The USPS needs to get rid of it's bad, underutilised services and focus on it's core, money making units.
While this sounds like standard business management turnaround strategy, I fear from my experience that the USPS is in much worse condition and that deeper, more fundamental issues need to be addressed. Like firing 90% of the workforce and replacing them with motivated employees who realize that if they do not perform, their jobs will end. Currently, in every single USPS office I have been to in the last few years, and I go in to an office every two weeks or so, the staff are largely slow, inefficient, do not speak English well, are often surly, and don't give one whit about customer service. There are employees who are the occasional exception, but the large picture remains. I've had people staffing the special services window tell me they can't staple a form together, or that they're too busy to make a photo copy even though they clearly aren't working on any pressing tasks, or that I used the wrong color pen *after* I've filled out a form, and no, they have no black pens I could use. Somehow they don't understand that there are alternatives to the USPS for many services, that they face stiff and growing competition, and that the future is bleak for the USPS existing at all, threatening the long-term prospects for the employees having jobs.
There are exceptions, naturally, but given that such laziness is blatantly apparent in customer-facing employees, I can only imagine the sloth behind the scenes. Such horrible attitude develops if employment is essentially guaranteed for life and (a) it is impossible to advance by working hard, and (b) it is impossible to get fired for not working.
The USPS needs a new work force. The idea of a national letter delivery service is an excellent one, and I continue to be amazed that they can deliver a letter anywhere in the US for under half a dollar. Imagine if they had efficient, motivated employees!
Actually I find this less of an issue. Check out grepmail [sourceforge.net] and mboxgrep [mboxgrep.org]. I use these pretty regularly and they're very useful for doing e.g. grepmail 'foo.*bar[a-z]' ~/Mail/mbox.gz >/tmp/messages; mutt -f/tmp/messages
HOLYCRAPTHANKYOU!!!! You have no idea how much time and headache you've saved me. Well, maybe you do, given that you can imagine how much of a pain it is to use plain grep on mbox and similar files.
2000? Hell, I have my email back to the early 1980s.
The real problem is that back then it was OK to put all messages in one file, and having one message per file is far more useful for searching with grep.
While that might be interesting, perhaps you can try doing it by remote control a large handful of millions of miles away, in near-total vacuum, with instruments that have survived 15G vibrational acceleration followed by months of punishing high-energy radiation while going through complex orbital dynamics to insert around a body that has about 2% the gravitational attraction of Earth. That would be impressive, just like the NASA video.
Each time NASA releases images from some distant planet or asteroid, I'm floored. The number of things that have to go right, that have to not fail, millions of miles away, is immense. Kudos to the scientists and engineers who worked on imaging Vesta. Fantastic results!
The real question is why would anyone (like Shanon Boase quoted in the summary) think that the total amount of greenhouse gasses created by the entire waste stream is going to magically go down if some of the stream is composted? Composting is, to first approximation, slow-rate burning that's driven by biological processes. It requires lots of oxygen. If you take a given amount of organic waste and bury it in a landfill instead of composting it, it has a much smaller tendency to decay, and often completely stops doing so. Sequestering carbon is better than releasing it, if you're worried about creating greenhouse gasses.
Thanks to the power of javascript and web2.0, you can again await the day when we'll be able to push a 6502 into the realm of the megahertz! [visual6502.org]
That is one of the coolest projects I have ever seen! I wish we had had something like that when I was taking intro to digital design.
That's a nice Hollywood-inspired vision you have there.
Reality is that research costs money. A lot of money. Being passionate and driven, in the Hollywood sense, is largely irrelevant because that does not get you research dollars. Money for health-related research comes from the NIH, and only the NIH, to first approximation. Yes, there are other sources, but the NIH dwarfs them all. Sure, an extraordinarily motivated researcher might be able to convince George Soros to give him a few million dollars to pursue a multi-year plan on a new drug target, but that's the Hollywood fantasy again. The vast majority of biology researchers get their money from the standard NIH grant mechanism called an R01 (pronounced ARR-OH-ONE). That would be your tax dollars at work.
As another poster pointed out, that's only the first step. A drug target has been identified by university research. Now, the hard part begins where multiple animal models are tested in large scale, followed by Phase I clinical trials with a small cohort to demonstrate that the drug causes no harm, then Phase II trials with a slightly larger cohort to determine effective doses, then, perhaps, another animal study or two because the results didn't work as well as anticipated in humans, followed by Phase I again on a reformulated drug, then more research to figure out why there were horrible side-effects, back to Phase I, then Phase II, and, if the developer is lucky, Phase III. We're talking years after the initial discovery now, with lots of hospital costs, lots of salaries, and *then* the legal stuff starts with the FDA to get approval for general release. Next, lobbying starts on the insurance companies, especially Medicare and Medicaid, to cover treatment with the drug.
Put it this way, there is an entire industry focused specifically on clinical trials, and most drug candidates don't make it through. Because we've set the bar so high to get a drug approved, and the success rate is so low, there must be substantial reward for many people to justify the expense. One researcher having a dream is not enough, despite what Hollywood would have you believe.
You're missing one point here. Foreigners cannot arbitrarily work for a US company: they must have visas to enter the States that specifically allow employment. Those are much harder to get than visitor visas. And, unlike US citizens who can freely go to just about any country on the planet without getting a visa first, in general non-US citizens require a visa (citizens of countries with visa waiver treaties with the US are the exception: like US citizens in many countries abroad, such foreigners essentially get a visa on-the-spot at immigration control; unlike US citizens visiting other countries, foreigners visiting the US also get fingerprinted and have their photo taken). Moreover, a US company cannot arbitrarily hire a foreign citizen.
So if a foreigner wants to start a company in the US, and, especially, employ other foreigners, it's kind of hard without the right visa. If you've got a startup, then physical proximity to potential investors and industrial partners is important.
Think of it this way: if Canada had ultra-lax laws for foreign workers, then Vancouver would be a prime location to house high-tech startups, what with its proximity to Seattle.
I wonder how they're going to handle the atmospherics.
Plumbing, water, the other stuff, that's easy and well-understood. We already know how to pump liquid stuff to change its level by a thousand feet reliably, inside a building. CO2, however, is denser than air, and would tend to accumulate at the bottom. Sure there are plenty of potential answers, I just wonder if they've thought about it since it sounds like the design is not far from a large open hole.
Yes it is a visual dictionary and if it is a cache-miss, then the fallback behavior is to re-parse the word slowly and sound it out. After a few encounters with a strange word it becomes visually cached as well. Parsing a word is far slower, of course, and is not the default behavior.
I have had the opportunity to go from an illiterate but fluent speaker in a language (Greek) to a literate one recently. I had been raised speaking the language, but was never taught to read it. With the advent of children whom I wish to also speak my first language, I've had to teach myself to read. This has been an interesting experience, and I can, from personal experience, 100% confirm the above quoted behavior. There is an additional layer, however, that of context. If you are familiar with the words you are about to read in any way, even if it's because you've read the first word in a commonly used phrase that has appeared in an expected context, memorization of the familiar trumps whole-word recognition.
Yes, another rather face-planting editorial failure. The summary should have been made clear that it was talking about electrically charged gas as a sterilization agent, since growth media derived from bovine blood plasma is standard stuff for filling petri dishes.
You haven't been paying much attention, apparently.
We live in a near police state where citizens are repeatedly urged by friendly, paternalistic voices over loudspeakers in places of common congregation that "if you see something, say something." We are video recorded everywhere (at least in London, but I suspect the same is true in most cities these days). While we don't yet have observation systems directly in our houses, we're coming remarkably close what with technology that sees quite well through walls these days. We must undergo examinations to ride on an airplane (and, soon, trains and busses, if TSA has their way) that 20 years ago would have been thought ludicrous. And if you didn't notice during the Bush administration (the second one), history was rewritten left and right for the convenience of the government. We have always been at war with Eastasia. Perhaps you missed the subtle linguistic shifts during the Bush / Blair partnership? Double plus ungood. The only thing Orwell missed was the speed with which it would happen: it's been somewhat slower than he anticipated, but we're without a doubt on the predicted path.
And trivial to manufacture with readily available low-grade materials. Heck, all you really need is a saltpeter mine, a source of charcoal, a junk yard, and some bronze-age ingenuity.
There are three programs that are vital to my professional life: MS Office, MSIE, and Photoshop.
The vast majority of the people with whom I share MS Office documents are running Windows, and since these people are often holding multi-million dollar purse strings, compatibility is a very serious concern. Sharing documents that have problems with tables or figures not rendering correctly is not an option; I must have a box with Windows on my desk to run MS Office. LibreOffice / OpenOffice is fine for basic, non-critical vanilla documents (and because of a long-standing bug with subscripts and saving to DOC format files, I mean really very basic vanilla documents) but anything beyond that requires MS Office. When you're sending in a grant application to a big governmental agency and the figures don't render correctly, or the document flow suddenly makes your submission a page too long, the governmental agency doesn't cut you slack for using open source software, but instead tosses your application in the trash bin for being out of spec.
In addition, there are a number of internal web applications that our company uses that really do not work well under anything other than MSIE on Windows. And, vitally, if I'm having problems with one of those applications and am not running MSIE on Windows, the IT department will tell me to suck eggs (but they use nicer words). If my use of the web applications isn't successful, my lab gets shut down. Why should I screw around with trying to get these web applications working on Firefox / Konqueror / Chrome / Opera on Linux to work when the stakes are so high?
Because that box is on my desk for MS Office and MSIE, I can run Photoshop on it as well. GIMP is nowhere near good enough (no 16 bit, no CMYK, barely passable cloning tools, inadequate color manipulation, inadequate selection and masking, etc). While I could run Photoshop on a Mac, I only have so much room on my desk, and as my main box runs Fedora, Photoshop gets run on the Windows box.
In a similar vein, I've often wondered how accurate and extensive the medical knowledge I've gained over the years of watching ER, House, and the other medical dramas on television is. From time to time, I check on things that are said, and usually they're right, if sometimes over-dramatized. But then, would that not be a good way to educate the general public about medical issues? Or, really, anything? I don't know the symptoms of stroke from those stupid acronym signs on the subway (PACE? HELP? STROKE? TIME? whatever -- the advertising campaign didn't work), I know it from watching medical dramas.
Anyone else reminded of the episode of The West Wing where President Bartlet speaks candidly after an interview is supposedly over but the cameras are still rolling? In that fictional case his staff went apoplectic trying to put spin on how it was a mistake, etc., when in reality it was a calculated and savvy move where the President acted in full knowledge that the camera was still on.
The practicalities of limited resources ensure that when committing scarce manpower to monitoring an individual's movements by direct observation, the choice of suspects is expected to be justified by substantial existing evidence, even if warrants have not been obtained. Although we might debate the appropriateness of legal requirements for governmental monitoring of an individual's movements with traditional methods, the limited resource effect ensures that we have a system that at least approximates due process.
With tracking devices, these limitations are lifted and the threshold of justification drops substantially such that, in practical, real terms, we approach a standard of constant monitoring of every individual's every movement when in public, without due process. When it becomes easy and inexpensive to monitor any individual, and the hand of practicality no longer guides selectivity of police action, we must impose legal requirements to protect our citizens' rights. Police are not allowed to go on fishing expeditions, by and large, looking for evidence without any justification.
Please mod the parent up, that's a remarkably informative reply, especially from an AC.
The important part, if I understand the technique then, is not that we're painting the surface pixel by pixel, as one might expect for an image produced by scanning a focused beam across the asteroid surface to create a 2D image, and as I expected to see in the photo. Instead of a scanning beam, there's a single pulse that gets sent out with some impressively sophisticated processing on the echo allows that signal to be broken down by delay and doppler shift, and those two parameters map only approximately to a side-view relative to an Earth based observer.
It also explains the suspiciously spherical view of the asteroid, and implies that if the asteroid were not spinning relative to the Earth, the technique would degenerate to a 1D image of average reflectivity at a given distance.
I read the articles. I watched the video. But, I'm confused: why does the asteroid appear side-lit in the images?
If we're imaging the asteroid based on radar that's transmitted from the Earth, and the asteroid is heading nearly directly toward us, then we should be able to see images of the asteroid nearly full face on, rather than it appearing like a crescent moon with illumination from the sun, right? The radar illumination is from a source that spatially coincides with the receiving apparatus, so the image should appear more like the full moon.
In fact, at the end of the article, in the concluding paragraph, the last line of serious content is:
Once you've ridden the learning curve and spent some time actually getting to know the innards, you may decide you'd be better off running FreeBSD on the next set of Web servers, SMTP relays, or application servers you build
Not in a single place does the article advocate that any variant of BSD is appropriate for the desktop, save Apple's.
So, as these were "novel" or "neat" "brain-teasers" - they really assessed only if you were some mathematical savant - and had nothing to do with your coding, or even problem-solving abilities.
The real reason to ask questions like that is not to assess the knowledge level of the candidate, but their ability to reason with limited knowledge. You, the potential employer, want to see how they think. For a programmer, that's going to be more important than anything else.
Slashdot Mirror
The USPS needs to get rid of it's bad, underutilised services and focus on it's core, money making units.
While this sounds like standard business management turnaround strategy, I fear from my experience that the USPS is in much worse condition and that deeper, more fundamental issues need to be addressed. Like firing 90% of the workforce and replacing them with motivated employees who realize that if they do not perform, their jobs will end. Currently, in every single USPS office I have been to in the last few years, and I go in to an office every two weeks or so, the staff are largely slow, inefficient, do not speak English well, are often surly, and don't give one whit about customer service. There are employees who are the occasional exception, but the large picture remains. I've had people staffing the special services window tell me they can't staple a form together, or that they're too busy to make a photo copy even though they clearly aren't working on any pressing tasks, or that I used the wrong color pen *after* I've filled out a form, and no, they have no black pens I could use. Somehow they don't understand that there are alternatives to the USPS for many services, that they face stiff and growing competition, and that the future is bleak for the USPS existing at all, threatening the long-term prospects for the employees having jobs.
There are exceptions, naturally, but given that such laziness is blatantly apparent in customer-facing employees, I can only imagine the sloth behind the scenes. Such horrible attitude develops if employment is essentially guaranteed for life and (a) it is impossible to advance by working hard, and (b) it is impossible to get fired for not working.
The USPS needs a new work force. The idea of a national letter delivery service is an excellent one, and I continue to be amazed that they can deliver a letter anywhere in the US for under half a dollar. Imagine if they had efficient, motivated employees!
Actually I find this less of an issue. Check out grepmail [sourceforge.net] and mboxgrep [mboxgrep.org]. I use these pretty regularly and they're very useful for doing e.g. grepmail 'foo.*bar[a-z]' ~/Mail/mbox.gz >/tmp/messages; mutt -f /tmp/messages
HOLYCRAPTHANKYOU!!!! You have no idea how much time and headache you've saved me. Well, maybe you do, given that you can imagine how much of a pain it is to use plain grep on mbox and similar files.
2000? Hell, I have my email back to the early 1980s.
The real problem is that back then it was OK to put all messages in one file, and having one message per file is far more useful for searching with grep.
While that might be interesting, perhaps you can try doing it by remote control a large handful of millions of miles away, in near-total vacuum, with instruments that have survived 15G vibrational acceleration followed by months of punishing high-energy radiation while going through complex orbital dynamics to insert around a body that has about 2% the gravitational attraction of Earth. That would be impressive, just like the NASA video.
Each time NASA releases images from some distant planet or asteroid, I'm floored. The number of things that have to go right, that have to not fail, millions of miles away, is immense. Kudos to the scientists and engineers who worked on imaging Vesta. Fantastic results!
The real question is why would anyone (like Shanon Boase quoted in the summary) think that the total amount of greenhouse gasses created by the entire waste stream is going to magically go down if some of the stream is composted? Composting is, to first approximation, slow-rate burning that's driven by biological processes. It requires lots of oxygen. If you take a given amount of organic waste and bury it in a landfill instead of composting it, it has a much smaller tendency to decay, and often completely stops doing so. Sequestering carbon is better than releasing it, if you're worried about creating greenhouse gasses.
Thanks to the power of javascript and web2.0, you can again await the day when we'll be able to push a 6502 into the realm of the megahertz! [visual6502.org]
That is one of the coolest projects I have ever seen! I wish we had had something like that when I was taking intro to digital design.
That's a nice Hollywood-inspired vision you have there.
Reality is that research costs money. A lot of money. Being passionate and driven, in the Hollywood sense, is largely irrelevant because that does not get you research dollars. Money for health-related research comes from the NIH, and only the NIH, to first approximation. Yes, there are other sources, but the NIH dwarfs them all. Sure, an extraordinarily motivated researcher might be able to convince George Soros to give him a few million dollars to pursue a multi-year plan on a new drug target, but that's the Hollywood fantasy again. The vast majority of biology researchers get their money from the standard NIH grant mechanism called an R01 (pronounced ARR-OH-ONE). That would be your tax dollars at work.
As another poster pointed out, that's only the first step. A drug target has been identified by university research. Now, the hard part begins where multiple animal models are tested in large scale, followed by Phase I clinical trials with a small cohort to demonstrate that the drug causes no harm, then Phase II trials with a slightly larger cohort to determine effective doses, then, perhaps, another animal study or two because the results didn't work as well as anticipated in humans, followed by Phase I again on a reformulated drug, then more research to figure out why there were horrible side-effects, back to Phase I, then Phase II, and, if the developer is lucky, Phase III. We're talking years after the initial discovery now, with lots of hospital costs, lots of salaries, and *then* the legal stuff starts with the FDA to get approval for general release. Next, lobbying starts on the insurance companies, especially Medicare and Medicaid, to cover treatment with the drug.
Put it this way, there is an entire industry focused specifically on clinical trials, and most drug candidates don't make it through. Because we've set the bar so high to get a drug approved, and the success rate is so low, there must be substantial reward for many people to justify the expense. One researcher having a dream is not enough, despite what Hollywood would have you believe.
You're missing one point here. Foreigners cannot arbitrarily work for a US company: they must have visas to enter the States that specifically allow employment. Those are much harder to get than visitor visas. And, unlike US citizens who can freely go to just about any country on the planet without getting a visa first, in general non-US citizens require a visa (citizens of countries with visa waiver treaties with the US are the exception: like US citizens in many countries abroad, such foreigners essentially get a visa on-the-spot at immigration control; unlike US citizens visiting other countries, foreigners visiting the US also get fingerprinted and have their photo taken). Moreover, a US company cannot arbitrarily hire a foreign citizen.
So if a foreigner wants to start a company in the US, and, especially, employ other foreigners, it's kind of hard without the right visa. If you've got a startup, then physical proximity to potential investors and industrial partners is important.
Think of it this way: if Canada had ultra-lax laws for foreign workers, then Vancouver would be a prime location to house high-tech startups, what with its proximity to Seattle.
I wonder how they're going to handle the atmospherics.
Plumbing, water, the other stuff, that's easy and well-understood. We already know how to pump liquid stuff to change its level by a thousand feet reliably, inside a building. CO2, however, is denser than air, and would tend to accumulate at the bottom. Sure there are plenty of potential answers, I just wonder if they've thought about it since it sounds like the design is not far from a large open hole.
And it was all started by ArsDigita University way back when.
Yes it is a visual dictionary and if it is a cache-miss, then the fallback behavior is to re-parse the word slowly and sound it out. After a few encounters with a strange word it becomes visually cached as well. Parsing a word is far slower, of course, and is not the default behavior.
I have had the opportunity to go from an illiterate but fluent speaker in a language (Greek) to a literate one recently. I had been raised speaking the language, but was never taught to read it. With the advent of children whom I wish to also speak my first language, I've had to teach myself to read. This has been an interesting experience, and I can, from personal experience, 100% confirm the above quoted behavior. There is an additional layer, however, that of context. If you are familiar with the words you are about to read in any way, even if it's because you've read the first word in a commonly used phrase that has appeared in an expected context, memorization of the familiar trumps whole-word recognition.
Yes, another rather face-planting editorial failure. The summary should have been made clear that it was talking about electrically charged gas as a sterilization agent, since growth media derived from bovine blood plasma is standard stuff for filling petri dishes.
Orwell got most of it wrong
You haven't been paying much attention, apparently.
We live in a near police state where citizens are repeatedly urged by friendly, paternalistic voices over loudspeakers in places of common congregation that "if you see something, say something." We are video recorded everywhere (at least in London, but I suspect the same is true in most cities these days). While we don't yet have observation systems directly in our houses, we're coming remarkably close what with technology that sees quite well through walls these days. We must undergo examinations to ride on an airplane (and, soon, trains and busses, if TSA has their way) that 20 years ago would have been thought ludicrous. And if you didn't notice during the Bush administration (the second one), history was rewritten left and right for the convenience of the government. We have always been at war with Eastasia. Perhaps you missed the subtle linguistic shifts during the Bush / Blair partnership? Double plus ungood. The only thing Orwell missed was the speed with which it would happen: it's been somewhat slower than he anticipated, but we're without a doubt on the predicted path.
Bullets are small and cheap.
And trivial to manufacture with readily available low-grade materials. Heck, all you really need is a saltpeter mine, a source of charcoal, a junk yard, and some bronze-age ingenuity.
There are three programs that are vital to my professional life: MS Office, MSIE, and Photoshop.
The vast majority of the people with whom I share MS Office documents are running Windows, and since these people are often holding multi-million dollar purse strings, compatibility is a very serious concern. Sharing documents that have problems with tables or figures not rendering correctly is not an option; I must have a box with Windows on my desk to run MS Office. LibreOffice / OpenOffice is fine for basic, non-critical vanilla documents (and because of a long-standing bug with subscripts and saving to DOC format files, I mean really very basic vanilla documents) but anything beyond that requires MS Office. When you're sending in a grant application to a big governmental agency and the figures don't render correctly, or the document flow suddenly makes your submission a page too long, the governmental agency doesn't cut you slack for using open source software, but instead tosses your application in the trash bin for being out of spec.
In addition, there are a number of internal web applications that our company uses that really do not work well under anything other than MSIE on Windows. And, vitally, if I'm having problems with one of those applications and am not running MSIE on Windows, the IT department will tell me to suck eggs (but they use nicer words). If my use of the web applications isn't successful, my lab gets shut down. Why should I screw around with trying to get these web applications working on Firefox / Konqueror / Chrome / Opera on Linux to work when the stakes are so high?
Because that box is on my desk for MS Office and MSIE, I can run Photoshop on it as well. GIMP is nowhere near good enough (no 16 bit, no CMYK, barely passable cloning tools, inadequate color manipulation, inadequate selection and masking, etc). While I could run Photoshop on a Mac, I only have so much room on my desk, and as my main box runs Fedora, Photoshop gets run on the Windows box.
Frankly, the original question is a bit naive.
In a similar vein, I've often wondered how accurate and extensive the medical knowledge I've gained over the years of watching ER, House, and the other medical dramas on television is. From time to time, I check on things that are said, and usually they're right, if sometimes over-dramatized. But then, would that not be a good way to educate the general public about medical issues? Or, really, anything? I don't know the symptoms of stroke from those stupid acronym signs on the subway (PACE? HELP? STROKE? TIME? whatever -- the advertising campaign didn't work), I know it from watching medical dramas.
Anyone else reminded of the episode of The West Wing where President Bartlet speaks candidly after an interview is supposedly over but the cameras are still rolling? In that fictional case his staff went apoplectic trying to put spin on how it was a mistake, etc., when in reality it was a calculated and savvy move where the President acted in full knowledge that the camera was still on.
The practicalities of limited resources ensure that when committing scarce manpower to monitoring an individual's movements by direct observation, the choice of suspects is expected to be justified by substantial existing evidence, even if warrants have not been obtained. Although we might debate the appropriateness of legal requirements for governmental monitoring of an individual's movements with traditional methods, the limited resource effect ensures that we have a system that at least approximates due process.
With tracking devices, these limitations are lifted and the threshold of justification drops substantially such that, in practical, real terms, we approach a standard of constant monitoring of every individual's every movement when in public, without due process. When it becomes easy and inexpensive to monitor any individual, and the hand of practicality no longer guides selectivity of police action, we must impose legal requirements to protect our citizens' rights. Police are not allowed to go on fishing expeditions, by and large, looking for evidence without any justification.
I'm not a lawyer, though.
Please mod the parent up, that's a remarkably informative reply, especially from an AC.
The important part, if I understand the technique then, is not that we're painting the surface pixel by pixel, as one might expect for an image produced by scanning a focused beam across the asteroid surface to create a 2D image, and as I expected to see in the photo. Instead of a scanning beam, there's a single pulse that gets sent out with some impressively sophisticated processing on the echo allows that signal to be broken down by delay and doppler shift, and those two parameters map only approximately to a side-view relative to an Earth based observer.
It also explains the suspiciously spherical view of the asteroid, and implies that if the asteroid were not spinning relative to the Earth, the technique would degenerate to a 1D image of average reflectivity at a given distance.
I read the articles. I watched the video. But, I'm confused: why does the asteroid appear side-lit in the images?
If we're imaging the asteroid based on radar that's transmitted from the Earth, and the asteroid is heading nearly directly toward us, then we should be able to see images of the asteroid nearly full face on, rather than it appearing like a crescent moon with illumination from the sun, right? The radar illumination is from a source that spatially coincides with the receiving apparatus, so the image should appear more like the full moon.
What am I missing here?
In fact, at the end of the article, in the concluding paragraph, the last line of serious content is:
Once you've ridden the learning curve and spent some time actually getting to know the innards, you may decide you'd be better off running FreeBSD on the next set of Web servers, SMTP relays, or application servers you build
Not in a single place does the article advocate that any variant of BSD is appropriate for the desktop, save Apple's.
someone accidentally caused a nuclear fission without taking proper precautions at a lab
That happened, at least twice. See the WIkipedia entries for Harry Daghlian, Jr and Louis Slotin.
So, as these were "novel" or "neat" "brain-teasers" - they really assessed only if you were some mathematical savant - and had nothing to do with your coding, or even problem-solving abilities.
The real reason to ask questions like that is not to assess the knowledge level of the candidate, but their ability to reason with limited knowledge. You, the potential employer, want to see how they think. For a programmer, that's going to be more important than anything else.
tell me why you think I should hire you
This is arguably the best interview question of all time.