No you don't. You do need a properly randomised sample though. Asian people are generally shorter than Caucasian, and people from North Korea even more so, so a sample which was not properly random (say taking 30 men from North Korea and 30 women from rural Germany) would be misleading. But a sample of 30 random people across the entire planet would not, even if we didn't control for height. Weighting for regional height differences would give us a better understanding, and increase our statistical power, but you do not need to control for all variables, and in fact in most research controlling for all variables is impossible. Even in something like basic physics what you are suggesting is impossible. All g-2 measurements for the electron for instance have been conducted on Earth. We have no way of knowing if that works outside of a stars gravity field or in a black hole. It is still reasonable to assume it is.

The sample in climate research is adequately controlled, sufficiently random and the noise is well enough understood that it is likely most of the conclusions of the research are useful and predictive.

Mmmmm those cherries are so good, I see why you picked them:
http://www.climate4you.com/GlobalTemperatures.htm#Global temperature trends
Care to admit why you picked 10 years and not 15 or 20?

If you grab a sample of 2 women and 2 men you may well find the women are taller, and you wont be able to say based on that sample if men or women are taller on average. But given 20 or 30 women and 20 or 30 men the answer becomes obvious.

Very strongly depends on what you are doing with MATLAB vs. scipy and numpy as to which is faster, although my experience is that on the whole MATLAB is a bit faster (although my experience is pre-MKL so maybe MATLAB is much faster now, if so my colleagues haven't noticed). A factor of 2-10 seems reasonable. On the other hand I have regularly ran into memory problem with MATLAB when I've used it with large data sets that I don't encounter with python, of course that is probably because I'm much better with python than MATLAB (which is the other problem with MATLAB vs. python, I find it much easier to vectorise stuff for numpy so it might be theoretically possible to do stuff faster in MATLAB but in practice I just cant get the behaviour I need).

Octave is nice, but painfully slow, at least when I've used it. I mostly use it when I have to teach students who have used MATLAB and I need a substitute. If you are in a position where you need MATLAB day to day I don't think octave is there as a replacement yet.

Prototyping in MATLAB is okay, but I find it is at the ugly spot in my discipline where it is no faster than python for the quick stuff because fractions of a second don't matter, and too slow for the slow stuff. I also think python has a slightly nicer syntax and structure, and is easier to teach to students, so long as I find them a text editor that doesn't balls up the whitespace. I find they learn much faster and their code is generally cleaner.

Bottom line is if I needed speed that badly I would write in C++, C, or FORTRAN (yup, trained physicist). For me a factor of 10 isn't worth the extra time it would take me to prototype, and if I need speed that badly, I probably need a lot of speed.

Will look into gnuradio, hasn't crossed my radar before.

Simply using the word 'shortage' is advocating against workers. That is arguing for anti-worker actions. Education funding is directed in part in accordance with what businesses say they need. You saying there is a shortage when there is none does damage to workers interests.

You say you wish that but I suspect your actions speak otherwise. How much did your business donate to your local university for the purposes of supporting elite students? How much do you spend on open ended training of your employees (and no, indentured servitude where they have to agree to stay with the company for X years in return for training does not count, I'm talking about them getting training and then you paying them more as a result with them remaining free to leave if you try to swindle them). Mentoring your employees doesn't count either, I'm talking about something they can take to the bank, something that gets them paid better.

The only way someone who owns and runs a business gets to complain and have me take it seriously is if they put large sums of their money where their mouth is, otherwise they are just trying to screw us.

Then pay seven figures. You pay your CEO seven figures so if you really need someone with an advanced skill set you are prepared to pay for it. If they aren't that valuable to you then you don't need them and there is no shortage. Is basic economics really this hard to understand?

Look it is very simple, to show a shortage (which is a short term phenomena) you need to show some event which is causing either supply or demand to sky rocket and a resulting rapid shift in price. For example after an oil shock there is a shortage of oil because oil is partly controlled by a cartel so the market is slow to respond and can be manipulated. If say a virus went round killing 90% of all STEM workers then we would have a shortage while we trained up new people and in the mean time STEM workers salary would rise until it matched the highest offers on the left of the demand curve. You want to show me a shortage, show me that STEM workers salaries have ballooned, show me where all the dead tech workers are, the rapid drop in STEM graduates, the sudden rise in people looking to employ STEM workers. You want to know what a shortage looks like? It looks like the dot com boom. That was a shortage of tech workers. Price of tech labour sky-rocketed to absurd levels, a new source of demand rapidly opened up faster than the labour market could respond and we had a shortage.

Show me a dot com boom or a supply shock or shut up and stop trying to make arguments whose obvious purpose is to build support for policies that will screw labour.

If there is no problem, there is no shortage. Don't call it a shortage. Don't argue for anti-worker actions that would address a non-existent shortage.

I want chocolate ice cream in a cone. I'm not under the delusion I don't have to pay for it though. And when I walk into the store and don't see them priced at 20 cents a piece I don't complain there is a shortage of them. I don't try to get government to give me a subsidy on chocolate ice cream. I shut the fuck up and pay the market price. Shut the fuck up and pay the market price.

Don't care how you do it, the free market will solve the problem. If you have an actual shortage you would do it. You are not doing it, ergo you don't have a shortage. The market sets the price, your rates are not reasonable if you cant fill them, by definition. This is the lesson people in business have taught worker, forced on worker, mandated to workers for years now.

Is differentiating the great from the adequate? Pay for universities to run harder more advance courses so you have an easier time differentiating candidates. If it mattered to you, you would pay for it. Or any number of other options you could do. You don't, so it doesn't.

Then you go out of business. Welcome to market economics. Become more efficient like workers are constantly told they should.

You do not have a shortage of good applicants, such a shortage is impossible in a market system like we have. What you have is too low a price point. Quadrupedal the offered pay rate and you will find plenty of such applicants, because you will be able to poach them from other companies for a start. I cant help but feel that any employer who ever mentions the word 'shortage' in relation to labour should be immediately required to increase the pay they give the relevant employees by 20% and handed a leaflet explaining exactly how market economies work.

Hanlon's Razor doesn't really work when each individual entity has a reason to be malicious. Don't rule out malice after all.

Your like the corrupt cop who says "When I take bribes to arrest my paymasters political enemies people hate me, but when I let the perp go because 'fuck it' people still hate me! I cant win!". No shit bombing people on behalf of Haliburton, propping up dictators while ignoring genocide is going to get you on peoples naughty list.

I think the MATLAB vs. python environment is something of a horses for courses situation. Spyder does a decent job replicating the MATLAB feel, and I generally prefer working in ipython than in the MATLAB terminal because the MATLAB terminal is generally speaking more sluggish. Of course I'm running it on top of Linux, so I don't know what it is like on Windows, and I suspect the situation is probably reversed.

As for importing everything in one go: ipython --pylab gets you 90% of the way there. The remaining 10% is really not worth it as it will import a whole bunch of packages you don't need, slow down things a bunch and makes it very difficult to work out the dependencies of your program (unless you keep the namespaces, which can be a royal pain in the arse in and of itself).

Who started it matters a great deal.

And your own case works against you. The German strategic bombing efforts were poor, the British knew they were poor and they knew the only effect of them was to piss off civilians and motivate them further to fight. Neither side had any idea what a massive bombing campaign which tore the heart out of a country would do to civilian moral (although both were pretty sure what it would do to industrial output). There were reasons to suspect it would diminish morale if you could strike hard enough, and Speer admitted that there were a few occasions when he was concerned for civilian morale, although you would rightly point out that he asserted the allies didn't have the capacity to maintain the pressure to realise his fears. Those reasons, as again you rightly point out, turned out to be fallacious. That is why I would oppose a strategic bombing campaign in a contemporary war unless extreme methods were used to reduce civilian casualties. The allies had no way of knowing for certain if their strategic bombing campaigns were working though. And they certainly damaged Germany's productive capacity even if their effect on morale was a net negative for the allies.

The Germans and Japanese don't get to use the fact that they were shit at strategic bombing as an excuse for why the allies shouldn't do it. They normalised it, they had to live with the consequences. The Second World War reminded us of an important and valuable lesson, don't normalise an atrocity, because your opponents capacity to actualise it might be greater than yours and once you have made it a normal part of the war, you might suffer it far worse than your enemy.

As for atrocity always being counter-productive to war effort, sadly it is easy to imagine scenarios where this is not the case. Consider if, during the tension in 1980's the Soviet Union had launched a massive nuclear strike on NATO in response to Able Archer. Clearly an atrocity, but what would happen if the West did not respond in kind (and responding in kind would also be a clear atrocity)? NATO would lose.

Not disagreeing, but usability is only one factor. In addition the MATLAB stats package are individually easier to learn, but have no consistent interface making them globally more of a chore. If you only ever do one kind of ANOVA then MATLAB has you covered. I know plenty of academics who only ever use an ANOVA, although some of those really, really should be using something else some of the time.

Never said we were the good guys. I said you cant commit an atrocity then act shocked when that atrocity is normalised. If Germany and Japan didn't want the bombing of cities done to them, then they needed to not do it to others first.