, a good low-fat protein source. The only reason you'd worry about them in food is that, unless the insects are there intentionally as one of the ingredients, they indicate unsanitary preparation conditions and the presence of other things [bacteria, etc.] that may not be quite as safe to eat.
I thought it was an unusually informative reply to the question. It applies here, since TFA is talking about insects specifically grown and sold as food, not as contaminants in other food.
One of the biological curiosities about humans is that we don't consume (very many) insects. Our closest relatives, chimps and bonobos, eat insects regularly. One of the textbook examples of chimps making and using tools is their common technique for capturing termites. They break off a stem of grass, strip it of its leaves (thus engaging in "tool construction"), insert it into an opening in a termite hive, wiggle it around a bit to get the termites to attack it as an intruder, then pull it back out covered when the attacking insects. They then stick the stem in their mouth and strip off the termites. Yum. Field studies have reported that chimp troops get up to 50% of their protein by eating insects and small animals that they catch and eat.
But for some reason, humans mostly stick to larger game, and ignore the smaller packages of protein all around us. We're big-game hunters, competitors of the lions, jackals and wolves. TFA is basically suggesting that we rethink this evolutionary choice, and take advantage of a large food source that we've generally dismissed as annoying pests. From a strictly biological viewpoint, it makes good sense. But our evolutionary heritage as a large, top-level predator probably goes against it, and most human societies won't easily accept the idea.
So what do they do/produce, if anything? That web site is as opaque and uninformative as any I've seen; it reads like a prototype for a "shell" or "front" corporation. And this discussion (or TFA) hasn't imparted any information about the company, either.
I suspect it doesn't stand for ElectroMagnetic Conformance, Environmental Modeling Center, or Emmaus Moravian Cemetery.
So what sort of precedent do we have for prosecuting a non-citizen for treason? Assange is an Australian citizen, after all, not American.
It could be interesting to watch them attempt to extradite another country's citizen and charge him with treason. (Recently, the usual approach has been to just grab the person and spirit him off to an undisclosed location for "questioning", then dump him on a hillside in Macedonia.;-)
During the Cold War it was standard to brief military and government employees to be wary of espionage attempts and trust no one.
Yeah, and people generally went along with it. But there's a major difference between that and the current issue. Back then, there was a real foreign enemy trying to harm us. Today, the purported enemy is just ordinary citizens who are trying to expose their own government's corruption to its own voters. I suspect that a large part of our population understands this. The government employees, who see the corruption first hand, certainly do.
(Well, OK, in the case of wikileaks, it's mostly citizens of other countries. But those people and their governments aren't at war with us; they're just trying to tell us about our own government's corruption.)
The memo also suggests that agencies require all their employees to report any contacts with members of the news media they may have.
But they've been telling us all along that the wikileaks folks don't qualify as "journalists" and don't deserve the legal protections that most democracies give to "news media". Employees in contact with such online information sources can easily think that such requirements don't apply, since they've been specifically told that such organizations aren't news media.
Maybe they should think of a better way of expressing what they want their employees to do. Or stop the pretense that, since there are no printing presses involved, people working on informing the public online don't qualify for legal protections such as the US's First Amendment guarantee of freedom of the press.
Customers are not spending money on all those alternatives. Therefore the market is saying exactly what escort wagon asserted.
Well, I was at an event yesterday where there were a number of iPads visible and, remembering the comments here, I asked the people who had one why they'd bought an iPad and not another brand tablet. Every one of them responded with a blank look, and/or said something like "What other tablets?". I also heard a few people commenting that they were thinking of buying one, and when I asked them the same question, they were also puzzled because they didn't know of any others.
So I'd say (based on this tiny sample) that The Market just might be saying something like "If you want people to buy your new product, you have to market it so they know it exists".
I ran across a similar, but funnier, example a while back, when Android started getting traction. I'd ask people why they'd bought an Android gadget rather than one running linux. Usually, they'd actually heard of linux, and gave answers making it clear that linux is a toy, not something that anyone would actually buy. Or it doesn't have windows, so they couldn't use it. But Android was a cool new product that finally did a lot of things right.
I think that what The Market is often really saying is "I'm an idiot". (A number of economists have drawn the same inference, but they usually use a lot more words to express it.)
When a company has to litigate to get customers/compete, it's not a good sign.
On the other hand, when the government simply decides to go with an existing vendor without considering other options, it's also not a good sign. After all, we don't want government wasting money on inferior solutions, do we?
Nah; we expect them to first do the required investigation of alternative vendors. Then we expect them to dutifully buy from the corporation that made the biggest contributions to the campaigns of whoever appointed them.
Of course, since the 2000 elections, one of the biggest campaign contributors in the US has been Microsoft. Mostly to Republicans, of course, but they also contributed to a lot of Democratic campaign funds, just in case. Despite all that, a government agency isn't supposed to simply buy from them without first making some sort of pretense of asking for bids and investigating the alternatives.
How often do top conferences/journals publish a result like "Researchers repeat non-controversial experiment, find exactly the same results"?
About as often as they publish "Researchers tested the statistical correlation between X and Y, and found that they are independent and unrelated."
The difficulty of getting "negative" results published is quite well known, even when a negative result is highly significant (in the practical, not the statistical sense;-). If anything, this is a bigger problem than the difficulty with being the second or third to report an important result.
I feel that many researchers publish far too many papers with each one being an epsilon improvement on the previous.
Google the phrase "minimum publishable unit" (MPU) for lots of explanations of this phenomenon.
I would rather they wait and produce one good well-written paper rather than a string of ten sequential papers.
Those are published. They're they "summary" articles that you occasionally see, summing up the results of a particular line of research. It's common to wish we could see only the summaries, and not bother with the MPU papers.
Actually, some people have supported the MPU + summary approach, primarily on the grounds that it keeps a research community in touch with each other, and encourages discussion and rapid feedback between people working on related topics. But reading all those MPU papers can be a real time sinkhole. It's often easier to just scan the abstracts, and read only papers that strike you as especially interesting. But then you run the risk of missing something that turns out to be important.
There are signs that the move to the Internet is encouraging a somewhat improved approach, in which the "MPU" papers are replaced by blog-like reports on a web site. With good indexing, this can make it easier for people involved in related research to keep track of each others' work and results, and the blog software implements rapid feedback from colleagues (and sometimes spammers or religious fundies;-). This approach could be a real time saver for everyone, while making the information more easily available. We'll see how it works out.
Besides, its hard to get to work in the morning without breaking ANY law. How would you even know? You could be violating the law RIGHT NOW just by reading this.
I've run across a number of articles explaining why, in most of the world, it's logically impossible to not be violating a law. The reasons are often somewhat funny, and typically dependent on specific interpretations of the law.
One place I lived, a journalist gave an example with local laws. He asked if you have any money in your pockets. If not, you can be arrested for vagrancy (and having a credit card didn't help you). If so, are you aware of the various gambling games based on money? There are simple coin-tossing games, of course, and with bills, there are several games based on comparing the serial numbers according to various schemes. But the local laws made it illegal to be "in possession of gambling instruments", so carrying paper money or coins was technically illegal. So you're either a penniless vagrant, or you're carrying evidence of being involved in an unregistered gambling operation.
When I moved to Massachusetts in the early 1980s, I ran across a funny example. At the time, it was technically illegal for a new resident to register their automobile here. I learned this when I got a "runaround" while trying to register our car. At the registry, they told me that I couldn't register the car because I didn't have proof of insurance from a local insurance agent. So I dutifully went to an auto insurance dealer, asked if I could get insurance - and they told me I couldn't because the car wasn't registered in the state. I went back to the registry, where they repeated their story, and I told them I'd just been to an insurance agent. The registry person just grinned and registered the car. But this was an unauthorized "overlooking" of the letter of the law; they should have continued to refuse to register the car. Then, after the legal timeout interval had passed, I could be arrested for being a legal resident but failing to register my car. A few years later, I read that the laws had been revised so that it was legal for people moving to the state to register their car.
Anyway, chances are pretty good that wherever you live, you are violating one or more laws right now, and if you cease to violate them, you'll be in violation of some other laws. That's the way legal systems usually work. There's always a valid legal ground for arresting you, because you are a criminal, as is everyone around you.
Presumable a "cloud boot" will go directly to the internet and boot from there. I just wonder if it will be encrypted...
If not, we can trust that it'll only be a matter of a few weeks before people find that their ISP (think Comcast for example) is inserting their own software into the download. And it may only be a matter of days before lots of Windows botnets have inserted themselves into the boot process, making all Windows "cloud computers" part of one gigantic spam/phish System -- let's just call it "skynet" and be done with it.
Of course, the way most commercial network encryption goes, even if a cloud boot is encrypted, the encryption will quickly be cracked. But we won't read about it for a year or three, so we won't know it's happened.
Well, yeah, I read that. But he also suggested why a vigilante approach might not work too well. In particular, the phishing part is more and more being run by the organized crime crowd, who in many places function much like the government: If you hurt their people, they simply kill you. So we might want to be careful about which spammers and phishers we approach with our torches, pitchforks and clubs.
A much safer approach would be to spread the existing (open, free) software that helps spot email bearing malware and trashes it. That's a lot harder for the crime bosses to pin on specific anonymous victims, since the effect is to silently ignore the malware. They don't know you're in on it, so they won't send Vinnie and Joe around to take you out.
But we should take care of this problem. Otherwise we can't claim that there is any real benefit to the citizens in using the internet that we have so painstakingly created.
Um, we (the folks who brought you the Internet, including email) have done it. On the machine where my primary email address lives, the email software runs a program that does a pretty good job of testing each message for problems and giving a "spam" rank. I have my reader automatically file everything above a threshold in a "junk" folder, which I check occasional for false positives. I can also add things like keywords (e.g., certain commercial domains) to the list of suspicious content patterns.
This is hardly anything unusual. There are quite a lot of email packages out there that do a good job of fingering spam and malicious email. And there are a lot of people fighting the battle of keeping up with the bad guys and improving the ranking software.
But there's one thing we can't (legally;-) do: We can't force people to use such software. If people want an iPhone, they get email software that's approved by Apple, and we have no power over Apple. No matter how good our anti-malware code is, Apple can nix it, and iPhone users won't have it.
So don't say that we should take care of the problem. The people who should take care of it are the ones who decide what software will be delivered on the machines they sell to their customers. If they choose not to bother with the easily-available software that checks for spam, phishing, and other malware, there's not a thing we can do to force them to change their ways.
Well, OK, we can publicize their failures. Like we've done here. Publicity isn't always successful, especially against corporations with billion-collar marketing budgets, but it's really the only tool we have. Just providing the software doesn't work; we have copious examples to support that claim. So publicity is our only real recourse.
One of the problems with publicity is that the corporations routinely respond by saying that we should tell only them, and let them fix the problem. We also have copious evidence that this doesn't work. Even the most responsive corporations (which does include Apple and Google) will sit on unannounced exploits for months or years, until we announce the exploits to the world. So the only real solution is to not bother telling them about problems. Or rather, we should be looking for exploits, sending them the details of each exploit, and shortly thereafter, announcing each exploit in an appropriate public forum. Experience show that that's the only way we can get them to actually permit us to fix the problems.
(I wonder if/. would be such a forum. We should watch and see if this discussion gets any response from the guilty parties.;-)
So the lesson is, if you use an iPhone - don't click on that link until you check it out the full email header on a PC.
And this is a good hint at a major problem with mobile email: The user isn't generally allowed to see the full headers. I have a G1 (Android) with gmail installed. I've tried to find the email headers on several occasions, and as far as I can tell, there's no way to see them. And this isn't just a problem on Android; I also read my gmail from my linux and Mac computers, and I can't see the headers there, either. This is why my preferred email address is on an academic unix (FreeBSD) machine where I can run any of several mail readers, all of which show me the headers. And I can also use the low-level text-only mail(1) command. And I can read my mail with vi.
I get the impression that most GUI email readers don't show the headers because their authors consider their users too stupid to understand email headers. For those of us that aren't that stupid, it's not hard to see the symptoms of a phishing attack -- if we are permitted access to the full email message. But I suppose we're a small minority, so the suppliers of commercial email software see no reason to cater to us.
The frustrating part of this is that you know the software has all the headers, and could show them to us as easily as it shows the contents of the From: and Subject: lines. So denying us access to the rest of the headers is done with malice aforethought, and leads to this sort of susceptibility to phishing. Either that, or users learn to not read mail from strangers when using the limited (or limiting) commercial email readers.
I mean, DST change and year change? It's some of the most obvious and basic corner cases that you write tests for, especially in an application that specifically deals with time! It's practically textbook stuff, or an interview question for QA position. And so it's extremely surprising when that kind of thing goes wrong in production.
Huh? I learned years ago that DST and timezone changes are exactly when you'd expect the code to be wrong. In particular, I always look for clock weirdness at a DST change -- and I've rarely been disappointed.
You might expect the companies that develop the software to thoroughly test such things. Based on experience, I don't expect this at all. I expect them to sell their gadgets, and only bother getting it right when they get over some threshold number of bug reports.
... but how do you then deal with things like timezones? What if the phone's moved timezone since the alarm was set?
Some years back, shortly after I got my first cell phone whose clock synced automagically to the cell towers' clocks, I was on a trip to northern Arizona. The state of Arizona doesn't use DST; it's always in the MST (Mountain Standard) zone. But the Indian reservations, which are legally independent of the state and answer only to the US government, use DST. I found that, when driving around in the countryside, my phone would frequently jump back and forth between MST and MDT, depending on whether the tower it was talking to was on a reservation or not. When I mentioned this to locals, they'd just grin, and suggest turning off the time syncing while I was in the area.
I think some cell phones have gotten smarter about this since then. My current phone lets me turn DST on/off separately from syncing, so in boundary cases like this, it will stay consistent with itself. It also delays the syncing a while after crossing a time-zone border, apparently to make sure that you're really in the new time zone. But this is only a heuristic, and I expect there might be some situations that would confuse it.
Well, I certainly wouldn't argue against replacing our idiotic calendar system with something sane. But that's really a different problem than getting computers to deal with date/time values correctly. I ran across an elegant explanation of the problem some years back, when I got involved with my first java project, and the first time we crossed a DTS boundary, the code's time calculations got all screwed up. I did a bit of digging online, and came up with the simple explanation from a writer who said, in essence: "Look at the java specs. They require that the clock routines store the date+time internally in local time. This is all you need to know to correctly predict that code based on these routines won't work, and will never be debugged. In particular, new bugs will always appear when going in or out of DST. If the basic, internal time representation is in local time, these bugs will always exist, and no programmer will ever be able to fix them. As long as java decrees local time internally, its clock routines will never quite work right."
The writer went on to explain why the only sensible internal representation of time is in a universal format that's the same everywhere. Such a time value can be translated to any "user-friendly" time notation desired, in the software's human-readable output. But translating the other direction isn't always possible, and even when it is, the algorithms are often so complex that programmers just won't get them right.
There was a further observation that, if you want to do date/time arithmetic (and you will), the universal format should be a single number in a format that the hardware can use directly. This will save you the cpu cycles needed to convert to a hardware format for calculations, but is otherwise immaterial. The number can be the typical second counter, as used in unix, vms, and other OSs. It can be the floating-point day counter used by astronomers. It could be the ISO time format. The exact form isn't material; what's important is that you can do arithmetic with it efficiently, and that it be the same inside every machine (whose clock is correct;-). But if you need sub-second precision, it's nice if the representation can handle that.
I once worked on a project where the management decreed ISO timestamps - the long character string, like "20110102194617". We software guys groaned, but we quickly found an excuse of the form "one of our products uses Standard X, which requires the unix time value" to write the two appropriate conversion routines. From then on, we could use the unix time value during in-memory calculations, and the only cost was an ISO-utime conversion during input, and a utime-ISO conversion when writing to the database where it would be visible to non-programmers.
I've also worked on numerous projects where the management decreed local time. On those projects, we just admitted openly that the date/time calculations would never be correct, but we'd do our best to patch as many of the inevitable bugs as we could find time for. In a few, we arranged things so that managers could never actually see the internal data representations, and we hid the universal timestamps behind "format for display" routines to show the data in management's favorite format(s).
Actually, deflecting discussions of the internal time formats to discussions of calendar reform could be an interesting way of preventing management from learning how you're actually getting the dates and times correct. And if enough of us start doing it, maybe we can get some calendar reform.
Clock problems are a lasting source of geek humor...
Here's the music: the sun's output is weaker than expected and yet 2010 looks like one of the hottest years on record. What is this solar minimum proving again?
That human industry is now more powerful than the sun!
Actually, this is somewhat of an old story. When I was a kid, living in Seattle in the 1950s and 1960s I often heard or read comments about how the Pacific Northwest (from roughly Oregon to British Columbia) was getting cooler. One of the clear signs of this was that for the first half of the 20th century, the ice fields and glaciers in the Cascade and Olympic mountains had slowly grown. This was recognized as an anomaly, because scientists had good data showing that most of the planet had been getting warmer since at least 1800. Why the Pacific Northwest was an exception wasn't understood. There were a number of hypotheses, but no really good evidence to say which (if any) was right.
This ended during the 1970s, when the snow caps stopped their growth and started shrinking. When I went back recently, I was impressed by the shrinkage in the white tops of the mountains. Especially noticable is Mount Ranier, the giant volcano standing apart from the other mountains, and visible anywhere in the area (when the rain clears up;-). Its white top is visibly much smaller than it was 40 years ago. This is significant, because glaciers are pretty good at averaging the temperature over several years, and this is a highly visible sign of the warming over the past 30 years or so. Glaciers don't mistake a week or two of abnormally warm or cold weather for a decades-long trend.
Part of the current Global Warming story is that this story is repeated all around the world. It's true that there are a few areas that are slightly cooling now. But there aren't many, and they're a lot smaller than the cooling areas even 50 years ago. And the mechanisms are a lot better understood than 50 years ago.
Still, as the universal reminder in scientific papers remind us, further research is needed. We have a lot more data supporting the general warming scenario, of course, but we don't fully understand the insanely complex weather/climate mechanisms. Some important parts of our lives are affected by this, and it would be useful for everyone if we understood the whole mess a lot better.
Slashdot Mirror
I hope if that man ends up on American soil that the citizens of this country (US) riot and raise fucking hell. What our government plans to do is wrong, it's illegal, and they know it. So does every citizen and every member of the press.
How many cases can you list where any significant number of US citizens raised such a fuss in defense of someone?
I'm not saying it couldn't happen; I'd just be interested in the precedent(s) that would give you such hope.
It was given to him.
Ahh. So your assertion is that if it is "given" to him, as opposed to his asking for it, that's not espionage?
Hey, how 'bout you post your address here, so that a few of us readers can "give" you things that will make you guilty of assorted crimes. I don't actually have anything like that in my posession right now (that I know of), but I'm sure I could make something up given a bit of time.
Hmmm ... I think you mean you can run a Windows Mobile phone about a week at idle before it gets cut off by a 500MB cap. If you actually use it for what it's advertised for, you'll hit that cap in a much shorter time. If you visit any site that sends a video ad, you might hit the cap in a single day.
They have managed to turn the free airways into the most expensive form of communication ever.
You should google the phrase "whatever the market will bear". It's the primary explanation of most corporate behavior. If people will pay a price, that's the price that will be asked. If people will accept a decrease in service with no price change, then that's what will happen. As people keep reminding us here, the primary (and often only) intent of any corporation is maximizing income while decreasing expenses. Supplying a service is an expense, so you'd expect them to supply the minimum service that doesn't result in a major loss of paying customers. And you should expect them to constantly probe the market to see if they can charge higher prices for the same service.
There's nothing at all amazing about this behavior. It's the expected behavior.
... but I could not help noticing that it was the physicists, not the voodoo priests, who had made possible the life-promoting technology we enjoy today.
A hypothesis has been proposed that the important thing explaining the success of "science" isn't its logic or its experimental method. Rather, its real innovation was open publication.
Various historians have pointed out that the "scientific method" (or, more accurately, methods) have been discovered by people throughout history. Thus, lots of "primitive" societies have had locally developed medicine that is effective to various degrees, and such medicine is discovered by the expected observation and experimentation. But most of the groups that developed such methods had one serious limitation: They have generally been controlled by "guilds" that kept their information as a closely-held secret. This is especially clear with medicine, which has existed thoughout history, and is almost always a specialty that requires admittance to the exclusive club. That club might discover various useful medicines and medical techniques, but they rarely shared them with their neighbors, who were treated as competitors.
This attitude has pervaded our history. But, the hypothesis explains, in Europe during the 15th and 16th centuries, an unusual practice developed in Europe. Some of the scholarly types there, including the astrologers, doctors, engineers, and others, developed a practice of publishing their results and distributing the information to anyone who could afford a subscription. Most of the publications ended up in libraries, where they were permanently available to anyone who was literate. The result was that, for the first time in history, European scholars could easily build on each others' work, and nothing was forgotten because it was only known to a tiny group in some remote village.
Our modern science and technology, the hypothesis suggests, is merely a result of several centuries of the "standing on the shoulders of giants" approach, which is merely a rephrasing of the ethic of open publication.
An implication of this hypothesis, if it is true, is that the Western world's current attempt to privatize all knowledge by increasingly more repressive copyright and patent law should lead to a loss of the lead in scientific development. These legal restrictions amount to blocking the use of published information, ending the usefulness of the open-publication model of development. We've already seen what could be the start of this, with the reported uses of patent and copyright law to prevent competitors from building on the "IP" owner's knowledge. For example, medical researchers have been prevented from doing studies comparing a commercial drug with potential replacement drugs, by requiring a license from the owner of a patented drug to use it in scientific studies.
If this hypothesis is accurate, we'd expect scientific development to slowly migrate to countries that don't impose such restrictions on the practice of building on others' patented results. It might be interesting to see if there is evidence to support such predictions ...
Eat anything preprocessed? insects are in them, ground up with the rest of it.
Some years back, there was an interesting Q/A in Consumer Reports on this. A letter replied to a recent article on the limits to allowed insect fragments in food, and asked what was dangerous about eating insects. The letters editor referred the question to one of their medical experts, who replied: Actually, there's nothing dangerous about eating most insects. In fact, they're quite nutritious, a good low-fat protein source. The only reason you'd worry about them in food is that, unless the insects are there intentionally as one of the ingredients, they indicate unsanitary preparation conditions and the presence of other things [bacteria, etc.] that may not be quite as safe to eat.
I thought it was an unusually informative reply to the question. It applies here, since TFA is talking about insects specifically grown and sold as food, not as contaminants in other food.
One of the biological curiosities about humans is that we don't consume (very many) insects. Our closest relatives, chimps and bonobos, eat insects regularly. One of the textbook examples of chimps making and using tools is their common technique for capturing termites. They break off a stem of grass, strip it of its leaves (thus engaging in "tool construction"), insert it into an opening in a termite hive, wiggle it around a bit to get the termites to attack it as an intruder, then pull it back out covered when the attacking insects. They then stick the stem in their mouth and strip off the termites. Yum. Field studies have reported that chimp troops get up to 50% of their protein by eating insects and small animals that they catch and eat.
But for some reason, humans mostly stick to larger game, and ignore the smaller packages of protein all around us. We're big-game hunters, competitors of the lions, jackals and wolves. TFA is basically suggesting that we rethink this evolutionary choice, and take advantage of a large food source that we've generally dismissed as annoying pests. From a strictly biological viewpoint, it makes good sense. But our evolutionary heritage as a large, top-level predator probably goes against it, and most human societies won't easily accept the idea.
EMC presumably refers to www.emc.com;
So what do they do/produce, if anything? That web site is as opaque and uninformative as any I've seen; it reads like a prototype for a "shell" or "front" corporation. And this discussion (or TFA) hasn't imparted any information about the company, either.
I suspect it doesn't stand for ElectroMagnetic Conformance, Environmental Modeling Center, or Emmaus Moravian Cemetery.
So what sort of precedent do we have for prosecuting a non-citizen for treason? Assange is an Australian citizen, after all, not American.
It could be interesting to watch them attempt to extradite another country's citizen and charge him with treason. (Recently, the usual approach has been to just grab the person and spirit him off to an undisclosed location for "questioning", then dump him on a hillside in Macedonia. ;-)
During the Cold War it was standard to brief military and government employees to be wary of espionage attempts and trust no one.
Yeah, and people generally went along with it. But there's a major difference between that and the current issue. Back then, there was a real foreign enemy trying to harm us. Today, the purported enemy is just ordinary citizens who are trying to expose their own government's corruption to its own voters. I suspect that a large part of our population understands this. The government employees, who see the corruption first hand, certainly do.
(Well, OK, in the case of wikileaks, it's mostly citizens of other countries. But those people and their governments aren't at war with us; they're just trying to tell us about our own government's corruption.)
The memo also suggests that agencies require all their employees to report any contacts with members of the news media they may have.
But they've been telling us all along that the wikileaks folks don't qualify as "journalists" and don't deserve the legal protections that most democracies give to "news media". Employees in contact with such online information sources can easily think that such requirements don't apply, since they've been specifically told that such organizations aren't news media.
Maybe they should think of a better way of expressing what they want their employees to do. Or stop the pretense that, since there are no printing presses involved, people working on informing the public online don't qualify for legal protections such as the US's First Amendment guarantee of freedom of the press.
Customers are not spending money on all those alternatives. Therefore the market is saying exactly what escort wagon asserted.
Well, I was at an event yesterday where there were a number of iPads visible and, remembering the comments here, I asked the people who had one why they'd bought an iPad and not another brand tablet. Every one of them responded with a blank look, and/or said something like "What other tablets?". I also heard a few people commenting that they were thinking of buying one, and when I asked them the same question, they were also puzzled because they didn't know of any others.
So I'd say (based on this tiny sample) that The Market just might be saying something like "If you want people to buy your new product, you have to market it so they know it exists".
I ran across a similar, but funnier, example a while back, when Android started getting traction. I'd ask people why they'd bought an Android gadget rather than one running linux. Usually, they'd actually heard of linux, and gave answers making it clear that linux is a toy, not something that anyone would actually buy. Or it doesn't have windows, so they couldn't use it. But Android was a cool new product that finally did a lot of things right.
I think that what The Market is often really saying is "I'm an idiot". (A number of economists have drawn the same inference, but they usually use a lot more words to express it.)
When a company has to litigate to get customers/compete, it's not a good sign.
On the other hand, when the government simply decides to go with an existing vendor without considering other options, it's also not a good sign. After all, we don't want government wasting money on inferior solutions, do we?
Nah; we expect them to first do the required investigation of alternative vendors. Then we expect them to dutifully buy from the corporation that made the biggest contributions to the campaigns of whoever appointed them.
Of course, since the 2000 elections, one of the biggest campaign contributors in the US has been Microsoft. Mostly to Republicans, of course, but they also contributed to a lot of Democratic campaign funds, just in case. Despite all that, a government agency isn't supposed to simply buy from them without first making some sort of pretense of asking for bids and investigating the alternatives.
How often do top conferences/journals publish a result like "Researchers repeat non-controversial experiment, find exactly the same results"?
About as often as they publish "Researchers tested the statistical correlation between X and Y, and found that they are independent and unrelated."
The difficulty of getting "negative" results published is quite well known, even when a negative result is highly significant (in the practical, not the statistical sense ;-). If anything, this is a bigger problem than the difficulty with being the second or third to report an important result.
I feel that many researchers publish far too many papers with each one being an epsilon improvement on the previous.
Google the phrase "minimum publishable unit" (MPU) for lots of explanations of this phenomenon.
I would rather they wait and produce one good well-written paper rather than a string of ten sequential papers.
Those are published. They're they "summary" articles that you occasionally see, summing up the results of a particular line of research. It's common to wish we could see only the summaries, and not bother with the MPU papers.
Actually, some people have supported the MPU + summary approach, primarily on the grounds that it keeps a research community in touch with each other, and encourages discussion and rapid feedback between people working on related topics. But reading all those MPU papers can be a real time sinkhole. It's often easier to just scan the abstracts, and read only papers that strike you as especially interesting. But then you run the risk of missing something that turns out to be important.
There are signs that the move to the Internet is encouraging a somewhat improved approach, in which the "MPU" papers are replaced by blog-like reports on a web site. With good indexing, this can make it easier for people involved in related research to keep track of each others' work and results, and the blog software implements rapid feedback from colleagues (and sometimes spammers or religious fundies ;-). This approach could be a real time saver for everyone, while making the information more easily available. We'll see how it works out.
Besides, its hard to get to work in the morning without breaking ANY law. How would you even know? You could be violating the law RIGHT NOW just by reading this.
I've run across a number of articles explaining why, in most of the world, it's logically impossible to not be violating a law. The reasons are often somewhat funny, and typically dependent on specific interpretations of the law.
One place I lived, a journalist gave an example with local laws. He asked if you have any money in your pockets. If not, you can be arrested for vagrancy (and having a credit card didn't help you). If so, are you aware of the various gambling games based on money? There are simple coin-tossing games, of course, and with bills, there are several games based on comparing the serial numbers according to various schemes. But the local laws made it illegal to be "in possession of gambling instruments", so carrying paper money or coins was technically illegal. So you're either a penniless vagrant, or you're carrying evidence of being involved in an unregistered gambling operation.
When I moved to Massachusetts in the early 1980s, I ran across a funny example. At the time, it was technically illegal for a new resident to register their automobile here. I learned this when I got a "runaround" while trying to register our car. At the registry, they told me that I couldn't register the car because I didn't have proof of insurance from a local insurance agent. So I dutifully went to an auto insurance dealer, asked if I could get insurance - and they told me I couldn't because the car wasn't registered in the state. I went back to the registry, where they repeated their story, and I told them I'd just been to an insurance agent. The registry person just grinned and registered the car. But this was an unauthorized "overlooking" of the letter of the law; they should have continued to refuse to register the car. Then, after the legal timeout interval had passed, I could be arrested for being a legal resident but failing to register my car. A few years later, I read that the laws had been revised so that it was legal for people moving to the state to register their car.
Anyway, chances are pretty good that wherever you live, you are violating one or more laws right now, and if you cease to violate them, you'll be in violation of some other laws. That's the way legal systems usually work. There's always a valid legal ground for arresting you, because you are a criminal, as is everyone around you.
Presumable a "cloud boot" will go directly to the internet and boot from there. I just wonder if it will be encrypted ...
If not, we can trust that it'll only be a matter of a few weeks before people find that their ISP (think Comcast for example) is inserting their own software into the download. And it may only be a matter of days before lots of Windows botnets have inserted themselves into the boot process, making all Windows "cloud computers" part of one gigantic spam/phish System -- let's just call it "skynet" and be done with it.
Of course, the way most commercial network encryption goes, even if a cloud boot is encrypted, the encryption will quickly be cracked. But we won't read about it for a year or three, so we won't know it's happened.
What, me cynical? Nah ...
Well, yeah, I read that. But he also suggested why a vigilante approach might not work too well. In particular, the phishing part is more and more being run by the organized crime crowd, who in many places function much like the government: If you hurt their people, they simply kill you. So we might want to be careful about which spammers and phishers we approach with our torches, pitchforks and clubs.
A much safer approach would be to spread the existing (open, free) software that helps spot email bearing malware and trashes it. That's a lot harder for the crime bosses to pin on specific anonymous victims, since the effect is to silently ignore the malware. They don't know you're in on it, so they won't send Vinnie and Joe around to take you out.
But we should take care of this problem. Otherwise we can't claim that there is any real benefit to the citizens in using the internet that we have so painstakingly created.
Um, we (the folks who brought you the Internet, including email) have done it. On the machine where my primary email address lives, the email software runs a program that does a pretty good job of testing each message for problems and giving a "spam" rank. I have my reader automatically file everything above a threshold in a "junk" folder, which I check occasional for false positives. I can also add things like keywords (e.g., certain commercial domains) to the list of suspicious content patterns.
This is hardly anything unusual. There are quite a lot of email packages out there that do a good job of fingering spam and malicious email. And there are a lot of people fighting the battle of keeping up with the bad guys and improving the ranking software.
But there's one thing we can't (legally;-) do: We can't force people to use such software. If people want an iPhone, they get email software that's approved by Apple, and we have no power over Apple. No matter how good our anti-malware code is, Apple can nix it, and iPhone users won't have it.
So don't say that we should take care of the problem. The people who should take care of it are the ones who decide what software will be delivered on the machines they sell to their customers. If they choose not to bother with the easily-available software that checks for spam, phishing, and other malware, there's not a thing we can do to force them to change their ways.
Well, OK, we can publicize their failures. Like we've done here. Publicity isn't always successful, especially against corporations with billion-collar marketing budgets, but it's really the only tool we have. Just providing the software doesn't work; we have copious examples to support that claim. So publicity is our only real recourse.
One of the problems with publicity is that the corporations routinely respond by saying that we should tell only them, and let them fix the problem. We also have copious evidence that this doesn't work. Even the most responsive corporations (which does include Apple and Google) will sit on unannounced exploits for months or years, until we announce the exploits to the world. So the only real solution is to not bother telling them about problems. Or rather, we should be looking for exploits, sending them the details of each exploit, and shortly thereafter, announcing each exploit in an appropriate public forum. Experience show that that's the only way we can get them to actually permit us to fix the problems.
(I wonder if /. would be such a forum. We should watch and see if this discussion gets any response from the guilty parties. ;-)
So the lesson is, if you use an iPhone - don't click on that link until you check it out the full email header on a PC.
And this is a good hint at a major problem with mobile email: The user isn't generally allowed to see the full headers. I have a G1 (Android) with gmail installed. I've tried to find the email headers on several occasions, and as far as I can tell, there's no way to see them. And this isn't just a problem on Android; I also read my gmail from my linux and Mac computers, and I can't see the headers there, either. This is why my preferred email address is on an academic unix (FreeBSD) machine where I can run any of several mail readers, all of which show me the headers. And I can also use the low-level text-only mail(1) command. And I can read my mail with vi.
I get the impression that most GUI email readers don't show the headers because their authors consider their users too stupid to understand email headers. For those of us that aren't that stupid, it's not hard to see the symptoms of a phishing attack -- if we are permitted access to the full email message. But I suppose we're a small minority, so the suppliers of commercial email software see no reason to cater to us.
The frustrating part of this is that you know the software has all the headers, and could show them to us as easily as it shows the contents of the From: and Subject: lines. So denying us access to the rest of the headers is done with malice aforethought, and leads to this sort of susceptibility to phishing. Either that, or users learn to not read mail from strangers when using the limited (or limiting) commercial email readers.
I mean, we're not talking about some Indian outsourcing sweatshop here. It's Apple. They're supposed to be better than this.
I think you mean that their advertising leads you to believe that they're better than this. ;-)
I mean, DST change and year change? It's some of the most obvious and basic corner cases that you write tests for, especially in an application that specifically deals with time! It's practically textbook stuff, or an interview question for QA position. And so it's extremely surprising when that kind of thing goes wrong in production.
Huh? I learned years ago that DST and timezone changes are exactly when you'd expect the code to be wrong. In particular, I always look for clock weirdness at a DST change -- and I've rarely been disappointed.
You might expect the companies that develop the software to thoroughly test such things. Based on experience, I don't expect this at all. I expect them to sell their gadgets, and only bother getting it right when they get over some threshold number of bug reports.
... but how do you then deal with things like timezones? What if the phone's moved timezone since the alarm was set?
Some years back, shortly after I got my first cell phone whose clock synced automagically to the cell towers' clocks, I was on a trip to northern Arizona. The state of Arizona doesn't use DST; it's always in the MST (Mountain Standard) zone. But the Indian reservations, which are legally independent of the state and answer only to the US government, use DST. I found that, when driving around in the countryside, my phone would frequently jump back and forth between MST and MDT, depending on whether the tower it was talking to was on a reservation or not. When I mentioned this to locals, they'd just grin, and suggest turning off the time syncing while I was in the area.
I think some cell phones have gotten smarter about this since then. My current phone lets me turn DST on/off separately from syncing, so in boundary cases like this, it will stay consistent with itself. It also delays the syncing a while after crossing a time-zone border, apparently to make sure that you're really in the new time zone. But this is only a heuristic, and I expect there might be some situations that would confuse it.
Well, I certainly wouldn't argue against replacing our idiotic calendar system with something sane. But that's really a different problem than getting computers to deal with date/time values correctly. I ran across an elegant explanation of the problem some years back, when I got involved with my first java project, and the first time we crossed a DTS boundary, the code's time calculations got all screwed up. I did a bit of digging online, and came up with the simple explanation from a writer who said, in essence: "Look at the java specs. They require that the clock routines store the date+time internally in local time. This is all you need to know to correctly predict that code based on these routines won't work, and will never be debugged. In particular, new bugs will always appear when going in or out of DST. If the basic, internal time representation is in local time, these bugs will always exist, and no programmer will ever be able to fix them. As long as java decrees local time internally, its clock routines will never quite work right."
The writer went on to explain why the only sensible internal representation of time is in a universal format that's the same everywhere. Such a time value can be translated to any "user-friendly" time notation desired, in the software's human-readable output. But translating the other direction isn't always possible, and even when it is, the algorithms are often so complex that programmers just won't get them right.
There was a further observation that, if you want to do date/time arithmetic (and you will), the universal format should be a single number in a format that the hardware can use directly. This will save you the cpu cycles needed to convert to a hardware format for calculations, but is otherwise immaterial. The number can be the typical second counter, as used in unix, vms, and other OSs. It can be the floating-point day counter used by astronomers. It could be the ISO time format. The exact form isn't material; what's important is that you can do arithmetic with it efficiently, and that it be the same inside every machine (whose clock is correct ;-). But if you need sub-second precision, it's nice if the representation can handle that.
I once worked on a project where the management decreed ISO timestamps - the long character string, like "20110102194617". We software guys groaned, but we quickly found an excuse of the form "one of our products uses Standard X, which requires the unix time value" to write the two appropriate conversion routines. From then on, we could use the unix time value during in-memory calculations, and the only cost was an ISO-utime conversion during input, and a utime-ISO conversion when writing to the database where it would be visible to non-programmers.
I've also worked on numerous projects where the management decreed local time. On those projects, we just admitted openly that the date/time calculations would never be correct, but we'd do our best to patch as many of the inevitable bugs as we could find time for. In a few, we arranged things so that managers could never actually see the internal data representations, and we hid the universal timestamps behind "format for display" routines to show the data in management's favorite format(s).
Actually, deflecting discussions of the internal time formats to discussions of calendar reform could be an interesting way of preventing management from learning how you're actually getting the dates and times correct. And if enough of us start doing it, maybe we can get some calendar reform.
Clock problems are a lasting source of geek humor ...
Here's the music: the sun's output is weaker than expected and yet 2010 looks like one of the hottest years on record. What is this solar minimum proving again?
That human industry is now more powerful than the sun!
Actually, this is somewhat of an old story. When I was a kid, living in Seattle in the 1950s and 1960s I often heard or read comments about how the Pacific Northwest (from roughly Oregon to British Columbia) was getting cooler. One of the clear signs of this was that for the first half of the 20th century, the ice fields and glaciers in the Cascade and Olympic mountains had slowly grown. This was recognized as an anomaly, because scientists had good data showing that most of the planet had been getting warmer since at least 1800. Why the Pacific Northwest was an exception wasn't understood. There were a number of hypotheses, but no really good evidence to say which (if any) was right.
This ended during the 1970s, when the snow caps stopped their growth and started shrinking. When I went back recently, I was impressed by the shrinkage in the white tops of the mountains. Especially noticable is Mount Ranier, the giant volcano standing apart from the other mountains, and visible anywhere in the area (when the rain clears up ;-). Its white top is visibly much smaller than it was 40 years ago. This is significant, because glaciers are pretty good at averaging the temperature over several years, and this is a highly visible sign of the warming over the past 30 years or so. Glaciers don't mistake a week or two of abnormally warm or cold weather for a decades-long trend.
Part of the current Global Warming story is that this story is repeated all around the world. It's true that there are a few areas that are slightly cooling now. But there aren't many, and they're a lot smaller than the cooling areas even 50 years ago. And the mechanisms are a lot better understood than 50 years ago.
Still, as the universal reminder in scientific papers remind us, further research is needed. We have a lot more data supporting the general warming scenario, of course, but we don't fully understand the insanely complex weather/climate mechanisms. Some important parts of our lives are affected by this, and it would be useful for everyone if we understood the whole mess a lot better.