Although I have seen some kernel patches for AMD power management, they don't work on any of my numerous systems, and I've heard reports that on systems where they will work, it causes serious system instability.
Well, I have an older 500 MHz Athlon at home that hangs the network cards when I try to run 'fvcool' (Linux), but my (now quite old) 2200+ here works beautifully. Stable as a rock, and no performance degradation either. Before I ran 'fvcool' it would remain constantly at 72 dec C, but without it it's down to an easier 38 dec C. (Don't ask, I was limited to one builder and they weren't AMD savy.)
Scratchbuilding the V would give us the workhorse that NASA has been looking for that will put us back on the moon to stay, as well as other missions.
Yes well, I'm not saying that a big honking rocket isn't the way to go. Combined with a wingless capsule type craft to put people in, it'll be both safer and more cost efficient. However, once you're done 'rebuilding' the Saturn V with modern technology, you'll in effect have built a new rocket. It's no longer the Saturn V. And the Russians have some nifty 'hot pump' engines you'd like to put on it, though the design for those is quite old. Thinking that you're only using tested parts and that'll speed up reconstruction won't work as the first Ariane 5 failure demonstrated. It's the knowledge you can reuse, not the design.
And having the blue prints isn't much help in that respect. Lots of the engineering knowledge that went into the Saturn V have been lost. You have the design now, but much of the reasoning that went into that design is gone. And you really need that when you're going to modify/update the design. The Soviets made a few spectacular errors in this regard, stealing western aircraft designs (or downed planes) and building something that didn't work at all as advertised (Though it should be noted that the TU-144 Concordski crash was really due to other reasons).
So by all means, go for the big honking rocket, but name it the Saturn VI to help remembering what it is that you're really building.
I heard someone in NASA management had all the blueprints to the Apollo rockets destroyed so there would be no choice but to build the shuttle. So it would cost billions to reconstitute the Saturn program -- they'd have to basically start from scratch.
And many others have heard that also, so many in fact that it qualifies as an urband legend.
It's not true however. They're still there. The problem is that you couldn't get the parts (sixties vintage) today, and the launch pads have been rebuilt. We've also learnt a thing or two since the sixties. Once you've resourced the parts and rebuilt the launch pads you might as well have started from scratch and gotten a better vehicle for it.
Caching is controlled completely by the CPU, transparent of the programmer.
Not really true these days. Some architectures do have an explicit 'cache preload' instruction, such as the SPARC V9 and the ARM9E to mention two. These allow the programmer to preload a D-cache line before it is needed.
As the speed of the CPU has increased much faster than the speed of main memory (and hence increasing the relative cost of a cache mis), compiler based techniques to emit cache preload instructions in advance, before the data is actually needed, has been the subject of some research in the past 7-8 years. The main reason to do it is software, instead of hardware, is that the compiler have a greater knowledge of the layout of the entire task, as it can 'look ahead' in the source code. The main disadvantage is that any static analysis, of course won't have access to dynamic (run time) data about the program as it is running.
If you wish to go further, you could do worse than to start with my former colleague Magnus Karlsson's PhD thesis on the subject:
Magnus Karlsson: "Data Prefetching Techniques Targeting Single and a Network of Processing Nodes". Ph. D. thesis, Department of Computer Engineering, Chalmers University of Technology, December 1999.
And of course as always, Google and citeseer is your friend.
In other words, just trust the CPU. It knows what it's doing:).
Well, actually, it doesn't these days...:-) Trust the compiler instead:-) (Yeah, right).
But it isn't a measure of how much energy is transmitted, at twice the distance, a fourth the pressure is there, that means any sound pressure, dB, whatever is useless without a distance from the "center" of the transmitter.
Ah, we're into nit picking are we? In that case, you have to qualify that you're thinking of a point source in a completely echo free environment. Which won't typically be the case in a computer case.
No, seriously, you make a good point. Of course if you started your fan in Mongolia it could be pretty powerful and still be considered silent here (for most definitions of 'here' not actually close to Mongolia). Hence, how the measurement is made really is an important factor. I'd take a 21dB fan at 10cm over a 14dB fan at 1.5 meter any day.
Perhaps instead of reading my post with such sarcastic bitterness you should have seen it for what it is and was
Oh, I wouldn't be too hard on the guy. You did come across quite a bit like that to me to.
My machine is set up simlaraly to yours, and knowing what I know now, I would have taken some of the extra money I spent on soundproofing my machine and put it into upgrading some of its other components instead.
Well, you may want to think about your wording in the future then, that's not at all how I got the tone of your post. Just a friendly advice, take it or leave it.
Myself, I'd argue the opposite angle. Even my old computer is plenty fast/big for anything I need/want to do with it, even the (few) games I play. If it hadn't been quiet (I built an MDF sound proof hood for it) I couldn't have it on all the time conveniently situated in my living room, which makes TV-out a delight for example.
I mean, what use is an epsilon faster computer, costing you hundreds of dollars more for very little speed up, if it's so damn noisy that you think twice before turning it on (not to mention the complaints from the rest of the family, especially later in the evening), and turning it off as fast as you can? Spend a reasonable amount of dosh on quiet I say, and choose a slower processor, slightly smaller disk etc. You wont regret it.
Cranking up the TV/stereo to drown out the noise of the computer is not an option for most of us.
No, he's saying - as hard as it is to believe - that Bush is smarter than a mail server.
Actually, he's saying that the guy that works in the whitehouse mail room is smarter than a mail server. Whether that makes him smarter than his boss, I think is a question of some debate, though I know that I'm certainly leaning towards that possiblitity.
The kinds of mines that the plants are probably used to find are anti-personnel mines dug into the ground. These usually consist of an explosive charge 10cm wide and about 10cm high with some shrapnelling stuff in them.
It should be noted that these are by now the old style anti personell mine, not in wide spread use since the early/mid eighties. The problem with providing a casing (or simply a plate) to provide shrappnel is that it increases the weight of the mine, and also the metallic content which makes clearing easier (via the use of mine detectors).
Modern anti personell mines have close to zero metallic content (at least ferrous metals) to make the use of mine detectors unfeasible. Instead the designers realized that changing the trigger so that the mine detonates when you step off it instead of step onto it, bringing the body over the mine instead of behind it at the time of detonation, and reducing the amount (and 'speed' i.e. brisance) of the explosives, enables the 'use' of the small bones of the middle foot as shrappnell (mainly impacting the groin area).
This is as mentioned a tripple win. It makes the use of metal detectors as mine detectors unfeasible, it reduces the weight of the mine so more can be carried, and it reduces the cost of the mine since there is less (relatively expensive) material and less explosives. What's left is a plastic container, explosives and a (very simple) trigger with initiator. Dirt cheap, light weight, lasts for ages and virtually impossible to detect.
Well, that's rather my point. It's anywhere from impossible to damn difficult to 'prove' that the situation was dangerous. There simply wasn't any experience/work done to investigate this scenario in the design phase, since it out of design specification, and simply wasn't supposed to happen.
If someone says 'prove it' and you can't (and you know you'll be labeled a 'PITA' if you voice concerns anyway) then what are you supposed to do? Engineers like to deal with the known, yelling that the thing isn't 'safe' on the top of your lungs when you just don't know, and management says; 'Well you know, if you had anything concrete we'd listen to you.' is a bit much.
Hence, I still would blame management for fostering a culture where 'we're right, and if you ever want to question that then you'd better have all your t:s crossed and i:s dotted. That's not the right kind of organisation to handle space flight. Just like the army, you don't blame the ineptitude on your troops on them, since you were also responsible for training them.
And it's not like the engineers didn't mention their misgivings time and time again, but they didn't have the facts (and also failed to present the ones they had in an accessible manner, see Tuftes arguments on the visualisation of the O-ring damage in for example the introduction to, "Readings in information visualisation", MacKinley et al.) They just didn't know. They told management so, and management responded 'well come back when you know it's unsafe.' It's a bit much to ask the engineers to commit ritual suicide on such weak evidence. It's not too much to ask of management to say: 'Gee, you say the whole thing was built on the assumption that hot gasses where never supposed to get near the O-ring seal, you're right let's take a step back and review the new situation untill we are certain there's no additional risk from running out of design specs in such an obviously critical part of the flight.'
The US does, however, clean up areas that it's mined once it's done with them. I doubt it's a perfect job, but it's considerably better than the vast number of military forces that use mines and don't clean them up (which is where the issue has come from).
The problem is one of economy. It cost about $10 to $20 to lay an anti personell mine, and can cost as much as $800 to clear one. Hence any nation the US included can easily put itself into a position where they simply cannot afford to 'clean up after themselves'. That the US want to keep the mines on the South Korean border has little to do with staving off an invasion (anti vehicle mines have a role to play, but anti personell doesn't really give pause). Granted, if you've saved the mine charts and the mines haven't shifted too much you're not going to spend the full $800 but it's still considerably more expensive to clear than to lay.
Troops stop invasions, mines give troops time to react but will not hinder a determined foe much, especially in a place like the Korean border where the mine fields are clearly marked. Command detonated AP mines, that aren't part of the treaty, could serve just as well with a small increase in the manpower needed to operate them.
AP mines don't kill infantry. Artillery kills infantry. Mines will at their very best make the aiming a little easier. Remote sensing equipment could work just as well. And the detection capabilities that mines provide can easily be replaced e.g. with buried fiber optics.
P.S. And tell the Vietnameese and the Cambodians that the US always cleans up areas that it's mined once it's done with them and you'll get a right bloody laugh in you face. Unless they have first hand experience in which case they may bloody your nose instead.
How is knowing you're being invaded a deterrent to invasion?
"Hey! You in the tank! I see you!"
Because it's "Hey, you in the tank, did you see my anti tank missle before it hit you?!"
Troops stop invasions. Mines only slow down or at best channel the attacker. If the military could somehow magically learn where were the enemy troops were all the time and what they were doing, they'd gladly give up all the anti personell mines in the world.
Anti vehicle mines actually have more of a tactical value, but they're not that much of a problem after the war, being bigger and not that easy to set of, so no-one's asking for them to be handed over yet.
The largest mines are probably anti-personnel claymore mines. They are something like 25cm high and 50cm wide and produce a huge (50-100m) conical storm of shrapnel. They are however usually attached to a tree or similar structure so that the effect acts horizontally taking out a lot of infantry and perhaps unarmored vehicles. The plants are of no use when looking for these.
Depending on exactly how you use them, they're not that much of a problem either. If retained as a command detonated device only (as we have done with the Swedish inventory), i.e. with no 'boby trap' trigger, they pose little threat to post war inhabitants, less so than ordinary duds, such as artillery and mortar shells and the like.
These are not dug into the ground so I doubt the plant will be of much use in finding them. Now someone is thinking that such small mines that are above the ground shouldn't be a problem, but account for the fact that the a plane can drop several thousands in one run. Combine that with toxic gas and/or napalm and it's a really big problem.
If they stayed above ground for time eternal, they wouldn't be a problem for generations to come that's for certain. However, they don't. Rain, erosion, accumulation of plant debri etc all work together to bury them over time, and then you have a problem. That there are many of them, and napalm etc is mostly just a tactical problem. And we're not trying to ban mines to save soldiers.
The kinds of mines that the plants are probably used to find are anti-personnel mines dug into the ground.
That and other types of unexploded ammunition. Cluster bomblets in particular tend to have a very high failure rate (figures of 10%-20% are sometimes heard) and being armed and buried on impact they get to be a real problem.
In my experience engineers are the least likely people to overlook apparently inconsequential problems when they have hints or evidence of problems ; but if you keep the absurd requirement of 100% failsafe equipement, they're obviously going to never approve anything, expecially when they know blame will be shifted on them. Add the second stress factor of working under a budget, knowing that you'll not be listened to if the problem you spotted is going to cause a budget rescheduling.
And that's not even the issue here. No NASA engineer belives 'safe.' They do belive risk analysis, fault tree analysis etc ad nauseaum. Flying into space is not about avoiding risk, it's about taking a calculated risk.
Problem then becomes when engineers find a new variable, one previously never thougt of, that threatens to upset the old risk calculations. A 'safety' concious culture, then says "Hmm, that's not good, lets do the experiments and calculations and see where that leads us." Instead NASA management has an affinity for saying "Well, until you can prove that these here old numbers aren't correct, and I mean prove, I'm not going to listen to that and just proceed as if the situation hasn't changed.
And partially as a result the shuttle is a complete failure from a risk management perspective. It was designed (as a requirement) to be much safer than previous generations, and it hasn't come close to that design goal.
The problem isn't that it's risky to fly into space using the shuttle. The problem is that NASA management belives against their better judgement that it's much safer that it actually is.
The also fit the pattern of murder by engineer. When asked by management to make their case as regards to the O-rings, they were unable to come up with a coherent case. When asked by management to make their case for the imaging, they simply walked away without even attempting to do so.
Well, that's a little harsh IMHO. In the case of the O-rings, they showed signs of erosion to a maximum of about a third of their radius. However, as the engineers pointed out the seals were designed to not have hot enough gasses near the O-rings for them to erode at all. That should be a big red flag right there. We are operating out of design spec.
As part of the investigation it was concluded that failure would not come unless the whole radius of the O-ring was burned through, and management then chose to read this as having a safety factor of three against failure. Then demanding the engineers to prove otherwise.
That's not how to run a safety concious organisation (in fact Feynman likend it to a child having run accross a busy road saying; "Look ma, nothing happened, it's not dangerous"). In fact it should have been handled the other way around. Back to the drawing board and finding the root cause analysis of why the O-rings showed signs of erosion. Not concluding out of the blue that they had a 'safety' factor of three built into them.
It's interesting that poor presentation played a part in the O-ring failure as well as the Columbia accident, but it's really an aside. You should have to prove that the operation is within design parameters (not 'safe' as flying into space required quite a considerable measure of calculated risk), not having to counter argue that you're operating out of them.
I have not conducted much investigation here but a cursory glance would indicate that the kernel calls made by X are restricted to some bulk AGP/PCI data transfers whereby one context switch occurs per large block of data and on the other hand the APIs exposed to the applications deal with much smaller and frequently used objects like window elements and widgets. So it would probably be a net loss of performance.
Like the IO optimisations that take place in a (decent) C library, by blocking transfers. Of course, never thought about that but it makes sense. If you have the time to investigate further I think many people would be interested in the results.
As an aside, it's all this talk about NT being a 'microkernel' that irks me. With the graphics subsystem in the kernel proper it's even less of a 'microkernel' than a traditional UNIX monolith. After all many critical system functions do take place in user land daemons in a traditional UNIX kernel. I can't really see a difference there. Especially if one compares with a proper microkernel, like QNX, VSTA, L3 or the like.
As a side note, the attitude in your response indicates
that you are not willing to compromise every aspect
of computer operation for ease of use and responsiveness.
And when that 'aspect' is stability, as in 'no BSOD's then he's completely correct.
Having windows dragging a few percent faster is not an improvement if it means that you're going to lose even an hours work once. Then all that 'saved' time is more than used up, and then some.
I therefore hope that you do not get involved in
developing desktop anything.
On the contrary, we need more people like him developing for the desktop. People who for example understand that developing a GUI application in 'C' just so that it can crash from a random null pointer reference as soon as the user turns his back, on the theory that 'code that draws must be fast' is not the way to do it.
Repeat after me; an application (or whole computer if you're under windows) that crashes is supremely unusable. Always was, always will be. First the functional requirements must be met, and then the non-functional. If you can only have one, then go for the functional. Fast and wrong is never right.
Can't be the only reason though, since Gripen was one of their alternatives.
No, it's probably a combination of. 1) They were allowed to buy american for the first time, having disolved their forced defensive pact with the Russians. 2) The US heavily subsidises the sale of military aircraft (in part as a way to subsidise their civilian aircraft industry, which they couldn't do otherwise), so it was a better deal than Gripen, which has to sell on merit alone (well mostly anyway).
It's hard to deny that Germany would not have been stopped in Europe without the US entry into that theater.
I'd say it is hard actually. The main battle was on the eastern front, a battle that dwarfed all others. For the whole duration of the war some 80% of the German army's resources (esp. personell) was spent on the eastern front.
So to parapharse: "If it weren't for the Americans we'd all be speaking Russian" is not an unlikely scenario.
Not that I'd have very much cared for Stalinist rule instead of a free Western Europe, but I think it could be argued that the Nazis were doomed when Hitler broke the Molotov-Ribbentrop act.
come on slashdot, infoworld, this isn't news, this is a PRESS RELEASE
Well, it probably originated from a press release. But then the journalist actually went to the trouble of confirming the claims with independed experts, all of which seemed to agree that: yes there is a problem and yes Valence does seem to have a safer solution to it.
That's not actually that shoddy journalism as such go these days. The reporter actually put the facts up front and went to some trouble to corroborate them, instead of rewriting the press release into something resembling a researched piece; all too common these days.
Just because we're fed up with exagerated claims and 'marketing' contorting everything into unrecognisability, that doesn't mean that once in a while someone actually does invent a better mousetrap. Unfortunately they have to use the same channels as everybody else to spread the word. Not all 'innovation' is of the Microsoft kind.
Now, whether the effect leading to exploding batteries is significant enough to actually worry about is another matter. So even if Valence have a solution to the problem that solution may still be irrelevant from a bigger perspective, but that's a slightly different question.
Wrong. The Norwegian legal system normally has a jury of ten persons, in criminal cases, tried at the court level Jon Johansen was just aquitted in. I don't know why this wasn't the case here.
Hmm, that's not what my Norwegian workmates tell me. They tell me it's closer to the Swedish system, i.e. with 'jurors' that sit for a term, i.e. aren't called in at random for each trial. But I'm not Norwegian, so I really can't say.
As far as I know the supreme court can't overturn the aquittal.
That sounds odd. I was under the impression that the Norwegian system was like the Swedish in that the highest court was the only one that could set a precedent. That's not going to work if they cannot convict. That they may never do so is another kettle of fish.
Since you asked. This is in no way exhustive. You legal system is more complex than you tax codes, but here is the basics:
Oh, it was only the one detail really. I already know quite a bit about the US legal system, if for nothing else courtesy of Hollywood. But there was a tidbit or two I didn't know about.
And interestingly the Hollywood factor is quite strong. Something like 60%-70% of all Swedes belive we have jury trials in Sweden. (We don't, except in cases where the freedom of the press is involved). Swedish police introduced badges only a few years back since people were always surprised they didn't have any (just an ordinary ID). etc.
Since you brought up the civil side. The differences there are that loser pays here, i.e. if you lose you pay the cost of the legal representation of the winning side (within reason). And if you are awarded less than half of what you originally claimed you're considered the loser. So that tends to keeps exhorbitant claims in check. (And no punitive damages, if your behaviour is criminal it belongs in criminal court).
Also there are a number of special courts, i.e. where the workplace, where the renting of an appartment, or the appeal of a government decision is concerned. Bringing a case before any of these courts is "risk free" for the defendant, i.e. you've already 'lost' something otherwise you wouldn't be there. And there are special public servants with prosecutory power (ombudsman) that can bring suit on your behalf if you cannot do that on your own (or belong to an association such as a union or the renters association that will do so for you). Also, product liability is handled by a branch of government, the 'consumers dept' ("Consumer report" in Sweden is a government publication).
There's basically a special court for every situation in which a 'normal' citizen would go to court, so the overwhelming majority of people will never see the inside of a criminal/civil court, or even know anybody that has.
We don't have juries on the theory that they will tend to lead to "show trials", where oratory performance will become too much of a deciding factor, a common complaint when freedom of the press cases are concerned.
Slashdot Mirror
Well, I have an older 500 MHz Athlon at home that hangs the network cards when I try to run 'fvcool' (Linux), but my (now quite old) 2200+ here works beautifully. Stable as a rock, and no performance degradation either. Before I ran 'fvcool' it would remain constantly at 72 dec C, but without it it's down to an easier 38 dec C. (Don't ask, I was limited to one builder and they weren't AMD savy.)
So, it works for some people at least.
Having a bit of a sense-of-humour failure are we?
Yes well, I'm not saying that a big honking rocket isn't the way to go. Combined with a wingless capsule type craft to put people in, it'll be both safer and more cost efficient. However, once you're done 'rebuilding' the Saturn V with modern technology, you'll in effect have built a new rocket. It's no longer the Saturn V. And the Russians have some nifty 'hot pump' engines you'd like to put on it, though the design for those is quite old. Thinking that you're only using tested parts and that'll speed up reconstruction won't work as the first Ariane 5 failure demonstrated. It's the knowledge you can reuse, not the design.
And having the blue prints isn't much help in that respect. Lots of the engineering knowledge that went into the Saturn V have been lost. You have the design now, but much of the reasoning that went into that design is gone. And you really need that when you're going to modify/update the design. The Soviets made a few spectacular errors in this regard, stealing western aircraft designs (or downed planes) and building something that didn't work at all as advertised (Though it should be noted that the TU-144 Concordski crash was really due to other reasons).
So by all means, go for the big honking rocket, but name it the Saturn VI to help remembering what it is that you're really building.
And many others have heard that also, so many in fact that it qualifies as an urband legend.
It's not true however. They're still there. The problem is that you couldn't get the parts (sixties vintage) today, and the launch pads have been rebuilt. We've also learnt a thing or two since the sixties. Once you've resourced the parts and rebuilt the launch pads you might as well have started from scratch and gotten a better vehicle for it.
Not really true these days. Some architectures do have an explicit 'cache preload' instruction, such as the SPARC V9 and the ARM9E to mention two. These allow the programmer to preload a D-cache line before it is needed.
As the speed of the CPU has increased much faster than the speed of main memory (and hence increasing the relative cost of a cache mis), compiler based techniques to emit cache preload instructions in advance, before the data is actually needed, has been the subject of some research in the past 7-8 years. The main reason to do it is software, instead of hardware, is that the compiler have a greater knowledge of the layout of the entire task, as it can 'look ahead' in the source code. The main disadvantage is that any static analysis, of course won't have access to dynamic (run time) data about the program as it is running.
If you wish to go further, you could do worse than to start with my former colleague Magnus Karlsson's PhD thesis on the subject:
And of course as always, Google and citeseer is your friend.
Well, actually, it doesn't these days... :-) Trust the compiler instead :-) (Yeah, right).
Ah, we're into nit picking are we? In that case, you have to qualify that you're thinking of a point source in a completely echo free environment. Which won't typically be the case in a computer case.
No, seriously, you make a good point. Of course if you started your fan in Mongolia it could be pretty powerful and still be considered silent here (for most definitions of 'here' not actually close to Mongolia). Hence, how the measurement is made really is an important factor. I'd take a 21dB fan at 10cm over a 14dB fan at 1.5 meter any day.
Oh, I wouldn't be too hard on the guy. You did come across quite a bit like that to me to.
Well, you may want to think about your wording in the future then, that's not at all how I got the tone of your post. Just a friendly advice, take it or leave it.
Myself, I'd argue the opposite angle. Even my old computer is plenty fast/big for anything I need/want to do with it, even the (few) games I play. If it hadn't been quiet (I built an MDF sound proof hood for it) I couldn't have it on all the time conveniently situated in my living room, which makes TV-out a delight for example.
I mean, what use is an epsilon faster computer, costing you hundreds of dollars more for very little speed up, if it's so damn noisy that you think twice before turning it on (not to mention the complaints from the rest of the family, especially later in the evening), and turning it off as fast as you can? Spend a reasonable amount of dosh on quiet I say, and choose a slower processor, slightly smaller disk etc. You wont regret it.
Cranking up the TV/stereo to drown out the noise of the computer is not an option for most of us.
Actually, he's saying that the guy that works in the whitehouse mail room is smarter than a mail server. Whether that makes him smarter than his boss, I think is a question of some debate, though I know that I'm certainly leaning towards that possiblitity.
It should be noted that these are by now the old style anti personell mine, not in wide spread use since the early/mid eighties. The problem with providing a casing (or simply a plate) to provide shrappnel is that it increases the weight of the mine, and also the metallic content which makes clearing easier (via the use of mine detectors).
Modern anti personell mines have close to zero metallic content (at least ferrous metals) to make the use of mine detectors unfeasible. Instead the designers realized that changing the trigger so that the mine detonates when you step off it instead of step onto it, bringing the body over the mine instead of behind it at the time of detonation, and reducing the amount (and 'speed' i.e. brisance) of the explosives, enables the 'use' of the small bones of the middle foot as shrappnell (mainly impacting the groin area).
This is as mentioned a tripple win. It makes the use of metal detectors as mine detectors unfeasible, it reduces the weight of the mine so more can be carried, and it reduces the cost of the mine since there is less (relatively expensive) material and less explosives. What's left is a plastic container, explosives and a (very simple) trigger with initiator. Dirt cheap, light weight, lasts for ages and virtually impossible to detect.
Well, that's rather my point. It's anywhere from impossible to damn difficult to 'prove' that the situation was dangerous. There simply wasn't any experience/work done to investigate this scenario in the design phase, since it out of design specification, and simply wasn't supposed to happen.
If someone says 'prove it' and you can't (and you know you'll be labeled a 'PITA' if you voice concerns anyway) then what are you supposed to do? Engineers like to deal with the known, yelling that the thing isn't 'safe' on the top of your lungs when you just don't know, and management says; 'Well you know, if you had anything concrete we'd listen to you.' is a bit much.
Hence, I still would blame management for fostering a culture where 'we're right, and if you ever want to question that then you'd better have all your t:s crossed and i:s dotted. That's not the right kind of organisation to handle space flight. Just like the army, you don't blame the ineptitude on your troops on them, since you were also responsible for training them.
And it's not like the engineers didn't mention their misgivings time and time again, but they didn't have the facts (and also failed to present the ones they had in an accessible manner, see Tuftes arguments on the visualisation of the O-ring damage in for example the introduction to, "Readings in information visualisation", MacKinley et al.) They just didn't know. They told management so, and management responded 'well come back when you know it's unsafe.' It's a bit much to ask the engineers to commit ritual suicide on such weak evidence. It's not too much to ask of management to say: 'Gee, you say the whole thing was built on the assumption that hot gasses where never supposed to get near the O-ring seal, you're right let's take a step back and review the new situation untill we are certain there's no additional risk from running out of design specs in such an obviously critical part of the flight.'
The problem is one of economy. It cost about $10 to $20 to lay an anti personell mine, and can cost as much as $800 to clear one. Hence any nation the US included can easily put itself into a position where they simply cannot afford to 'clean up after themselves'. That the US want to keep the mines on the South Korean border has little to do with staving off an invasion (anti vehicle mines have a role to play, but anti personell doesn't really give pause). Granted, if you've saved the mine charts and the mines haven't shifted too much you're not going to spend the full $800 but it's still considerably more expensive to clear than to lay.
Troops stop invasions, mines give troops time to react but will not hinder a determined foe much, especially in a place like the Korean border where the mine fields are clearly marked. Command detonated AP mines, that aren't part of the treaty, could serve just as well with a small increase in the manpower needed to operate them.
AP mines don't kill infantry. Artillery kills infantry. Mines will at their very best make the aiming a little easier. Remote sensing equipment could work just as well. And the detection capabilities that mines provide can easily be replaced e.g. with buried fiber optics.
P.S. And tell the Vietnameese and the Cambodians that the US always cleans up areas that it's mined once it's done with them and you'll get a right bloody laugh in you face. Unless they have first hand experience in which case they may bloody your nose instead.
Because it's "Hey, you in the tank, did you see my anti tank missle before it hit you?!"
Troops stop invasions. Mines only slow down or at best channel the attacker. If the military could somehow magically learn where were the enemy troops were all the time and what they were doing, they'd gladly give up all the anti personell mines in the world.
Anti vehicle mines actually have more of a tactical value, but they're not that much of a problem after the war, being bigger and not that easy to set of, so no-one's asking for them to be handed over yet.
Depending on exactly how you use them, they're not that much of a problem either. If retained as a command detonated device only (as we have done with the Swedish inventory), i.e. with no 'boby trap' trigger, they pose little threat to post war inhabitants, less so than ordinary duds, such as artillery and mortar shells and the like.
If they stayed above ground for time eternal, they wouldn't be a problem for generations to come that's for certain. However, they don't. Rain, erosion, accumulation of plant debri etc all work together to bury them over time, and then you have a problem. That there are many of them, and napalm etc is mostly just a tactical problem. And we're not trying to ban mines to save soldiers.
That and other types of unexploded ammunition. Cluster bomblets in particular tend to have a very high failure rate (figures of 10%-20% are sometimes heard) and being armed and buried on impact they get to be a real problem.
And that's not even the issue here. No NASA engineer belives 'safe.' They do belive risk analysis, fault tree analysis etc ad nauseaum. Flying into space is not about avoiding risk, it's about taking a calculated risk.
Problem then becomes when engineers find a new variable, one previously never thougt of, that threatens to upset the old risk calculations. A 'safety' concious culture, then says "Hmm, that's not good, lets do the experiments and calculations and see where that leads us." Instead NASA management has an affinity for saying "Well, until you can prove that these here old numbers aren't correct, and I mean prove, I'm not going to listen to that and just proceed as if the situation hasn't changed.
And partially as a result the shuttle is a complete failure from a risk management perspective. It was designed (as a requirement) to be much safer than previous generations, and it hasn't come close to that design goal.
The problem isn't that it's risky to fly into space using the shuttle. The problem is that NASA management belives against their better judgement that it's much safer that it actually is.
Don't think 'safe', think 'calculated risk'.
Well, that's a little harsh IMHO. In the case of the O-rings, they showed signs of erosion to a maximum of about a third of their radius. However, as the engineers pointed out the seals were designed to not have hot enough gasses near the O-rings for them to erode at all. That should be a big red flag right there. We are operating out of design spec.
As part of the investigation it was concluded that failure would not come unless the whole radius of the O-ring was burned through, and management then chose to read this as having a safety factor of three against failure. Then demanding the engineers to prove otherwise.
That's not how to run a safety concious organisation (in fact Feynman likend it to a child having run accross a busy road saying; "Look ma, nothing happened, it's not dangerous"). In fact it should have been handled the other way around. Back to the drawing board and finding the root cause analysis of why the O-rings showed signs of erosion. Not concluding out of the blue that they had a 'safety' factor of three built into them.
It's interesting that poor presentation played a part in the O-ring failure as well as the Columbia accident, but it's really an aside. You should have to prove that the operation is within design parameters (not 'safe' as flying into space required quite a considerable measure of calculated risk), not having to counter argue that you're operating out of them.
Like the IO optimisations that take place in a (decent) C library, by blocking transfers. Of course, never thought about that but it makes sense. If you have the time to investigate further I think many people would be interested in the results.
As an aside, it's all this talk about NT being a 'microkernel' that irks me. With the graphics subsystem in the kernel proper it's even less of a 'microkernel' than a traditional UNIX monolith. After all many critical system functions do take place in user land daemons in a traditional UNIX kernel. I can't really see a difference there. Especially if one compares with a proper microkernel, like QNX, VSTA, L3 or the like.
And when that 'aspect' is stability, as in 'no BSOD's then he's completely correct.
Having windows dragging a few percent faster is not an improvement if it means that you're going to lose even an hours work once. Then all that 'saved' time is more than used up, and then some.
On the contrary, we need more people like him developing for the desktop. People who for example understand that developing a GUI application in 'C' just so that it can crash from a random null pointer reference as soon as the user turns his back, on the theory that 'code that draws must be fast' is not the way to do it.
Repeat after me; an application (or whole computer if you're under windows) that crashes is supremely unusable. Always was, always will be. First the functional requirements must be met, and then the non-functional. If you can only have one, then go for the functional. Fast and wrong is never right.
Dead? They're pretty important, though granted they're not acetylcholine.
No, it's probably a combination of. 1) They were allowed to buy american for the first time, having disolved their forced defensive pact with the Russians. 2) The US heavily subsidises the sale of military aircraft (in part as a way to subsidise their civilian aircraft industry, which they couldn't do otherwise), so it was a better deal than Gripen, which has to sell on merit alone (well mostly anyway).
I'd say it is hard actually. The main battle was on the eastern front, a battle that dwarfed all others. For the whole duration of the war some 80% of the German army's resources (esp. personell) was spent on the eastern front.
So to parapharse: "If it weren't for the Americans we'd all be speaking Russian" is not an unlikely scenario.
Not that I'd have very much cared for Stalinist rule instead of a free Western Europe, but I think it could be argued that the Nazis were doomed when Hitler broke the Molotov-Ribbentrop act.
I distinctly remember wiring up the signal ground as well, for a total of three pins. Was that one too many?
Well, it probably originated from a press release. But then the journalist actually went to the trouble of confirming the claims with independed experts, all of which seemed to agree that: yes there is a problem and yes Valence does seem to have a safer solution to it.
That's not actually that shoddy journalism as such go these days. The reporter actually put the facts up front and went to some trouble to corroborate them, instead of rewriting the press release into something resembling a researched piece; all too common these days.
Just because we're fed up with exagerated claims and 'marketing' contorting everything into unrecognisability, that doesn't mean that once in a while someone actually does invent a better mousetrap. Unfortunately they have to use the same channels as everybody else to spread the word. Not all 'innovation' is of the Microsoft kind.
Now, whether the effect leading to exploding batteries is significant enough to actually worry about is another matter. So even if Valence have a solution to the problem that solution may still be irrelevant from a bigger perspective, but that's a slightly different question.
Hmm, that's not what my Norwegian workmates tell me. They tell me it's closer to the Swedish system, i.e. with 'jurors' that sit for a term, i.e. aren't called in at random for each trial. But I'm not Norwegian, so I really can't say.
That sounds odd. I was under the impression that the Norwegian system was like the Swedish in that the highest court was the only one that could set a precedent. That's not going to work if they cannot convict. That they may never do so is another kettle of fish.
Oh, it was only the one detail really. I already know quite a bit about the US legal system, if for nothing else courtesy of Hollywood. But there was a tidbit or two I didn't know about.
And interestingly the Hollywood factor is quite strong. Something like 60%-70% of all Swedes belive we have jury trials in Sweden. (We don't, except in cases where the freedom of the press is involved). Swedish police introduced badges only a few years back since people were always surprised they didn't have any (just an ordinary ID). etc.
Since you brought up the civil side. The differences there are that loser pays here, i.e. if you lose you pay the cost of the legal representation of the winning side (within reason). And if you are awarded less than half of what you originally claimed you're considered the loser. So that tends to keeps exhorbitant claims in check. (And no punitive damages, if your behaviour is criminal it belongs in criminal court).
Also there are a number of special courts, i.e. where the workplace, where the renting of an appartment, or the appeal of a government decision is concerned. Bringing a case before any of these courts is "risk free" for the defendant, i.e. you've already 'lost' something otherwise you wouldn't be there. And there are special public servants with prosecutory power (ombudsman) that can bring suit on your behalf if you cannot do that on your own (or belong to an association such as a union or the renters association that will do so for you). Also, product liability is handled by a branch of government, the 'consumers dept' ("Consumer report" in Sweden is a government publication).
There's basically a special court for every situation in which a 'normal' citizen would go to court, so the overwhelming majority of people will never see the inside of a criminal/civil court, or even know anybody that has.
We don't have juries on the theory that they will tend to lead to "show trials", where oratory performance will become too much of a deciding factor, a common complaint when freedom of the press cases are concerned.
Pining for the fjords are we? :-)