Given that the hydrogen-palladium issue is so wierd, and that nobody knows the exact details of how the supposed fusion mechanism is meant to work, and given the importance of the experimenters being able to check that the result isn't due to some wierd chemical effect,
then perhaps the priority here might be that the experimenters know a bit about electrochemistry.
Since the exact physics of the supposed effect is unknown, being a nuclear physicist might not give one an automatic advantage.
Perhaps what's more important is that the experimenters know palladium, palladium alloy structures, palladium chemistry and palladium surface chemistry inside-out, so that they can rule out known effects.
Perhaps we want the guys conducting these sorts of tests to be highly-trained electrochemists rather than physicists.
So, where would one find experts in electrode chemistry and metallic microstructure, with research experience in metals electrochemistry?
So I guess, if you wanted to use multiple images to boost the resolution of a given scene past the sensor resolution of your camera, you'd take a stack of near-identical photos, use an editing package to blow them all up by a factor of, say, five or ten, so that all the photos show big square single-colour blocks at high magnification that correspond to single pixels in the original images, and then when the software has found the "best-fit" alignment and rotations for the different images, the overlapping transparent blocks should give you "sub-block" detail corresponding to sub-pixel detail at the original resolution?
That sounds like an interesting way of taking ultra-high resolution pictures of small objects without needing hideously expensive lenses and intense lighting. you could even make a custom tripod mount attachment that lets you lock the camera sensor at a fixed distance, and then rotate or slide the camera while you fire off pics.
Ideally, I guess you'd have a "smart" camera with a piezo device that could deliberately waggle the sensor between shots so that you could lock the camera completely and have the camera's own onboard processor fire off a volley of shots and then assemble a higher-res image all by itself (after a wait).
I wonder how long it'll be before this starts turning up on Canon and Sony "consumer" cameras and on cameraphones as a standard feature? It's probably a lot cheaper than improving the optics or the sensors. Downside: you have a moving part to go wrong.
More thinks... if you were building the thing into the camera itself, I guess you could have the lens tilting so that the image is panned diagonally across the sensor. That'd give guaranteed extra resolution without rotating the image. I don't know whether you'd be able to do this accurately to sub-pixel resolution, or whether you'd just use "noise" to pan diagonally and hope for the best. Or just hope that the photographer has a wobbly tripod!:)
I had basically the same idea, but I was going to keep the colour information, blur, include a global colour/contrast value (obtained by resampling to "1x1"), use that to colour-correct the image, and then resample to maybe 5x5.
I figured that for web searches, that should probably be good enough to find lots of alternative images from the same photoshoot or photoset as a sample picture, pictures taken by other photographers of the same scene, or still images of the same scene from a movie.
If you relaxed the search dependency on the global colour value, you should be able to find differently-processed versions of the same image.
I was also going to strip the edge pixels (to remove borders). The use of colour-correction would mean that you wouldn't be wasting code resolution on data that was the same in each cell, and you could identify different version of the same image with different colour "casts", or where operators had played about with the contrast.
It was going to need a bit of R&D and a decent library of sample pictures to work out the best tradeoff between recognisability, color resolution per cell, and final number of cells ("5x5"?, "7x7"?), and once you'd done//that//, there'd probably want to be further R&D to consider possible ways of optimising and future-proofing the "ID code" file structure.
For instance, you might want to put the global colour value first (for finding "exact-match" candidates), followed by the centre cell value, then the surrounding cells ordered by proximity to the centre, based on the assumption that edge information is likely to be less important (if it turned out that the four corner cells weren't too helpful, your search algorithm could ignore the last four cell values).
And then you have issues over whether the number of cells in the ID ought to be fixed or variable, and if its variable, whether it should be recursive. You might base the system on "5x5", but allow the possibility of appended secondary ID code data that subdivides each cell and uses lower-res relative colour offset values for the sub-cells, based on the parent cell's value. Or you might not. It'd be up to the algorithm whether it was going to clip, degrade or weight cell values depending on their ordering.
If you were going along the "recursive" route, then instead of starting with a cell grid that then subdivides, you might divide the grid into concentric zones based on the picture's centre, for instance with a 5x5 grid, you could individually code the relative colours of the three zones: the central cell, the average of all the edge cells, and the average of all the intermediate cells. That'd give three codes that'd describe some of the colour variation across the image without being too sensitive to whether a human subject had moved an arm or leg between shots.
With more processing power you could start moving away from a square grid, and use proper concentric circular zones that are then subdivided, or a pattern of "test blobs" for sampling that approximate subdivided concentric zones (using pre-made bitmask templates for speed), but to start with, a grid approach would probably be more straightforward and simpler to execute.
One of those interesting projects I never got around to following through on...:(
Do the main linux variants have the ability to play mpeg2 video on a typical windows multimedia PC?
Just asking. 'Cos last time I looked they didn't seem to. Maybe that's changed.
I think that if you want people to switch to Linux, people have to be told in advance, honestly, which things are going to work and which aren't. Linux advocates aren't always a good source for this sort of information, they tend to stress all the wonderful things Linux does while forgetting to mention the things that a user might expect to be able to do that aren't properly supported.
OpenOffice is great for Linux in that people can try it on windows first, and gain an expectation of what they're likely to be able to do on Linux in this respect, without having to take some Linux enthusiast's word for it.
Likewise the ASUS EEE. Okay, so it's a limited machine, but the limitations are explicit. If ASUS is saying: "This machine won't to everything, but for the things you'd expect it to do, linux is absolutely fine", then that's good enough for a heck of a lot of buyers.
A lot of people are absolutely fine with the idea that a system doesn't do certain things, as long as it's absolutely clear from the outset what those things are, and that (I'd suggest) is where the Linux community has some credibility problems.
I think I know what's going on here, I've tracked how the amazon stock allocation system works:
I've also had emails from a number of people that have bought our books and report extremely bad service from Amazon, for example, delivery times of two months are common. I suspect that Amazon sits on orders and waits until there are enough from one specialty publisher to attempt to strong-arm the small publisher into a low price.
What's probably happening is that amazon are initially listing your book as out of stock, ETA for new orders (conservatively) 4-6 or 4-8 weeks.
"Customer A" orders a copy regardless, and you immediately send their copy to amazon. The amazon system then registers that copy as "in stock", and sits on it for four to six weeks, on the assumption that the customer doesn't need to have it before then.
During this time, amazon get to show the book as available from stock (next day delivery available). This makes it easier for them to get further sales.
After a week "Customer B" comes along, sees that the book is in stock at amazon, and buys a copy. Amazon then immediately send "Customer B" the copy that was originally ordered in for Customer A, with next day delivery as promised, and reorders another copy from you, for "A". You dutifully send the second copy to amazon immediately. Their system looks at the existing order, realises that it still doesn't have to be supplied for several weeks, and flags it again as as available stock. After another week they sell the second copy to "customer C", who's again seen it marked as in stock and available for immediate delivery.
So poor old Customer A gets screwed, because they ordered the book when it had a long quoted lead-time, and that quoted lead-time gets attached to their order - amazon treats their copy as unallocated and hangs onto it until the last possible moment, and all the successive customers, who are quoted a short lead time, get to leapfrog A's order. A's order is treated as "pseudostock" and only gets sent to A at the last possible moment... assuming that it hasn't just been sold again to someone else.
From amazon's point of view it means that everyone gets sent the book within the promised timeframe (unless A is especially unlucky, which can happen), and amazon get to "stock" the book at zero risk, because their "stock" is guaranteed pre-sold.
From B, C, D's... point of view, amazon are giving a good service, keeping the book in stock and supplying it promptly.
From A's point of view, they don't understand why their mates who ordered the same book after them get sent copies by amazon after a few days, while their copy, which was ordered first, takes two months to arrive.
It's just amazon's clever stock allocation system at work, sometimes it's just a little bit too clever for its own good.
Well, I recently worked at a middle-sized POD plant in their prepress department, and of the books that we were putting onto the system, the overwhelming majority that I saw seemed to be from household-name "traditional publishing companies" with big non-fiction back-catalogues.
I'd see the occasional self-published children's book or novel, and a small trickle of books by local history societies, but the bulk of the books going onto the system seemed to be titles that had already been published conventionally, and were being POD-ised by traditional publishers as a way of making sure that they stayed in print without anyone having to maintain inventory.
So this isn't just impacting on author mills who put out bad novels, it's also hitting the mainstream scientific and technical press and university presses, that put out specialist books on science, engineering, semiconductor fabrication, civil engineering, spacecraft electronics, robotics, mathematics, law, political history, economics, genetics, medical electronics, psychology, social work, marketing, archeology and particle accelerator design... all the non-fiction stuff dear to the hearts of hardcore SlashDotters is likely to be affected.
Amazon isn't preventing anyone from selling POD books in any other market - all they are saying is if you want us to sell your book it needs to be in the following format.
No, amazon recently bought their own POD company ("Booksurge") and the Booksurge reps have allegedly started telling publishers that in future, the only way to get a POD book sold through amazon will be to give amazon's own subsidiary Booksurge the contract to print the thing. So it's apparently not the format itself that amazon are objecting to, but the idea that competitors to their new acquisition are still selling rather a lot of POD books through amazon, without amazon's own POD company making any money from those sales. They are probably not happy that healthy POD sales through amazon are feeding POD companies that compete with their own POD comapny, and so the latest wheeze seems to be to have sales reps tell all the major publishers that in future, if they want to sell POD books through amazon, they'll have to sign up with Booksurge to do it.
Various publishers are expressing disbelief that amazon seem to be using their power as a retailer to dictate to publishers what printing companies they use to print their POD books, as a condition of those books being allowed to be sold by amazon.
If they said you can *only* publish with our vendor, then they might be stifling competition.
For POD books, that's supposed to be exactly what the Booksurge sales reps have been saying.
Apparently the plan for POD books printed by anyone else is that if the customer tries to buy them, they'll find that the "buy" button is mysteriously disabled. This has allegedly already started happening for some POD titles where sales reps have already strongly suggested to a publisher that they might like to start using Booksurge for their POD printing, and the publisher has declined the kind offer.
Terms like "blackmail" are being bandied about.
I think you misunderstand. A traditional publisher only accepts those books that meet certain standards of writing quality.
No, it's not about quality: it's about amazon recognising the growth of the POD book sector, buying a POD printing company themselves, finding that their printing business is failing to get as many contracts as hoped, and finally deciding that, as a desperate last resort to grab more of this market, they'll allow Booksurge reps to scare up contracts by threatening publishers that all their books that are produced using POD technology are going to be blacklisted from amazon, unless they start giving amazon's own POD company contracts to print at least some of them.
It's amazon saying: "Now that we've bought a specialised printing business, we don't want to contribute to the sales of any of that business's competitors, so any books that could have been produced by our new printing company, but aren't, will no longer be available for sale through amazon. We'll list those books to retain our status as a premier listing site, but we won't let people actually buy them as a matter of principle, even if those books are freely available through our standard distributors.
If you're a printing company in competition with our new acquisition, we'll do our best to use our position as a retailer, and our control over amazon's marketplace, to put you out of business: any books that you print for a publisher will effectively be barred from being sold on amazon. If you're a publisher, and want your books to continue to be available for sale on amazon, you'll have to make sure that they aren't printed by any company that we regard as competition for the printing company that we've bought."
It seems like it would require significant work to set up a line to and account with every print-on-demand service an author cares to use.
That argument might work if amazon were just targeting the small POD companies. In fact, they seem to be targeting some of the customers of the largest POD company, Ingram LSI.
Ingram are a major book distributor, and LSI can supply any of about half a million books straight into the distribution chain to both "bricks and mortar" bookshops and to online sellers like amazon. This massive catalogue includes large numbers of specialist academic titles from university publishers. The customers buying these books will often have no idea that they're being printed on POD technology rather than litho. In fact, if you buy an individual POD book through amazon, and it's printed by one of the larger printer/distributors like LSI, amazon may not actually ever see the book themselves. Their computers pass on the order and the payment to LSI, and LSI package it up in a nice amazon box and send it directly to the customer. With POD printing/distribution, not only do the nominal publishers not have to worry about warehousing and handling stock, neither do the online booksellers. It's a good system, that puts some of the more traditional distribution systems to shame. Laser-printed POD-technology books work out significantly more expensive per page than litho printing, so for "popular" titles, litho is still the way to go... but for the established academic presses that might have tens of thousands of "niche" books in their catalogues, migrating them to POD makes a lot of sense.
At this point in the story, almost everything in the garden looks happy. LSI are the largest most integrated supplier but have fixed printing options that don't please everyone: smaller specialist POD companies take up the slack for more specialist POD print jobs that require more human intervention: unusual sizes or cover options, foldouts, inserts, prestige paper, special inks, that sort of thing. Vanity publishers and print-your-own-book services run their own in-house POD printing plant rather than subcontracting, to keep the business in-house, as do certain other speciality publishers. Each has their own niche.
Where the business shakedown started to happen was with the larger independent POD/distribution startup companies that didn't have the niche business of the smaller companies, and couldn't compete with the slickly integrated production service offered by LSI (whose parent company, Ingrams, is one of the most important book industry corporations). One of these companies, "BookSurge", was ambitious, and had the print plant, but had trouble actually getting companies to sign up with them. What they offered wasn't as good as the larger LSI, or the smaller specialist companies. There was no obvious niche for them. So amazon saw an opportunity and bought them out.
And now amazon run their own print-on demand service built around BookSurge.
Snag is, it's not really all that good. It can't offer the flexibility or customer-friendly service of the smaller POD companies, and it can't achieve economies of scale or better integration than LSI, because LSI already take orders directly from the Amazon systems and ship direct. So amazon don't get any additional "process efficiency" by having books printed by their own POD company rather than LSI. What they do get is an extra share of the profits from being the printer and distributor as well as the seller.
Trouble is, that argument only works if their printer-distributor company actually//makes// money, and while Booksurge has had great publicity, it turns out that it doesn't actually seem to offer a sufficiently compelling service for enough people to want to sign up for it. Even with the amazon name behind them, they simply aren't sufficiently competitive.
And so, we have this new development that BookSurge sales reps have started making up lists of
What about this proposition makes it synonymous with nazism?
The "eugenics" part.
The idea that the state can improve society as a whole by compulsorily forcing people to participate in genetic assessments, and that those people may then be dealt with differently by society according to how well they scored, regardless of what their actual abilities are.
It reminds people of the points systems that the Nazis used to categorise degrees of ethnicity and supposed genetic inferiority.
If we thought that social ills were significantly genetic in origin, and that the "solution" was to identify individuals with a particular genetic pattern and do something about it, we'd tend to end up with social eugenics, and the most familiar example to a European of social eugenics as public policy, with enforced participation by the general population, was Nazi Germany.
So the idea that someone's suggesting a scheme where people who haven't done anything wrong have the state mandatorily taking samples from them, for use in some study into some possibly eugenics-related research, it doesn't go down well. Because, again, when people in Europe think of state-enabled doctors doing things to patients without their consent that may not be in the patients interests, they again tend to think of Nazi Germany. The idea that you as an individual don't have the option NOT to participate and don't have the right to decide what you do with your body then ticks the additional entries on the wiki list for "collectivism", "opposition to political liberalism" and "totalitarianism".
There was likely e-mail on those drives that could've had a massive negative effect on the President and his administration, thus it is in our national security interests to see that those records were destroyed.
No. In a democratic system, the needs of the country are more important than those of the President and his administration. The President and his administration are essentially disposable, the country and its constitution aren't.
The system recognises that presidents and administrations can be rotten, and that it's in the country's interests (including its national security interests) for the bulk electorate to have the ability every four years to kick a president out and vote in someone else who they think might do the job better. There are also mechanisms for getting rid of incapable, incompetent or criminal presidents in the middle of a term.
In fact, the system goes further than this: it currently recognises that entrenched centralised power represents such a potential threat to the democratic system that it currently refuses to allow a president to serve more than two full terms even if the electorate want them to. Presidential power and influence is reckoned to be so corrosive that the need to avoid its possible misuse even overrides the natural right of the electorate to vote in who they want.
So no, saving the reputation of a sitting President and his buddies when they've been up to no good is NOT a matter of national security. It's instead a matter of national security that if a bunch of wrong'uns do get into power and misbehave, that we flush the bastards out before they do any more damage to the system and to the country, and that we then try to learn as much as we can about what went wrong so that we are less likely to get caught out again.
When these guys exploited 9-11 and treated it as a blank cheque to allow them to do anything they wanted in the name of national security, they plunged the US into one of the most damaging periods in the country's history. It's a question of national security that we learn from the experience and try not to let anything as half-assed happen again. We need to find out why the system failed, so that we can examine whether further safeguards are needed.
What we can't have is people in power who've screwed things up, illegally destroying the evidence that would tell the US authorities what actually happened, in order to save their own necks and those of their friends. We currently seem to be at least three or four (maybe as many as six) layers deep into apparent illegal cover-ups of other apparent illegal coverups -- national security demands that if there is this level of apparent endemic criminal activity at the White House, whose main purpose seems to be to conceal incompetence, and to conceal criminality conducted not to save the country but to prevent the country from takng standard measures to protect itself from corruption, that we sort the thing out.
George W Bush once said in a speech (about Abu Ghraib) that bad things will always happen under every system, but that the important difference between the United States of America and Saddam Hussein's Iraq was that when people break the law in the US, they go to prison.
Bad things might happen sometimes under the US system, but when they did it was because of isolated individuals who were then caught and punished, no matter how well connected they were (as opposed to the systemic corruption of Saddam's regime, where the guys at the top could get away with anything).
Well, let's see this difference between us and a country we just went to war with to depose their corrupt government being demonstrated. Lets have a full investigation, lets prosecute anyone who hampers it for obstruction, let's have juries deciding whether there's sufficient evidence to send people to prison, and where there's a "grey area" over whether criminality or incompetence or naivete in "following criminal orders" was involved, let's imp
If no Americans want to go, they could subcontract the "personnel" bit to AQ.
The mission logo could be a crescent Mars with two small moons.
Given the choice of death through glorifying Islam by (1) trying to blow up a humvee in Iraq or (2) going to Mars,
I don't think AQ would have trouble finding lots of people who'd prefer to do the Mars thing.
But if a terrorist cell don't mind collateral damage, they don't need to identify individual people: they just need to be able to attack anything that moves at a given location at a given time, or try to destroy anything within a certain radius of a mobile phone signal. The point of autonomous vehicles is that they can get to a given location, at a given time, and do their stuff without needing a radio control signal that can be detected and jammed.
I think that some people in the US military get excited by this stuff because it offers the possibility of using cheap hardware to to mount attacks on high-security targets where a conventional manned attack would tend to be discovered and might have to be a suicide mission. You could get in and do your job, but you probably couldn't get out again. This lets you mount an attack on a building (or a part of a building) that has so much security that it can't be attacked conventionally without all your guys being wiped out.
Trouble is, if you look at the type of sites that these weapons would be most useful against, it'd seem that most of them would belong to the US or the US military. So the proliferation of this sort of technology would mean that the US military would gain the ability to attack a few types of site that are currently quite difficult to get at unless you are a suicide bomber, but the downside is that most US bases and installations (and civilian targets that a terrorist might be interested in) would be far more vulnerable than they are now.
We're effectively developing a technology that levels the playing field between superpowers and small guerilla groups, and if you're a superpower, and there are dozens of small groups out there that wish you harm, then that's probably not a great idea. By trying to make these sorts of attacks lower-cost (both in terms of technology and manpower), this technology also lowers the logistical threshold for attacks on US targets.
You've just hit on one of the factors that makes this technology more useful to terrorists than to conventional military.
If you have a mix of robots and conventional soldiers, then there's a high risk that some of the robots are going to start killing some of your own guys. But if you're mounting a terrorist attack on an installation, without any of your own personnel being present, then the robot can simply kill anyone it finds - targets, police, civilians, it doesn't matter. You don't need friend-of-foe systems when everyone is a target.
A second problem that affects superpowers but not terrorists is the question of what happens when you instigate a 600 million dollar automated UAV programme, and one of the things crashes and falls into enemy hands through sheer bad luck. Suddenly you then have a group out there somewhere with a stealthed machine that can take out presidents or CEOs of corporations, without you being able to see it coming. So having built your machines, you find yourself hardly ever using them, for fear of losing one. This is supposed to be one of the problems that hampered the Allies towards the end of WW2: they had jet-engined planes, but didn't dare fly them over Germany in case one got shot down, and the Germans then got to take a look at it.
A third problem is that these devices are great at circumventing security, so their ideal application is use by small groups against large high-security targets. In other words, they may make the US a little bit better at attacking terrorists, but they make terrorists much more effective at attacking US targets.
A little weeny UAV can fly over security fences and into secure bases. It can whizz over the concrete wall around the green zone and go looking for the visiting commander in chief. It can reach any window of the Pentagon, in- or out-facing, and it can attack any window of any skyscraper. You can be taking a crap in the exacutive loo on the fortieth floor of the most high-security building in New York, and one of these things can shoot you through the wall from outside. These things could hop over airport perimeter fences and take out airliners as they come in to land, shoot politicians as they give speeches, wipe out entire company boards while they are in session, or take out Air Force One while it's still on the tarmac.
These things are about as close as you can get to the perfect terrorist device, and we're promoting their development.
Remember, once we've ironed out all the bugs in the control systems and turned this into a 99.99% reliable technology, and conducted the "proof of concept" tests, the terrorists can tap into cheaper versions of that proven technology fairly easily.
They don't need the same reliability or range. We want a UAV with a 200km range that can execute a mission and return safely to base, they only need something that can go maybe a kilometre and blow up.
So... basically, the US is developing and promoting technologies whose main "killer application" would seem to be that they allow small groups to attack the US in ways they never could before, without having to risk their own lives. Our foreign policy assumes that we have superpower advantage, in that we can attack people who don't have the ability to attack us back in any meaningful way , but this is a a "leveling technology" that takes away that superpower advantage and means that anyone can attack anyone, without the people being attacked necessarily knowing who did it.
It's a deeply destabilising technology that favors countries and groups who don't like the current status quo, and where current superpowers have the most to lose, and because it doesn't require special hard-to-handle materials, its easily transferrable.
So the people who work on these research projects probably need to be monitored and tracked as carefully as nuclear weaponry scientists or bioweapons researchers.
They'd be great if you were stuck on a moonbase or a space station, and you desperately needed some little replacement widget to fix a piece of machinery. Suppose that you lose a 3mm hexagonal nut (worth a fraction of a cent), and the nearest replacement is six months away. Go to your fabricator, call up the specs of the machine being repaired, call up the component list, dial up the spec of a 3mm hex nut, and print one off.
Saves carrying along loads of redundant spare parts.
Every space colony should have one!
(actually, every space colony should have at least two, so that when one breaks down, the second machine can print spare parts for it).
Yeah, but I thought that "totality" usually referred to a combination of both time and space coordinates (in other words, that it only applied to observers within the zone being totally blacked out, at the time that it is totally blacked out).
It's like, if someone says that something is going to happen in the sky at a location "at sunrise", you tend to hope that they mean sunrise at that location, otherwise it's a bit of a misleading thing to say.
If the shooting down wasn't actually scheduled
"during the totality of Wednesday's lunar eclipse"
with respect to the area of the shooting-down zone, then I think it was a very misleading article. I mean, if I say that something happened in London "during a total eclipse", you're going to tend to assume that I'm referring to a total eclipse in London, aren't you? As opposed to something happening in London during a total eclipse in, say, Ecuador.
If you're saying that the article got its wires crossed, or took a few liberties with phrasing in order to make a better story, I'll take your word for it.
Yeah, but it's not meant for typing on, is it? 'Cos the people who spend so much time typing on a keyboard that they can consider blowing fifteen hundred bucks on one... are already going to know where all the keys are.
No, this is for people doing video editing or music production or other multimedia editing, where you might easily have a couple of hundred functions tucked away behind various Ctrl-Shift-Alt key combinations, and which change depending on which edit screen you're in, or which function key you just pressed. If you're in an audio editor, and you mark out a section of audio, there might easily be forty or fifty different functions that you might want to apply to that block: cut/copy/paste/save-as-file/silence/optimise/filter/replace/retune/add-to-library... the list goes on and on (when I was prototyping an all-out audio editor once, I think I had about sixty different region-edit functions).
If you're using one of these programs, the main function of the keyboard isn't inputting text, it's launching functions and actions by key-command shortcut so that the user doesn't have to dig through menus and dialog boxes. And of course, the big problem is that although a keyboard has enough buttons to launch all these functions, they aren't written on the keys, and even if you buy a custom keyboard for something like Logic (with the commands printed on the key-caps), you don't have context-sensitivity or proper customisability, and if the company adds or changes key-commands on a new software update, you're left behind.
If you use a couple of different audio editing apps and a couple of video editors, plus a few other bits of specialist software, plus photoshop, and you can't face the idea of ordering seven different custom keyboards and finding some way to switch between them, then this is probably a very nice gadget for a cramped pre-production studio.
Keep a cheap generic keyboard tucked away under the desk for those times that you need to do some serious typing.
LinuxGuy: "You know, I really don't know what's wrong with users. Here we have a perfectly usable graphics utility program on Linux, and it never seems to get the press it deserves. Okay, so it doesn't have a lot of the "pro" features that people tend to expect in the big commercial packages, and maybe it has a bit of a learning curve to it, but it does pretty much most of the stuff that people need day-to-day. But for some reason people just don't take it seriously. I really don't understand what it is that we're doing wrong."
WindowsGuy: "Well, earning an end-user's trust and respect isn't easy. Maybe it's your GUI, maybe you have poor online documentation, or perhaps there's some other problem with it that's more difficult to pin down. Perhaps there's some little flaw in the presentation that potential users find offputting, that gives out the wrong signals and makes them less likely to trust your program or to invest their time in it. Let me download it and take a look at it for you. What's this program called?"
Slashdot Mirror
Since the exact physics of the supposed effect is unknown, being a nuclear physicist might not give one an automatic advantage.
Perhaps what's more important is that the experimenters know palladium, palladium alloy structures, palladium chemistry and palladium surface chemistry inside-out, so that they can rule out known effects. Perhaps we want the guys conducting these sorts of tests to be highly-trained electrochemists rather than physicists.
So, where would one find experts in electrode chemistry and metallic microstructure, with research experience in metals electrochemistry?
Perhaps ... at a welding research institute?
Just sayin' ...
Yeah, or dropout brothers who own bicycle repair shops claiming to have built heavier-than-air flying machines. Ridiculous!
So I guess, if you wanted to use multiple images to boost the resolution of a given scene past the sensor resolution of your camera, you'd take a stack of near-identical photos, use an editing package to blow them all up by a factor of, say, five or ten, so that all the photos show big square single-colour blocks at high magnification that correspond to single pixels in the original images, and then when the software has found the "best-fit" alignment and rotations for the different images, the overlapping transparent blocks should give you "sub-block" detail corresponding to sub-pixel detail at the original resolution?
That sounds like an interesting way of taking ultra-high resolution pictures of small objects without needing hideously expensive lenses and intense lighting. you could even make a custom tripod mount attachment that lets you lock the camera sensor at a fixed distance, and then rotate or slide the camera while you fire off pics.
Ideally, I guess you'd have a "smart" camera with a piezo device that could deliberately waggle the sensor between shots so that you could lock the camera completely and have the camera's own onboard processor fire off a volley of shots and then assemble a higher-res image all by itself (after a wait).
I wonder how long it'll be before this starts turning up on Canon and Sony "consumer" cameras and on cameraphones as a standard feature? It's probably a lot cheaper than improving the optics or the sensors. Downside: you have a moving part to go wrong.
More thinks ... if you were building the thing into the camera itself, I guess you could have the lens tilting so that the image is panned diagonally across the sensor. That'd give guaranteed extra resolution without rotating the image. I don't know whether you'd be able to do this accurately to sub-pixel resolution, or whether you'd just use "noise" to pan diagonally and hope for the best. Or just hope that the photographer has a wobbly tripod! :)
The number of angels that could fit onto the surface of a 1mm square pinhead depends on the average information-content per angel.
>I dare you to tell me Hawking doesn't secretly control a robot army! Wow! How cool would that be as a plotline on Dr Who! Someone call his agent!
I had basically the same idea, but I was going to keep the colour information, blur, include a global colour/contrast value (obtained by resampling to "1x1"), use that to colour-correct the image, and then resample to maybe 5x5.
I figured that for web searches, that should probably be good enough to find lots of alternative images from the same photoshoot or photoset as a sample picture, pictures taken by other photographers of the same scene, or still images of the same scene from a movie.
If you relaxed the search dependency on the global colour value, you should be able to find differently-processed versions of the same image. I was also going to strip the edge pixels (to remove borders). The use of colour-correction would mean that you wouldn't be wasting code resolution on data that was the same in each cell, and you could identify different version of the same image with different colour "casts", or where operators had played about with the contrast.
It was going to need a bit of R&D and a decent library of sample pictures to work out the best tradeoff between recognisability, color resolution per cell, and final number of cells ("5x5"?, "7x7"?), and once you'd done //that//, there'd probably want to be further R&D to consider possible ways of optimising and future-proofing the "ID code" file structure.
For instance, you might want to put the global colour value first (for finding "exact-match" candidates), followed by the centre cell value, then the surrounding cells ordered by proximity to the centre, based on the assumption that edge information is likely to be less important (if it turned out that the four corner cells weren't too helpful, your search algorithm could ignore the last four cell values).
And then you have issues over whether the number of cells in the ID ought to be fixed or variable, and if its variable, whether it should be recursive. You might base the system on "5x5", but allow the possibility of appended secondary ID code data that subdivides each cell and uses lower-res relative colour offset values for the sub-cells, based on the parent cell's value. Or you might not. It'd be up to the algorithm whether it was going to clip, degrade or weight cell values depending on their ordering.
If you were going along the "recursive" route, then instead of starting with a cell grid that then subdivides, you might divide the grid into concentric zones based on the picture's centre, for instance with a 5x5 grid, you could individually code the relative colours of the three zones: the central cell, the average of all the edge cells, and the average of all the intermediate cells. That'd give three codes that'd describe some of the colour variation across the image without being too sensitive to whether a human subject had moved an arm or leg between shots.
With more processing power you could start moving away from a square grid, and use proper concentric circular zones that are then subdivided, or a pattern of "test blobs" for sampling that approximate subdivided concentric zones (using pre-made bitmask templates for speed), but to start with, a grid approach would probably be more straightforward and simpler to execute.
One of those interesting projects I never got around to following through on ... :(
Just asking. 'Cos last time I looked they didn't seem to. Maybe that's changed.
I think that if you want people to switch to Linux, people have to be told in advance, honestly, which things are going to work and which aren't. Linux advocates aren't always a good source for this sort of information, they tend to stress all the wonderful things Linux does while forgetting to mention the things that a user might expect to be able to do that aren't properly supported.
OpenOffice is great for Linux in that people can try it on windows first, and gain an expectation of what they're likely to be able to do on Linux in this respect, without having to take some Linux enthusiast's word for it.
Likewise the ASUS EEE. Okay, so it's a limited machine, but the limitations are explicit. If ASUS is saying: "This machine won't to everything, but for the things you'd expect it to do, linux is absolutely fine", then that's good enough for a heck of a lot of buyers.
A lot of people are absolutely fine with the idea that a system doesn't do certain things, as long as it's absolutely clear from the outset what those things are, and that (I'd suggest) is where the Linux community has some credibility problems.
"Customer A" orders a copy regardless, and you immediately send their copy to amazon. The amazon system then registers that copy as "in stock", and sits on it for four to six weeks, on the assumption that the customer doesn't need to have it before then. During this time, amazon get to show the book as available from stock (next day delivery available). This makes it easier for them to get further sales.
After a week "Customer B" comes along, sees that the book is in stock at amazon, and buys a copy. Amazon then immediately send "Customer B" the copy that was originally ordered in for Customer A, with next day delivery as promised, and reorders another copy from you, for "A". You dutifully send the second copy to amazon immediately. Their system looks at the existing order, realises that it still doesn't have to be supplied for several weeks, and flags it again as as available stock. After another week they sell the second copy to "customer C", who's again seen it marked as in stock and available for immediate delivery.
So poor old Customer A gets screwed, because they ordered the book when it had a long quoted lead-time, and that quoted lead-time gets attached to their order - amazon treats their copy as unallocated and hangs onto it until the last possible moment, and all the successive customers, who are quoted a short lead time, get to leapfrog A's order. A's order is treated as "pseudostock" and only gets sent to A at the last possible moment ... assuming that it hasn't just been sold again to someone else.
From amazon's point of view it means that everyone gets sent the book within the promised timeframe (unless A is especially unlucky, which can happen), and amazon get to "stock" the book at zero risk, because their "stock" is guaranteed pre-sold. ... point of view, amazon are giving a good service, keeping the book in stock and supplying it promptly.
From B, C, D's
From A's point of view, they don't understand why their mates who ordered the same book after them get sent copies by amazon after a few days, while their copy, which was ordered first, takes two months to arrive.
It's just amazon's clever stock allocation system at work, sometimes it's just a little bit too clever for its own good.
I'd see the occasional self-published children's book or novel, and a small trickle of books by local history societies, but the bulk of the books going onto the system seemed to be titles that had already been published conventionally, and were being POD-ised by traditional publishers as a way of making sure that they stayed in print without anyone having to maintain inventory.
So this isn't just impacting on author mills who put out bad novels, it's also hitting the mainstream scientific and technical press and university presses, that put out specialist books on science, engineering, semiconductor fabrication, civil engineering, spacecraft electronics, robotics, mathematics, law, political history, economics, genetics, medical electronics, psychology, social work, marketing, archeology and particle accelerator design ... all the non-fiction stuff dear to the hearts of hardcore SlashDotters is likely to be affected.
http://online.wsj.com/article/SB120667525724970997.html?mod=googlenews_wsj
For POD books, that's supposed to be exactly what the Booksurge sales reps have been saying. Apparently the plan for POD books printed by anyone else is that if the customer tries to buy them, they'll find that the "buy" button is mysteriously disabled. This has allegedly already started happening for some POD titles where sales reps have already strongly suggested to a publisher that they might like to start using Booksurge for their POD printing, and the publisher has declined the kind offer.http://www.writersweekly.com/the_latest_from_angelahoycom/004597_03272008.html
Terms like "blackmail" are being bandied about.
No, it's not about quality: it's about amazon recognising the growth of the POD book sector, buying a POD printing company themselves, finding that their printing business is failing to get as many contracts as hoped, and finally deciding that, as a desperate last resort to grab more of this market, they'll allow Booksurge reps to scare up contracts by threatening publishers that all their books that are produced using POD technology are going to be blacklisted from amazon, unless they start giving amazon's own POD company contracts to print at least some of them.
It's amazon saying: "Now that we've bought a specialised printing business, we don't want to contribute to the sales of any of that business's competitors, so any books that could have been produced by our new printing company, but aren't, will no longer be available for sale through amazon. We'll list those books to retain our status as a premier listing site, but we won't let people actually buy them as a matter of principle, even if those books are freely available through our standard distributors. If you're a printing company in competition with our new acquisition, we'll do our best to use our position as a retailer, and our control over amazon's marketplace, to put you out of business: any books that you print for a publisher will effectively be barred from being sold on amazon. If you're a publisher, and want your books to continue to be available for sale on amazon, you'll have to make sure that they aren't printed by any company that we regard as competition for the printing company that we've bought."
That argument might work if amazon were just targeting the small POD companies. In fact, they seem to be targeting some of the customers of the largest POD company, Ingram LSI.
Ingram are a major book distributor, and LSI can supply any of about half a million books straight into the distribution chain to both "bricks and mortar" bookshops and to online sellers like amazon. This massive catalogue includes large numbers of specialist academic titles from university publishers. The customers buying these books will often have no idea that they're being printed on POD technology rather than litho. In fact, if you buy an individual POD book through amazon, and it's printed by one of the larger printer/distributors like LSI, amazon may not actually ever see the book themselves. Their computers pass on the order and the payment to LSI, and LSI package it up in a nice amazon box and send it directly to the customer. With POD printing/distribution, not only do the nominal publishers not have to worry about warehousing and handling stock, neither do the online booksellers. It's a good system, that puts some of the more traditional distribution systems to shame. Laser-printed POD-technology books work out significantly more expensive per page than litho printing, so for "popular" titles, litho is still the way to go ... but for the established academic presses that might have tens of thousands of "niche" books in their catalogues, migrating them to POD makes a lot of sense.
At this point in the story, almost everything in the garden looks happy. LSI are the largest most integrated supplier but have fixed printing options that don't please everyone: smaller specialist POD companies take up the slack for more specialist POD print jobs that require more human intervention: unusual sizes or cover options, foldouts, inserts, prestige paper, special inks, that sort of thing. Vanity publishers and print-your-own-book services run their own in-house POD printing plant rather than subcontracting, to keep the business in-house, as do certain other speciality publishers. Each has their own niche.
Where the business shakedown started to happen was with the larger independent POD/distribution startup companies that didn't have the niche business of the smaller companies, and couldn't compete with the slickly integrated production service offered by LSI (whose parent company, Ingrams, is one of the most important book industry corporations). One of these companies, "BookSurge", was ambitious, and had the print plant, but had trouble actually getting companies to sign up with them. What they offered wasn't as good as the larger LSI, or the smaller specialist companies. There was no obvious niche for them. So amazon saw an opportunity and bought them out.
And now amazon run their own print-on demand service built around BookSurge.
Snag is, it's not really all that good. It can't offer the flexibility or customer-friendly service of the smaller POD companies, and it can't achieve economies of scale or better integration than LSI, because LSI already take orders directly from the Amazon systems and ship direct. So amazon don't get any additional "process efficiency" by having books printed by their own POD company rather than LSI. What they do get is an extra share of the profits from being the printer and distributor as well as the seller.
Trouble is, that argument only works if their printer-distributor company actually //makes// money, and while Booksurge has had great publicity, it turns out that it doesn't actually seem to offer a sufficiently compelling service for enough people to want to sign up for it. Even with the amazon name behind them, they simply aren't sufficiently competitive.
And so, we have this new development that BookSurge sales reps have started making up lists of
The "eugenics" part.
The idea that the state can improve society as a whole by compulsorily forcing people to participate in genetic assessments, and that those people may then be dealt with differently by society according to how well they scored, regardless of what their actual abilities are.
It reminds people of the points systems that the Nazis used to categorise degrees of ethnicity and supposed genetic inferiority.
If we thought that social ills were significantly genetic in origin, and that the "solution" was to identify individuals with a particular genetic pattern and do something about it, we'd tend to end up with social eugenics, and the most familiar example to a European of social eugenics as public policy, with enforced participation by the general population, was Nazi Germany.
So the idea that someone's suggesting a scheme where people who haven't done anything wrong have the state mandatorily taking samples from them, for use in some study into some possibly eugenics-related research, it doesn't go down well. Because, again, when people in Europe think of state-enabled doctors doing things to patients without their consent that may not be in the patients interests, they again tend to think of Nazi Germany. The idea that you as an individual don't have the option NOT to participate and don't have the right to decide what you do with your body then ticks the additional entries on the wiki list for "collectivism", "opposition to political liberalism" and "totalitarianism".
No. In a democratic system, the needs of the country are more important than those of the President and his administration. The President and his administration are essentially disposable, the country and its constitution aren't.
The system recognises that presidents and administrations can be rotten, and that it's in the country's interests (including its national security interests) for the bulk electorate to have the ability every four years to kick a president out and vote in someone else who they think might do the job better. There are also mechanisms for getting rid of incapable, incompetent or criminal presidents in the middle of a term.
In fact, the system goes further than this: it currently recognises that entrenched centralised power represents such a potential threat to the democratic system that it currently refuses to allow a president to serve more than two full terms even if the electorate want them to. Presidential power and influence is reckoned to be so corrosive that the need to avoid its possible misuse even overrides the natural right of the electorate to vote in who they want.
So no, saving the reputation of a sitting President and his buddies when they've been up to no good is NOT a matter of national security. It's instead a matter of national security that if a bunch of wrong'uns do get into power and misbehave, that we flush the bastards out before they do any more damage to the system and to the country, and that we then try to learn as much as we can about what went wrong so that we are less likely to get caught out again.
When these guys exploited 9-11 and treated it as a blank cheque to allow them to do anything they wanted in the name of national security, they plunged the US into one of the most damaging periods in the country's history. It's a question of national security that we learn from the experience and try not to let anything as half-assed happen again. We need to find out why the system failed, so that we can examine whether further safeguards are needed.
What we can't have is people in power who've screwed things up, illegally destroying the evidence that would tell the US authorities what actually happened, in order to save their own necks and those of their friends. We currently seem to be at least three or four (maybe as many as six) layers deep into apparent illegal cover-ups of other apparent illegal coverups -- national security demands that if there is this level of apparent endemic criminal activity at the White House, whose main purpose seems to be to conceal incompetence, and to conceal criminality conducted not to save the country but to prevent the country from takng standard measures to protect itself from corruption, that we sort the thing out.
George W Bush once said in a speech (about Abu Ghraib) that bad things will always happen under every system, but that the important difference between the United States of America and Saddam Hussein's Iraq was that when people break the law in the US, they go to prison. Bad things might happen sometimes under the US system, but when they did it was because of isolated individuals who were then caught and punished, no matter how well connected they were (as opposed to the systemic corruption of Saddam's regime, where the guys at the top could get away with anything).
Well, let's see this difference between us and a country we just went to war with to depose their corrupt government being demonstrated. Lets have a full investigation, lets prosecute anyone who hampers it for obstruction, let's have juries deciding whether there's sufficient evidence to send people to prison, and where there's a "grey area" over whether criminality or incompetence or naivete in "following criminal orders" was involved, let's imp
The mission logo could be a crescent Mars with two small moons.
Given the choice of death through glorifying Islam by (1) trying to blow up a humvee in Iraq or (2) going to Mars, I don't think AQ would have trouble finding lots of people who'd prefer to do the Mars thing.
I think that some people in the US military get excited by this stuff because it offers the possibility of using cheap hardware to to mount attacks on high-security targets where a conventional manned attack would tend to be discovered and might have to be a suicide mission. You could get in and do your job, but you probably couldn't get out again. This lets you mount an attack on a building (or a part of a building) that has so much security that it can't be attacked conventionally without all your guys being wiped out.
Trouble is, if you look at the type of sites that these weapons would be most useful against, it'd seem that most of them would belong to the US or the US military. So the proliferation of this sort of technology would mean that the US military would gain the ability to attack a few types of site that are currently quite difficult to get at unless you are a suicide bomber, but the downside is that most US bases and installations (and civilian targets that a terrorist might be interested in) would be far more vulnerable than they are now.
We're effectively developing a technology that levels the playing field between superpowers and small guerilla groups, and if you're a superpower, and there are dozens of small groups out there that wish you harm, then that's probably not a great idea. By trying to make these sorts of attacks lower-cost (both in terms of technology and manpower), this technology also lowers the logistical threshold for attacks on US targets.
If you have a mix of robots and conventional soldiers, then there's a high risk that some of the robots are going to start killing some of your own guys. But if you're mounting a terrorist attack on an installation, without any of your own personnel being present, then the robot can simply kill anyone it finds - targets, police, civilians, it doesn't matter. You don't need friend-of-foe systems when everyone is a target.
A second problem that affects superpowers but not terrorists is the question of what happens when you instigate a 600 million dollar automated UAV programme, and one of the things crashes and falls into enemy hands through sheer bad luck. Suddenly you then have a group out there somewhere with a stealthed machine that can take out presidents or CEOs of corporations, without you being able to see it coming. So having built your machines, you find yourself hardly ever using them, for fear of losing one. This is supposed to be one of the problems that hampered the Allies towards the end of WW2: they had jet-engined planes, but didn't dare fly them over Germany in case one got shot down, and the Germans then got to take a look at it.
A third problem is that these devices are great at circumventing security, so their ideal application is use by small groups against large high-security targets. In other words, they may make the US a little bit better at attacking terrorists, but they make terrorists much more effective at attacking US targets.
A little weeny UAV can fly over security fences and into secure bases. It can whizz over the concrete wall around the green zone and go looking for the visiting commander in chief. It can reach any window of the Pentagon, in- or out-facing, and it can attack any window of any skyscraper. You can be taking a crap in the exacutive loo on the fortieth floor of the most high-security building in New York, and one of these things can shoot you through the wall from outside. These things could hop over airport perimeter fences and take out airliners as they come in to land, shoot politicians as they give speeches, wipe out entire company boards while they are in session, or take out Air Force One while it's still on the tarmac.
These things are about as close as you can get to the perfect terrorist device, and we're promoting their development.
Remember, once we've ironed out all the bugs in the control systems and turned this into a 99.99% reliable technology, and conducted the "proof of concept" tests, the terrorists can tap into cheaper versions of that proven technology fairly easily. They don't need the same reliability or range. We want a UAV with a 200km range that can execute a mission and return safely to base, they only need something that can go maybe a kilometre and blow up.
So ... basically, the US is developing and promoting technologies whose main "killer application" would seem to be that they allow small groups to attack the US in ways they never could before, without having to risk their own lives. Our foreign policy assumes that we have superpower advantage, in that we can attack people who don't have the ability to attack us back in any meaningful way , but this is a a "leveling technology" that takes away that superpower advantage and means that anyone can attack anyone, without the people being attacked necessarily knowing who did it.
It's a deeply destabilising technology that favors countries and groups who don't like the current status quo, and where current superpowers have the most to lose, and because it doesn't require special hard-to-handle materials, its easily transferrable. So the people who work on these research projects probably need to be monitored and tracked as carefully as nuclear weaponry scientists or bioweapons researchers.
Saves carrying along loads of redundant spare parts.
Every space colony should have one! (actually, every space colony should have at least two, so that when one breaks down, the second machine can print spare parts for it).
The RAF sometimes makes its planes last even longer: http://en.wikipedia.org/wiki/De_Havilland_Comet
NOT the satellite being shot at when the moon goes in front of the Sun.
In which case ... what a desperately boring story!
It's like, if someone says that something is going to happen in the sky at a location "at sunrise", you tend to hope that they mean sunrise at that location, otherwise it's a bit of a misleading thing to say.
If the shooting down wasn't actually scheduled
with respect to the area of the shooting-down zone, then I think it was a very misleading article. I mean, if I say that something happened in London "during a total eclipse", you're going to tend to assume that I'm referring to a total eclipse in London, aren't you? As opposed to something happening in London during a total eclipse in, say, Ecuador.If you're saying that the article got its wires crossed, or took a few liberties with phrasing in order to make a better story, I'll take your word for it.
No, this is for people doing video editing or music production or other multimedia editing, where you might easily have a couple of hundred functions tucked away behind various Ctrl-Shift-Alt key combinations, and which change depending on which edit screen you're in, or which function key you just pressed. If you're in an audio editor, and you mark out a section of audio, there might easily be forty or fifty different functions that you might want to apply to that block: cut/copy/paste/save-as-file/silence/optimise/filter/replace/retune/add-to-library ... the list goes on and on (when I was prototyping an all-out audio editor once, I think I had about sixty different region-edit functions).
If you're using one of these programs, the main function of the keyboard isn't inputting text, it's launching functions and actions by key-command shortcut so that the user doesn't have to dig through menus and dialog boxes. And of course, the big problem is that although a keyboard has enough buttons to launch all these functions, they aren't written on the keys, and even if you buy a custom keyboard for something like Logic (with the commands printed on the key-caps), you don't have context-sensitivity or proper customisability, and if the company adds or changes key-commands on a new software update, you're left behind. If you use a couple of different audio editing apps and a couple of video editors, plus a few other bits of specialist software, plus photoshop, and you can't face the idea of ordering seven different custom keyboards and finding some way to switch between them, then this is probably a very nice gadget for a cramped pre-production studio.
Keep a cheap generic keyboard tucked away under the desk for those times that you need to do some serious typing.
WindowsGuy: "Well, earning an end-user's trust and respect isn't easy. Maybe it's your GUI, maybe you have poor online documentation, or perhaps there's some other problem with it that's more difficult to pin down. Perhaps there's some little flaw in the presentation that potential users find offputting, that gives out the wrong signals and makes them less likely to trust your program or to invest their time in it. Let me download it and take a look at it for you. What's this program called?"
LinuxGuy: "Cocksucker."
If it's really going to be happening during eclipse totality, that doesn't sound like "broad daylight" ...
Oooo! Clever!