Well, that. Also, let us assume that

1) Car theft was a legal minefield like downloading copyrighted material is.
2) The chances of getting away with it are rather good.
Now, add to that the notion that one would not deprive the previous owner of his car. Yeah...how many people would steal a car? For an example, look at the Bosten police strike:
http://en.wikipedia.org/wiki/Boston_Police_Strike#Strike

With the chances of getting caught going to zero, the incidence of crime skyrocketed.....I'll let you draw your own conclusions from that.

I think you have this figures messed up... condoms fail about 10-18% of the time. I believe that the 'rhythm' method is about 25% effective, now that means 75% failure rate. Also, this is not a per time figure, this is over a year of regular sex, no other contraceptives, 3 in 4 couples on the rhythm method get pregnant. Compare that to 1 in 5 couples for condoms. I know which I am using....

I think part of the reason for the cautiousness was this. In the US, everybody would be screwed (well, save a long drive), whereas Europe has an extensive railnet that is fast and efficient. Hence, we can afford to be overcautious.

Besides, I view this whole thing as an inherent risk of the flying bussiness. If the tides are too high for the ferries to go, no one compensates them either, nor do the bus companies get compensated for slippery roads, or any other transportation companies if problems arise. If everybody was going to go bankrupt over it, they might have a case (we need airlines), but they aren't.

That's complete BS. When it comes to live saving drugs, trials aren't placebo controlled for, obvious, ethical reasons. How would you like to tell the family that mommy could have lived another 10 years, but instead died in the interest of a stronger experimental procedure?

Bingo! I think most researchers are too busy with research and grant-finding to really bother...but the whole idea is insane.

Also, most people may not have the resources to do the research, nor the skills, when it comes to biochemistry. However, think of any CS field, or perhaps mathematics, theoretical physics and statistics. All you need to play along is a computer, some programming skills and the drive to read an article and study hard enough to understand it (well, brains help too). Especially in these fields, opening up science may greatly speed up the pace and scope of the science.

Finally, it has nothing to do with being able to understand it, being able to replicate it, or even being able to do anything meaningful with it.
1) The publishers cannot know whether or not I can do anything meanigful with it, I might be a major hobbyist in Biochem, and have a fully equipped lab in my basement....highly unlikely, but they cannot know that. 2) Tax dollars are involved, I want what I paid for.

I believe MechWarrior (at least 3 and 4) had this control scheme, you could pivot the arms on your mech separately from the torso.
This was advantageous because sometimes you'd hit another mech and this would prevent your torso from turning. Also, mindlessly blowing pulse lasers in some direction was occasionally useful (basically to prevent someone from standing still and lining up a shot).
In short, I can see uses for it, especially in tactical shooters with one-shot-one-kill dynamics where getting shot at is a very good reason to immediately seek cover. If we're talking UT here, yeah, the point becomes a bit moot.

Apparently, the legality of grooveshark is doubtfull: http://en.wikipedia.org/wiki/Grooveshark#Legal_issues Also, it keeps dying on my over here (Belgium), which sucks.

Of course, every market has its issues. This is true for industry as it is for science. No one in science denies this, but I think some people outside of science have a slightly too romantic view of it.
As for the multi-pub thing: The case is isn't that if you have some severly deficient research you should just publish it, more that you
1) Shouldn't see your paper as the last one (either by your lab, or by others)
2) Shouldn't try to cover ever inch of the matter.

As you state:

And in the light of this, your statement does make some more sense.

I think the main problem, with China and elsewhere, is that it is very, very difficult to assess the quality of someone's scientific work. As a result, pubs and impact factor have become the standard of choice, and it has brought forth a mercenary attitude. Because of this, people have attempted less than honest tactics. Sadly, the only solution I see is either removing the entire meritocracy in science, or a complete reworking of the system in some way (don't ask me how). In short, I doubt that the problem is really China or Chinese culture, but has more to do with the way science is currently organized.

Well, evidently you have never published, or if you have, you have never run a larger (multi-pub) project. In this case, you'd publish and then proceed to do the background checks. If the background checks fail, you can publish those as well. If your original research turns out fine, you tack on some additional (original) research and publish that. Also, given that it worked in a single case, you are evidently on to something. Ergo, checking again is, at this point, a waste of time. You share your findings with the world, and then have other people run with it as well. More on-topic: I have seen a lot of Chinese, and more generally Asian, papers in my field... but not one of them is original. Also, doubtful results do pass by from time to time (although verifying this is hard, when it comes to sattelite observations there's no doing it twice). It seems that Chinese scholars (based on the ones I know and the research I see) are more concerned with quantity, as it improves your scholarly standing very directly, than with quality. So reproducing research (in my field: doing data assimilation on soil moisture for the umphteenth time) is a quick and easy way to get this.

Well, they've been doing that for god knows how long, but not in the way you think. You see, most weed groweries bypass the meter so they don't have to pay gargantuan energy bills. So, instead the power company looks at the discrepancy between billing and consumption at the block level. If a large enough discrepancy is noted, i.e. something big, they inform the DEA. Note: I assume they do it like this in the USA, as this is how many countries (including my own) do it.

Well, in most social democracies smoking is good for the budget. People always whine about increased health care costs, but where do you think those smoker's pensions go when they kick the bucket 10 years earlier?

Being a PhD student who mostly works with statistics (even though I have little formal training in it), I can attest to the truth of this. Sometimes you get results that do not vibe with what you are seeing, or that are doubtfull in some other way. Of course, we can blindly listen to the stats, or we can find out why the results are as they are. I try to do the latter, but the 'SPSS effect' tends to promote the former.