The current Taiwanese president's (who was popularly elected) stance on Taiwan's status is that he wouldn't push for independence, instead preferring to tread the fine line of the current status quo. The Taiwanese aren't actually loathing for (re)union with China, but they aren't unequivocally for independence either.
That is called diplomacy. The issue of Taiwan is a source of nationalistic pride in China and the issue is used like September 11th is in the USA. When you're a country not much biger than Luxemborg on the border of a nation with one of the largest militaries in the world who is looking for an excuse to invade you choose your words carefully.
Somehow I doubt that most Taiwan citizens would prefer Chinese citizenship. It would be a complete step backwards for them. As it is they have most of the conveniences of Chinese citizenship (due to Chinese policy) and none of the downsides. Kind of like Puerto Rico in the US...
And I didn't read People's Daily about the Tienanmen Square event. Did you use CNN/Foxnews as your sources?
Uh, I never said anything about the People's Daily coverage of Tienamen Square. They were used in a different illustration. Even so, I'm sure their coverage of the event was VERY tame.
Have you watched this before? http://www.tsquare.tv/ (note: it isn't made by the Chinese)
Actually, I haven't - I may take the time to do so (thanks for the link). I noted that they did have a link to Frontline on that site and I did see their episode dedicated to the aftermath of the incident.
I actually watch fairly little TV news programming - I tend to get most of my news on the web from RSS feeds (a fair variety of them). I doubt you'd ever catch me watching Fox News. If anything I'd be more likely to watch BBC's World News America or something along those lines for a broader perspective (I DVR it even if I don't often watch it).
I do strongly disagree with your assessment that Democracy simply wouldn't work in China. I don't disagree that it would be disruptive at first. There is a lot of history that the West went through prior to Democracy taking off and in many places that history simply doesn't exist. That doesn't mean that Democracy isn't ultimately what is needed in China (and everywhere else for that matter). Democracy isn't really a great form of government, but it just happens to be better than every other form out there (save appointing me dictator of the world). One thing that does tend to improve it seems to be the proportional system of representation used in most European states - not having two entrenched parties seems to help reduce the vote-buying phenomenon which is a big problem in the USA. Sure, Americans are a lot wealthier than the average Chinese, but that just means the politicians have to tax people and buy their votes using their own money...
At work I've tended to percieve that Indians are treated essentially as slaves. They've paid next to nothing. Many actually end up walking to work from nearby slum-like apartments on streets without sidewalks (in what is otherwise a very nice middle-class area). Work is dumped on them and they're expected to be incompetant. They're watched closely. The only thing that is missing is the whip. It has gotten to the point that when somebody sees an Indian at work the assumption is that they're essentially part of an underclass.
A few years ago this attitude didn't exist at all. Indian coworkers weren't viewed any different from caucasians or East-asians or anybody else. It seems like we're inventing slavery all over again...:(
Re:Too afraid to see who they are
on
China Blocks iTunes
·
· Score: 1, Informative
Seriously, if anybody indeed has constructive ideas on how to substantially improve the situation in China, I'm sure a lot of people (including Chinese) would lend you their ears.
Uh, how about hold elections? The kind where anybody can run? And how about having a free press where one isn't punished for expressing one's opinions? That would be a good start and in the long run would substantially improve the situation.
In fact, westerners who are bashing China are usually completely out of touch with China.
Yup - you've hit that nail on the head. Why, most westerners actually think that the citizens of Taiwan actually want to be a separate country! What a crazy idea - fortunately, the People's Daily dubunks that rubbish and points out how they've been oppressed for decades and that the best thing the world can do is get out of the way and let mainland China liberate them.
The next thing those crazy westerners will be trying to tell us is that the Chinese governement actually masaccred students in Tienneman Square! What a load of propaganda!
Testing is actually a good part of quality development. However, the best testing is the kind of testing performed directly by the developers on small chunks of code (unit testing).
Too many companies view testing as something you do once the whole project is "done". The problem is that when you find problems at that point you end up just saying "oh well, we'll just live with it" and deploy it anyway. If you're going to do that, why bother testing at all? You need to spend your money on preventing problems, not discovering them.
Encouraging web browsers to ignore security irregularities and allow users to access sites that handle private information *without* bringing it to the user's attention is just irresponsible.
But that is exactly what firefox is doing. If a user connects to a site without using ssl they don't get any warnings and they can submit private information at will.
If a user connects to a site that uses a self-signed certificate, they are at less risk of interception than if they don't use any encryption at all, and yet the connection is treated as being "riskier".
If firefox were to disable all form submission to any site at all that doesn't use https with a trusted certificate at least I could see that they've being consistent...
Is unencrypted http useless? It offers communications that are neither secure against evesdropping nor prevention of MITM attacks.
At least https with a self-signed certificate offers one of these two.
If firefox wants to scream when a user could be subject to a man-in-the-middle attack that seems fine to me. If that is the design goal, however, the browser should yell at me every time I visit slashdot - since the connection isn't protected against anything.
All an attacker needs to do to bypass firefox's protection is to do a MITM attack but use a non-encrypted connection back to the browser and SSL to the server. The browser won't complain a bit and the server won't know that anything is wrong. The user will think that everything is fine since he is now conditioned to the browser yelling at him anytime anything could go wrong.
As for #6, that doesn't even make sense. If the people who DON'T lead meetings are not going to survive, then who's going to attend these meetings? Meetings generally don't work well when everyone is trying to do the leading. Same can be said for a company's general operations. Remember the old adage about "too many chefs in the kitchen". US companies are not going to survive if they think they're going to make a lot of money by just sitting around, leading meetings, and having all the real work done in other countries by contractors
The parent post was not telling you want is needed for US companies to survive. He was telling you what you need to do to keep an IT job in the current climate. Current US coprate hiring practices do not have anything to do with acquiring the skills they need to survive. Eventually Darwin will catch up and hiring practices will change. Unfortuantely the typical corporation can last longer without income than the typical employee can.
When you're not expecting heavy action you could also use a multi-tiered staffing approach. You'd have a team of basically-trained staff that pilots the drones to and from points of interest, or monitors them during loitoring. You'd have a surveilance staff monitoring feeds from all the drones. When things heat you you'd turn over individual drones to dedicated strike crews - maybe with some crews having an A-A or A-G focus. You might also have the "special ops" crew that is every bit as trained as an F22 pilot is today for really critical missions.
There is no reason you need one pilot from liftoff to landing on one of these drones.
I very much realize the mistakes of Vietnam, but I wonder if things have changed since then. Back in Vietnam it wouldn't unheard of to discharge an entire load of missiles to only have one or two track. Today the practice of firing off more than one missile is still common, but that is mainly because in a typical mission you'd expect to only get one opportunity to shoot and even if the missile is 90% accurate two are better than one. Modern communications also makes it easier to identify foes beyond visual range and engage them at a distance.
And if you have a ton of unmanned aircraft all you need to do is stage them at various distances from the line - even if the enemy manages to surprise one or two of your picket aircraft the bulk of your forces can easily engage BVR. The loss of two aircraft is much more acceptable when they don't have pilots onboard. Indeed - if the other aircraft had clear shots they'd just shoot down the enemy formation including the two friendlys if it made sense to do so.
Perhaps the dogfight's days aren't completely over, but it seems that they're largely behind us. It would really take a very serious war with more of an equal to bring it back, and even then missiles are getting closer to the point where they are simply unevadable once you are within range.
It wouldn't surprise me if standard operational procedure after putting out any fire doesn't include a complete walk-through of the building to ensure that nothing is still smouldering. After all, somebody will get sued if the fire dept clears the building and then it catches on fire again after everybody goes back in.
Black body radiation. Gotta love it - no getting around it. All objects emit EM radiation - the frequency distribution and intensity is related to temperature.
If you don't want to put IR then you need to keep the temperature of the surface of this thing cool. The person inside is generating heat, and that heat has to go somewhere (as another poster has pointed out, humans don't work well if they can't perspire away their heat). I guess you can carry a big tank of LN2 with you. If you want you could radiate the heat directionally and only be visible in IR from that side, but that is obviously less valuable than just being invisible.
I'm not sure how well IR imaging works in daylight, however. If you're outdoors on a desert I suspect the IR blackbody radiation will be lost in the wash of heat coming off the ground.
Probably the bigger problem is that in order to legally advertise that a drug can be used to treat a disease you need to prove via clinical trials that the drug works. Those trials cost a few tens of millions of dollars each at a minimum. A full battery of trials costs into the 9-figures.
So, the company didn't want to pay for the R&D. Sure, you might "know" that it will work, but the FDA and its other first-world peers won't accept that.
The rules exist for a reason, but they really do make it hard to make drugs for conditions that don't have a significant market. Regulators should really think about this and come up with a better framework for regulating these kinds of drugs. When you require 9-figure development costs you aren't promoting patient safety in these cases - you're condemning patients to non-treatment.
Do some googling on off-label drug use to find out more about these sorts of problems. Once a drug is on the market doctors often bend the rules for their patients.
The specific issue you raise might very well also be a factor - particularly with biotech companies that tend to have lots of molecules but without the ability to develop them.
If I had a dollar for every university researcher who "cured cancer" I'd be a very rich man. Ditto for HIV.
For the most part the only really effective drug patents cover one specific molecule, with obvious non-functional additions/changes to it (you can't just stick a methyl on some huge molecule and call it an innovation).
If a drug company already has this product patented, then why haven't they released a cure?
Researchers tend to claim "success" when they come up with some molecule that has some inhibitory reaction in some in-vitro test. That is often a great lead, but only a very small fraction of these molecules turn out to be useful. Often they start a new line of thinking/questioning that does lead to the correct answer. It is a bit much to say that "most of drug R&D is government funded." That's like saying that because a government lab invented the first polymer that they should be credited with the commercialization of teflon. No doubt one wouldn't happen without the other, but there was certainly a lot of time, effort, and money spent in-between.
Overly-broad patents are clearly harmful. However, if a company goes to the effort to fund clinical trials on a molecule they should stand to benefit from this - beyond having a few months on the market before the first competitor scales up their own production of the same thing.
Massive government funding is a fair alternative to patented private drug development. I'm all for exploring that further. However, volunteer contribution is a bit unrealistic on these kinds of projects - they involve hundreds of clinics and thousands of doctors and scientists, with a truck's worth of paperwork to support their safety testing. Linux on the desktop is a trivial undertaking in comparision and "bounties" haven't quite made that happen yet...
I've seen this kind of issue in other quality-sensitive organizations.
The problem is one of risk tolerance. To truly engineer to all contingincies is VERY expensive. So there is always a question of where to draw the line.
If you ask a good QA organization to investiage ANY process they'll find 1000 things that you missed. If you fix all those then they'll find another 1000 things that you missed. At any time there are going to be memos showing that somebody had predicted anything that could have gone wrong. 99.99% of the time those predictions turn out to be wrong. When they turn out to be right of course those memos end up being "smoking guns", but if those memos were taken seriously all the time the cost of the program would be 100X higher.
The problem is that modern software and complex systems like manned spacecraft have enormous quanties of branch points and points of failure. To rigorously test every possible part under every possible condition requires more in spending than even large nations find acceptable. So, your choices are to either not try, or to try your best and accept the consequences.
There is one really big problem with risk-tolerance - when human life is at stake it just isn't acceptable to talk about accepting preventable risk of death. Think about it - any time anybody goes to have surgery performed and they don't seek out the most qualified surgeon on the face of the earth they accept more risk of death than if they had done so. This still happens all the time, but we just don't talk about it, because nobody wants to admit that they didn't fly their aunt suzie out to Boston to see Dr. X to have their appendix removed in a routine procedure.
Likewise, if you want to be truly quantitative about risk you need to indicate what risk of loss of life you're willing to accept and be realistic about everything. What normally happens is that management bows to politics (and/or lawyers in the private world) and either says that they'll accept zero risk to human life, or something crazy like a 1 in 1 billion chance of failure. In the case of zero risk everybody is deceiving themselves - getting out of bed carries a greater than zero risk of instant death. In the case of the really astronomical figures like 1:1E9 everybody just exaggerates all the figures to prove how reliable a system will be. The shuttle on paper is supposed to have a loss of something like 1 in 10k missions, and they've lost 2 in about 50. The reality is that the failure rate is much higher, as is the acceptable level of risk. However, it isn't popular to get up TV as a president and say that you're targeting a 0.1% chance of death if you ride the shuttle.
In the end you need to stop listening to what people say and watch what they do. The same folks who point out endless memos about O-rings probably drive to work in the morning, which says soemthing about the risks they're really willing to take with their own lives and the lives of their kids. They might be horrified about some of the memos floating around Detroit or even Tokyo.
Obviously you're right that management needs to better manage risk and listen to the engineers. However, society needs to allow managers to be more realistic about risk so that they can do so effectively. Everybody says they want software without bugs, but nobody would want to pay for it. I didn't get a pony either...
As others have pointed out, all of this is really just a matter of national priority.
As a taxpayer I'd hate to have my national government pay billions of dollars to put on an athletic competition. I can see how providing standard police services and such are within the scope of a government, but throwing a huge party and entertainment show isn't.
Look, humans are humans. There is nothing saying that the Europeans couldn't have landed on the moon if they wanted to spend that kind of money. The Chinese could have as well. Granted, at any given time particular nations have economies that are in various levels of repair - Europe or China probably couldn't have landed on the moon in the 60s even with a massively dedicated effort. However, either could probably do it today just fine.
Ditto for throwing an entertainment event. It isn't like Chinese acrobats are any better than Mexican acrobats. Just look at the Soviet Chess program - it had huge state sponsorship and unsurprisingly they turned out far better chess players than nations in which people played chess for fun almost entirely without compensation. Today the program is a shadow of its former self - and it isn't becuase Russians are being born dumber.
What all of this really demonstrates is the power of authoritarian governments to mobilize their entire nations around goals that are decided upon by a handful of those in power. A country with the economy of North Korea can mobilize more artillery tubes pointed at its rival than any nation in Europe. It isn't like the French don't know how to fashion a rifled cannon (gee, they've only been doing that for a century or so) - they just see the value in having millions of them pointed at Belgium. Likewise, I'm sure Germany could throw a 20 billion euro party if it wanted to, but I suspect the locals would rather see that money going into healthcare or maybe just into their own pockets.
Granted, democratic nations do bread-and-circuses too (aka Iraq), but they at least need to convince their populations to go along with it - without the benefit of highly self-censored media (although clueless media helps with slight self-censorship).
Quality is essential in any complex machine. Suppose the shuttle has 1 million critical points of failure. If each of them is 99.999% reliable then the chances of a successful launch is 0.99999^1000000 or 0.0045%. If you want to get off the ground you need to either reduce the number of points of failure (add redundancy or simplify the design), or increase the reliability of the parts (aka quality control).
If you want your bolts to have a tolerance of 1 um then you need a lathe that is calibrated umpteen times per day. Those bolts get individually packed in cotton and the box it is carried in gets followed by a procession of monks. The wrench used to tighten the bolt is also crafted with similar care, and operated by a $30M robot and not a human. When so much can go wrong the only way to prevent problems is to take extraordinary care with every step of the process. That costs a lot of money.
Software is the same way - everything is engineered with specs and written in something like ADA with extremely paranoid compile-time checks. Every function is tested on every boundary condition, every function call is carefully traced to ensure that the parameters will be in-range, etc.
And even so they occasionally lose a launch vehicle - even the best designs. What can you say - it isn't a cheap business to be in. That doesn't rule out private investment, but it does rule out cheap investment. I think that the only way it could be done privately would be if a company had a guarantee of profit in the event they got off the ground - the initial costs are just so high nobody would spend them if NASA might just decide to stick with their own rockets.
If you're benefiting from the U.S. economy, you are benefiting from a system built on the backs of slaves and laid bare by genocide. Your attempt to absolve yourself because you are new on the scene rings hollow.
If you're benefiting from ANY economy, you are benefiting from a system built on the backs of slaves and laid bare by genocide. Your attempt to absolve yourself because you are new on the scene rings hollow.
Fixed that for you...
Pretty much every long-lived institution on earth has done something really bad to somebody over the last 1000 years or so. At what point do you draw the line? If you go back far enough the whole planet was one big nature reserve with people living only in something the size of a reservation - should we restore the animals into their proper place as masters of the Earth?
At some point to move forward you need to stop looking backwards. Sure, I'm all for learning from the past to prevent future evils. However, trying to fix the evils of the past isn't constructive - it causes more harm to everybody than it remedies. Suppose Fred's great-great grandfather stole the home of Sam's great-great grandfather? Should we today take Sam's home and give it to Fred? But what if Fred is a great surgeon who donates all his income to charity, and Sam is a bum who squanders everything he gets on booze? What benefit to society is it to distract Fred from doing the good he does to benefit somebody who will just waste what is given to him?
Life isn't fair. Some of us start out with more money than others, and some of us have to make do with what we have. Some of us are born with 180 IQs and others are born mentally retarded. Some of us were born the sons of slaves, and some of us were born the sons of slave-masters. The only thing government can hope to achieve is to allow everybody to make the most of who and what they are - to the benefit of everybody. Shifting around wealth tends to discourage productivity - sometimes it is necessary but it needs to be minimized. It certainly can't be used just to make everything "fair" - because nothing ever is truly fair. Justice may be blind, but God isn't - why we aren't all cookie-cutter people cut from the same mold is a mystery to me, but I didn't create the world - I just have to live in it...
I think that WMD means that you spend a million dollars and kill at least 10k people. Rods from God are more like spending a million dollars and killing fewer people than a well-placed grenade. I guess cost doesn't matter as much as impact - a chemical leak isn't that expensive to generate but can kill quite a few people.
They certainly offer certain capabilities that make them attractive but they're hardly WMDs. And that has nothing to do with nationality.
Knowing a little about this stuff I figured I'd comment.
Quality is a big concern with pharmaceuticals. Even if the stuff coming from China was half-decent it probably couldn't be legally used to manufacture pills. For very good reason just about all modern nations require some variant of the GMPs ("Good Manufacturing Practices") to be followed when manufacturing drugs. The foundation of these principles are that everything is done following standardized processes, so that the first pill ever made and the pill that comes off the line 10 years later can be assured to be nearly identical in composition. That applies to the pills, and every ingredient that goes into them. Look up any common household item that can be bought with or without a "USP" (or EP/JP/BP/etc) designation - the latter will probably cost 3x as much but for the most part it is probably the same stuff. The difference is all the paperwork required to prove to regulatory inspectors that it is safe for use in people.
If that factory in China isn't a GMP facility then nobody could use it as a raw material for forumlating a pill. It isn't considered acceptable to just grab a random feedstock and test it - every step of the process has to be controlled. They may also have issues with impurities. Sometimes the best way to make a drug isn't the best way to make a chemical in general. The ideal manufacturing process for a non-drug might be prone to leaving impurities that are toxic, whereas a less-ideal process might have a lower yield but result in a product whose impurities are not toxic. Also - manufacturing processes might yield a salt or crystal form of a compound that varies in absorbtion or stability.
Pharmaceuticals that aren't stable in the stomach aren't uncommon - there are lots of common ways for handling these situations. Enteric coatings are fairly standard fare in the world of pharmaceutical manufacture.
All of these are well-understood problems in the pharmaceutical industry. They are solved probably 100 times for every drug that makes it to market (for every drug that makes it probably around 100 got as far as being formulated but were abandoned). It certainly costs money, but the real costs come later with the clinical trials.
My understanding it is considered in bad form to invoke cloture in the Senate even if one has the votes to do so. That means that a fair amount of legislation can be blocked by committees/filibusters/etc just by the virtue of everybody caring more about being polite than democracy.
I think that half the time representatives use these maneuvers as excuses, as in "well, I would have voted for it but those evil (somebody else)'s didn't let it come up for a vote." They're just as happy that it didn't come up for a vote - because then they'd need to go on the record and tick somebody off...
There are lots of them. The problem is that there aren't any which don't cost $10/month or impose a bandwidth cap like 100MB/month.
For non-binary you have quite a few options out there. For binaries your options are very limited. There are a few super-cheap or free (or near-free - one time small charge) options that are suitable for fills only (they have great retention and huge group selection, but very small bandwidth caps - great if you're missing one post out of a million). What you won't find is some place where you download your daily linux iso from a binary group for free.
A simple checksum will instantly determine if there is ANY variation from 100% perfection, and most software has such error checks built-in. Why yours does not, I can't guess.
What software of mine are you referring to? I don't recall mentioning having used any software that does or doesn't do comparisons of anything.
Obviously I agree that given two sets of digital data it is straightforward to determine if they are identical. You don't even need a checksum - just a straight comparison will do (and will avoid hash collisions as well).
I heard a story from a guy in the digital broadcast industry that actually did make me take a little pause.
Apparently some major artist was listening to a stack of CDs that were right off the presses. He commented to the engineer that he liked the ones in the one pile, but not so much the ones in the other stack he had made. The engineers of course chuckled inwardly since of course this was a digital reproduction and obviously the sound content would be identical between CDs. Then to humor the artist they played it and watched the signal in a spectrum analyzer, at which point they actually did notice some degradation in the CDs the artist had identified. This was traced to a poorly-created master CD in one of the presses, which all the defective CDs had come from.
So, digital isn't necessarily infallible, because ultimately everything in real life is analog. It just has a much higher tolerance for error before even the slightest degradation occurs.
Now, I'm still skeptical anytime an audiophile makes certain claims, but I do try to keep in mind that while they might not fully understand what is going on it is possible that somewhere in the microphone-recording-mixing-mastering-pressing-distribution-reading-decoding-amplification-cables-headphone chain something might have happened to make that digital reproduction less than perfect.
I think the difference in perspective can depend a bit on your viewing conditions.
There is no question that a 15Mbps broadcast is going to be higher quality than a 14Mbps broadcast. The only question is whether you can see the difference.
If you have a 40" LCD screen on the other side of your living room you're not going to tell the difference that having an SD weather subchannel makes. On the other hand, if you have a 72" plasma just far enough away from your face that you can catch both sides of the TV in your peripheral vision, you might actually be able to appreciate the extra quality.
Subchannels should certainly be encouraged - it gives consumers more for their sacrifice of RF spectrum (which does belong to the people). However, I don't see some guidelines as being unreasonable - perhaps a requirement that the primary channel maintain at least a certain amount of bandwidth. Also - perhaps networks could be permitted more than one station allocation if there is more than one unused allocation in the region (and if they all become used then after a year or two networks would be forced to give up their extra allocations to free up space again - the idea is to keep room for competition but at least use the frequency space that we have).
I think the market will also help - to the degree that consumers watch off-air programming. I can't believe that people actually use QAM to watch local stations (except where distance prevents decent reception) - the off-air station is free and multiples better in quality...
Slashdot Mirror
The current Taiwanese president's (who was popularly elected) stance on Taiwan's status is that he wouldn't push for independence, instead preferring to tread the fine line of the current status quo. The Taiwanese aren't actually loathing for (re)union with China, but they aren't unequivocally for independence either.
That is called diplomacy. The issue of Taiwan is a source of nationalistic pride in China and the issue is used like September 11th is in the USA. When you're a country not much biger than Luxemborg on the border of a nation with one of the largest militaries in the world who is looking for an excuse to invade you choose your words carefully.
Somehow I doubt that most Taiwan citizens would prefer Chinese citizenship. It would be a complete step backwards for them. As it is they have most of the conveniences of Chinese citizenship (due to Chinese policy) and none of the downsides. Kind of like Puerto Rico in the US...
And I didn't read People's Daily about the Tienanmen Square event. Did you use CNN/Foxnews as your sources?
Uh, I never said anything about the People's Daily coverage of Tienamen Square. They were used in a different illustration. Even so, I'm sure their coverage of the event was VERY tame.
Have you watched this before? http://www.tsquare.tv/ (note: it isn't made by the Chinese)
Actually, I haven't - I may take the time to do so (thanks for the link). I noted that they did have a link to Frontline on that site and I did see their episode dedicated to the aftermath of the incident.
I actually watch fairly little TV news programming - I tend to get most of my news on the web from RSS feeds (a fair variety of them). I doubt you'd ever catch me watching Fox News. If anything I'd be more likely to watch BBC's World News America or something along those lines for a broader perspective (I DVR it even if I don't often watch it).
I do strongly disagree with your assessment that Democracy simply wouldn't work in China. I don't disagree that it would be disruptive at first. There is a lot of history that the West went through prior to Democracy taking off and in many places that history simply doesn't exist. That doesn't mean that Democracy isn't ultimately what is needed in China (and everywhere else for that matter). Democracy isn't really a great form of government, but it just happens to be better than every other form out there (save appointing me dictator of the world). One thing that does tend to improve it seems to be the proportional system of representation used in most European states - not having two entrenched parties seems to help reduce the vote-buying phenomenon which is a big problem in the USA. Sure, Americans are a lot wealthier than the average Chinese, but that just means the politicians have to tax people and buy their votes using their own money...
Interesting with the comparison to slavery
At work I've tended to percieve that Indians are treated essentially as slaves. They've paid next to nothing. Many actually end up walking to work from nearby slum-like apartments on streets without sidewalks (in what is otherwise a very nice middle-class area). Work is dumped on them and they're expected to be incompetant. They're watched closely. The only thing that is missing is the whip. It has gotten to the point that when somebody sees an Indian at work the assumption is that they're essentially part of an underclass.
A few years ago this attitude didn't exist at all. Indian coworkers weren't viewed any different from caucasians or East-asians or anybody else. It seems like we're inventing slavery all over again... :(
Seriously, if anybody indeed has constructive ideas on how to substantially improve the situation in China, I'm sure a lot of people (including Chinese) would lend you their ears.
Uh, how about hold elections? The kind where anybody can run? And how about having a free press where one isn't punished for expressing one's opinions? That would be a good start and in the long run would substantially improve the situation.
In fact, westerners who are bashing China are usually completely out of touch with China.
Yup - you've hit that nail on the head. Why, most westerners actually think that the citizens of Taiwan actually want to be a separate country! What a crazy idea - fortunately, the People's Daily dubunks that rubbish and points out how they've been oppressed for decades and that the best thing the world can do is get out of the way and let mainland China liberate them.
The next thing those crazy westerners will be trying to tell us is that the Chinese governement actually masaccred students in Tienneman Square! What a load of propaganda!
Agreed 100%
Testing is actually a good part of quality development. However, the best testing is the kind of testing performed directly by the developers on small chunks of code (unit testing).
Too many companies view testing as something you do once the whole project is "done". The problem is that when you find problems at that point you end up just saying "oh well, we'll just live with it" and deploy it anyway. If you're going to do that, why bother testing at all? You need to spend your money on preventing problems, not discovering them.
Encouraging web browsers to ignore security irregularities and allow users to access sites that handle private information *without* bringing it to the user's attention is just irresponsible.
But that is exactly what firefox is doing. If a user connects to a site without using ssl they don't get any warnings and they can submit private information at will.
If a user connects to a site that uses a self-signed certificate, they are at less risk of interception than if they don't use any encryption at all, and yet the connection is treated as being "riskier".
If firefox were to disable all form submission to any site at all that doesn't use https with a trusted certificate at least I could see that they've being consistent...
No, you need both or it is useless.
This is patently false.
Is unencrypted http useless? It offers communications that are neither secure against evesdropping nor prevention of MITM attacks.
At least https with a self-signed certificate offers one of these two.
If firefox wants to scream when a user could be subject to a man-in-the-middle attack that seems fine to me. If that is the design goal, however, the browser should yell at me every time I visit slashdot - since the connection isn't protected against anything.
All an attacker needs to do to bypass firefox's protection is to do a MITM attack but use a non-encrypted connection back to the browser and SSL to the server. The browser won't complain a bit and the server won't know that anything is wrong. The user will think that everything is fine since he is now conditioned to the browser yelling at him anytime anything could go wrong.
As for #6, that doesn't even make sense. If the people who DON'T lead meetings are not going to survive, then who's going to attend these meetings? Meetings generally don't work well when everyone is trying to do the leading. Same can be said for a company's general operations. Remember the old adage about "too many chefs in the kitchen". US companies are not going to survive if they think they're going to make a lot of money by just sitting around, leading meetings, and having all the real work done in other countries by contractors
The parent post was not telling you want is needed for US companies to survive. He was telling you what you need to do to keep an IT job in the current climate. Current US coprate hiring practices do not have anything to do with acquiring the skills they need to survive. Eventually Darwin will catch up and hiring practices will change. Unfortuantely the typical corporation can last longer without income than the typical employee can.
When you're not expecting heavy action you could also use a multi-tiered staffing approach. You'd have a team of basically-trained staff that pilots the drones to and from points of interest, or monitors them during loitoring. You'd have a surveilance staff monitoring feeds from all the drones. When things heat you you'd turn over individual drones to dedicated strike crews - maybe with some crews having an A-A or A-G focus. You might also have the "special ops" crew that is every bit as trained as an F22 pilot is today for really critical missions.
There is no reason you need one pilot from liftoff to landing on one of these drones.
I very much realize the mistakes of Vietnam, but I wonder if things have changed since then. Back in Vietnam it wouldn't unheard of to discharge an entire load of missiles to only have one or two track. Today the practice of firing off more than one missile is still common, but that is mainly because in a typical mission you'd expect to only get one opportunity to shoot and even if the missile is 90% accurate two are better than one. Modern communications also makes it easier to identify foes beyond visual range and engage them at a distance.
And if you have a ton of unmanned aircraft all you need to do is stage them at various distances from the line - even if the enemy manages to surprise one or two of your picket aircraft the bulk of your forces can easily engage BVR. The loss of two aircraft is much more acceptable when they don't have pilots onboard. Indeed - if the other aircraft had clear shots they'd just shoot down the enemy formation including the two friendlys if it made sense to do so.
Perhaps the dogfight's days aren't completely over, but it seems that they're largely behind us. It would really take a very serious war with more of an equal to bring it back, and even then missiles are getting closer to the point where they are simply unevadable once you are within range.
It wouldn't surprise me if standard operational procedure after putting out any fire doesn't include a complete walk-through of the building to ensure that nothing is still smouldering. After all, somebody will get sued if the fire dept clears the building and then it catches on fire again after everybody goes back in.
Black body radiation. Gotta love it - no getting around it. All objects emit EM radiation - the frequency distribution and intensity is related to temperature.
If you don't want to put IR then you need to keep the temperature of the surface of this thing cool. The person inside is generating heat, and that heat has to go somewhere (as another poster has pointed out, humans don't work well if they can't perspire away their heat). I guess you can carry a big tank of LN2 with you. If you want you could radiate the heat directionally and only be visible in IR from that side, but that is obviously less valuable than just being invisible.
I'm not sure how well IR imaging works in daylight, however. If you're outdoors on a desert I suspect the IR blackbody radiation will be lost in the wash of heat coming off the ground.
I suspect that there is more to it than this.
Probably the bigger problem is that in order to legally advertise that a drug can be used to treat a disease you need to prove via clinical trials that the drug works. Those trials cost a few tens of millions of dollars each at a minimum. A full battery of trials costs into the 9-figures.
So, the company didn't want to pay for the R&D. Sure, you might "know" that it will work, but the FDA and its other first-world peers won't accept that.
The rules exist for a reason, but they really do make it hard to make drugs for conditions that don't have a significant market. Regulators should really think about this and come up with a better framework for regulating these kinds of drugs. When you require 9-figure development costs you aren't promoting patient safety in these cases - you're condemning patients to non-treatment.
Do some googling on off-label drug use to find out more about these sorts of problems. Once a drug is on the market doctors often bend the rules for their patients.
The specific issue you raise might very well also be a factor - particularly with biotech companies that tend to have lots of molecules but without the ability to develop them.
If I had a dollar for every university researcher who "cured cancer" I'd be a very rich man. Ditto for HIV.
For the most part the only really effective drug patents cover one specific molecule, with obvious non-functional additions/changes to it (you can't just stick a methyl on some huge molecule and call it an innovation).
If a drug company already has this product patented, then why haven't they released a cure?
Researchers tend to claim "success" when they come up with some molecule that has some inhibitory reaction in some in-vitro test. That is often a great lead, but only a very small fraction of these molecules turn out to be useful. Often they start a new line of thinking/questioning that does lead to the correct answer. It is a bit much to say that "most of drug R&D is government funded." That's like saying that because a government lab invented the first polymer that they should be credited with the commercialization of teflon. No doubt one wouldn't happen without the other, but there was certainly a lot of time, effort, and money spent in-between.
Overly-broad patents are clearly harmful. However, if a company goes to the effort to fund clinical trials on a molecule they should stand to benefit from this - beyond having a few months on the market before the first competitor scales up their own production of the same thing.
Massive government funding is a fair alternative to patented private drug development. I'm all for exploring that further. However, volunteer contribution is a bit unrealistic on these kinds of projects - they involve hundreds of clinics and thousands of doctors and scientists, with a truck's worth of paperwork to support their safety testing. Linux on the desktop is a trivial undertaking in comparision and "bounties" haven't quite made that happen yet...
I've seen this kind of issue in other quality-sensitive organizations.
The problem is one of risk tolerance. To truly engineer to all contingincies is VERY expensive. So there is always a question of where to draw the line.
If you ask a good QA organization to investiage ANY process they'll find 1000 things that you missed. If you fix all those then they'll find another 1000 things that you missed. At any time there are going to be memos showing that somebody had predicted anything that could have gone wrong. 99.99% of the time those predictions turn out to be wrong. When they turn out to be right of course those memos end up being "smoking guns", but if those memos were taken seriously all the time the cost of the program would be 100X higher.
The problem is that modern software and complex systems like manned spacecraft have enormous quanties of branch points and points of failure. To rigorously test every possible part under every possible condition requires more in spending than even large nations find acceptable. So, your choices are to either not try, or to try your best and accept the consequences.
There is one really big problem with risk-tolerance - when human life is at stake it just isn't acceptable to talk about accepting preventable risk of death. Think about it - any time anybody goes to have surgery performed and they don't seek out the most qualified surgeon on the face of the earth they accept more risk of death than if they had done so. This still happens all the time, but we just don't talk about it, because nobody wants to admit that they didn't fly their aunt suzie out to Boston to see Dr. X to have their appendix removed in a routine procedure.
Likewise, if you want to be truly quantitative about risk you need to indicate what risk of loss of life you're willing to accept and be realistic about everything. What normally happens is that management bows to politics (and/or lawyers in the private world) and either says that they'll accept zero risk to human life, or something crazy like a 1 in 1 billion chance of failure. In the case of zero risk everybody is deceiving themselves - getting out of bed carries a greater than zero risk of instant death. In the case of the really astronomical figures like 1:1E9 everybody just exaggerates all the figures to prove how reliable a system will be. The shuttle on paper is supposed to have a loss of something like 1 in 10k missions, and they've lost 2 in about 50. The reality is that the failure rate is much higher, as is the acceptable level of risk. However, it isn't popular to get up TV as a president and say that you're targeting a 0.1% chance of death if you ride the shuttle.
In the end you need to stop listening to what people say and watch what they do. The same folks who point out endless memos about O-rings probably drive to work in the morning, which says soemthing about the risks they're really willing to take with their own lives and the lives of their kids. They might be horrified about some of the memos floating around Detroit or even Tokyo.
Obviously you're right that management needs to better manage risk and listen to the engineers. However, society needs to allow managers to be more realistic about risk so that they can do so effectively. Everybody says they want software without bugs, but nobody would want to pay for it. I didn't get a pony either...
As others have pointed out, all of this is really just a matter of national priority.
As a taxpayer I'd hate to have my national government pay billions of dollars to put on an athletic competition. I can see how providing standard police services and such are within the scope of a government, but throwing a huge party and entertainment show isn't.
Look, humans are humans. There is nothing saying that the Europeans couldn't have landed on the moon if they wanted to spend that kind of money. The Chinese could have as well. Granted, at any given time particular nations have economies that are in various levels of repair - Europe or China probably couldn't have landed on the moon in the 60s even with a massively dedicated effort. However, either could probably do it today just fine.
Ditto for throwing an entertainment event. It isn't like Chinese acrobats are any better than Mexican acrobats. Just look at the Soviet Chess program - it had huge state sponsorship and unsurprisingly they turned out far better chess players than nations in which people played chess for fun almost entirely without compensation. Today the program is a shadow of its former self - and it isn't becuase Russians are being born dumber.
What all of this really demonstrates is the power of authoritarian governments to mobilize their entire nations around goals that are decided upon by a handful of those in power. A country with the economy of North Korea can mobilize more artillery tubes pointed at its rival than any nation in Europe. It isn't like the French don't know how to fashion a rifled cannon (gee, they've only been doing that for a century or so) - they just see the value in having millions of them pointed at Belgium. Likewise, I'm sure Germany could throw a 20 billion euro party if it wanted to, but I suspect the locals would rather see that money going into healthcare or maybe just into their own pockets.
Granted, democratic nations do bread-and-circuses too (aka Iraq), but they at least need to convince their populations to go along with it - without the benefit of highly self-censored media (although clueless media helps with slight self-censorship).
Quality is essential in any complex machine. Suppose the shuttle has 1 million critical points of failure. If each of them is 99.999% reliable then the chances of a successful launch is 0.99999^1000000 or 0.0045%. If you want to get off the ground you need to either reduce the number of points of failure (add redundancy or simplify the design), or increase the reliability of the parts (aka quality control).
If you want your bolts to have a tolerance of 1 um then you need a lathe that is calibrated umpteen times per day. Those bolts get individually packed in cotton and the box it is carried in gets followed by a procession of monks. The wrench used to tighten the bolt is also crafted with similar care, and operated by a $30M robot and not a human. When so much can go wrong the only way to prevent problems is to take extraordinary care with every step of the process. That costs a lot of money.
Software is the same way - everything is engineered with specs and written in something like ADA with extremely paranoid compile-time checks. Every function is tested on every boundary condition, every function call is carefully traced to ensure that the parameters will be in-range, etc.
And even so they occasionally lose a launch vehicle - even the best designs. What can you say - it isn't a cheap business to be in. That doesn't rule out private investment, but it does rule out cheap investment. I think that the only way it could be done privately would be if a company had a guarantee of profit in the event they got off the ground - the initial costs are just so high nobody would spend them if NASA might just decide to stick with their own rockets.
If you're benefiting from ANY economy, you are benefiting from a system built on the backs of slaves and laid bare by genocide. Your attempt to absolve yourself because you are new on the scene rings hollow.
Fixed that for you...
Pretty much every long-lived institution on earth has done something really bad to somebody over the last 1000 years or so. At what point do you draw the line? If you go back far enough the whole planet was one big nature reserve with people living only in something the size of a reservation - should we restore the animals into their proper place as masters of the Earth?
At some point to move forward you need to stop looking backwards. Sure, I'm all for learning from the past to prevent future evils. However, trying to fix the evils of the past isn't constructive - it causes more harm to everybody than it remedies. Suppose Fred's great-great grandfather stole the home of Sam's great-great grandfather? Should we today take Sam's home and give it to Fred? But what if Fred is a great surgeon who donates all his income to charity, and Sam is a bum who squanders everything he gets on booze? What benefit to society is it to distract Fred from doing the good he does to benefit somebody who will just waste what is given to him?
Life isn't fair. Some of us start out with more money than others, and some of us have to make do with what we have. Some of us are born with 180 IQs and others are born mentally retarded. Some of us were born the sons of slaves, and some of us were born the sons of slave-masters. The only thing government can hope to achieve is to allow everybody to make the most of who and what they are - to the benefit of everybody. Shifting around wealth tends to discourage productivity - sometimes it is necessary but it needs to be minimized. It certainly can't be used just to make everything "fair" - because nothing ever is truly fair. Justice may be blind, but God isn't - why we aren't all cookie-cutter people cut from the same mold is a mystery to me, but I didn't create the world - I just have to live in it...
I think that WMD means that you spend a million dollars and kill at least 10k people. Rods from God are more like spending a million dollars and killing fewer people than a well-placed grenade. I guess cost doesn't matter as much as impact - a chemical leak isn't that expensive to generate but can kill quite a few people.
They certainly offer certain capabilities that make them attractive but they're hardly WMDs. And that has nothing to do with nationality.
Knowing a little about this stuff I figured I'd comment.
Quality is a big concern with pharmaceuticals. Even if the stuff coming from China was half-decent it probably couldn't be legally used to manufacture pills. For very good reason just about all modern nations require some variant of the GMPs ("Good Manufacturing Practices") to be followed when manufacturing drugs. The foundation of these principles are that everything is done following standardized processes, so that the first pill ever made and the pill that comes off the line 10 years later can be assured to be nearly identical in composition. That applies to the pills, and every ingredient that goes into them. Look up any common household item that can be bought with or without a "USP" (or EP/JP/BP/etc) designation - the latter will probably cost 3x as much but for the most part it is probably the same stuff. The difference is all the paperwork required to prove to regulatory inspectors that it is safe for use in people.
If that factory in China isn't a GMP facility then nobody could use it as a raw material for forumlating a pill. It isn't considered acceptable to just grab a random feedstock and test it - every step of the process has to be controlled. They may also have issues with impurities. Sometimes the best way to make a drug isn't the best way to make a chemical in general. The ideal manufacturing process for a non-drug might be prone to leaving impurities that are toxic, whereas a less-ideal process might have a lower yield but result in a product whose impurities are not toxic. Also - manufacturing processes might yield a salt or crystal form of a compound that varies in absorbtion or stability.
Pharmaceuticals that aren't stable in the stomach aren't uncommon - there are lots of common ways for handling these situations. Enteric coatings are fairly standard fare in the world of pharmaceutical manufacture.
All of these are well-understood problems in the pharmaceutical industry. They are solved probably 100 times for every drug that makes it to market (for every drug that makes it probably around 100 got as far as being formulated but were abandoned). It certainly costs money, but the real costs come later with the clinical trials.
My understanding it is considered in bad form to invoke cloture in the Senate even if one has the votes to do so. That means that a fair amount of legislation can be blocked by committees/filibusters/etc just by the virtue of everybody caring more about being polite than democracy.
I think that half the time representatives use these maneuvers as excuses, as in "well, I would have voted for it but those evil (somebody else)'s didn't let it come up for a vote." They're just as happy that it didn't come up for a vote - because then they'd need to go on the record and tick somebody off...
There are lots of them. The problem is that there aren't any which don't cost $10/month or impose a bandwidth cap like 100MB/month.
For non-binary you have quite a few options out there. For binaries your options are very limited. There are a few super-cheap or free (or near-free - one time small charge) options that are suitable for fills only (they have great retention and huge group selection, but very small bandwidth caps - great if you're missing one post out of a million). What you won't find is some place where you download your daily linux iso from a binary group for free.
Undoubtedly true. However, I've never created a master CD in my life - nor do I work in the audio engineering or duplication professions.
A simple checksum will instantly determine if there is ANY variation from 100% perfection, and most software has such error checks built-in. Why yours does not, I can't guess.
What software of mine are you referring to? I don't recall mentioning having used any software that does or doesn't do comparisons of anything.
Obviously I agree that given two sets of digital data it is straightforward to determine if they are identical. You don't even need a checksum - just a straight comparison will do (and will avoid hash collisions as well).
I heard a story from a guy in the digital broadcast industry that actually did make me take a little pause.
Apparently some major artist was listening to a stack of CDs that were right off the presses. He commented to the engineer that he liked the ones in the one pile, but not so much the ones in the other stack he had made. The engineers of course chuckled inwardly since of course this was a digital reproduction and obviously the sound content would be identical between CDs. Then to humor the artist they played it and watched the signal in a spectrum analyzer, at which point they actually did notice some degradation in the CDs the artist had identified. This was traced to a poorly-created master CD in one of the presses, which all the defective CDs had come from.
So, digital isn't necessarily infallible, because ultimately everything in real life is analog. It just has a much higher tolerance for error before even the slightest degradation occurs.
Now, I'm still skeptical anytime an audiophile makes certain claims, but I do try to keep in mind that while they might not fully understand what is going on it is possible that somewhere in the microphone-recording-mixing-mastering-pressing-distribution-reading-decoding-amplification-cables-headphone chain something might have happened to make that digital reproduction less than perfect.
I think the difference in perspective can depend a bit on your viewing conditions.
There is no question that a 15Mbps broadcast is going to be higher quality than a 14Mbps broadcast. The only question is whether you can see the difference.
If you have a 40" LCD screen on the other side of your living room you're not going to tell the difference that having an SD weather subchannel makes. On the other hand, if you have a 72" plasma just far enough away from your face that you can catch both sides of the TV in your peripheral vision, you might actually be able to appreciate the extra quality.
Subchannels should certainly be encouraged - it gives consumers more for their sacrifice of RF spectrum (which does belong to the people). However, I don't see some guidelines as being unreasonable - perhaps a requirement that the primary channel maintain at least a certain amount of bandwidth. Also - perhaps networks could be permitted more than one station allocation if there is more than one unused allocation in the region (and if they all become used then after a year or two networks would be forced to give up their extra allocations to free up space again - the idea is to keep room for competition but at least use the frequency space that we have).
I think the market will also help - to the degree that consumers watch off-air programming. I can't believe that people actually use QAM to watch local stations (except where distance prevents decent reception) - the off-air station is free and multiples better in quality...