Because the from address is invariably forged, you do nothing with a bounce. In fact, it's worse than nothing, because you create backscatter. I have suffered from backscatter and it is a pain - it just multiplies the spam problem. So, could I request that you just stop it!

If you actually know the person who is sending you the email then you should try a more personal approach rather than a passive aggressive bounce.

While the spammy advertisement would normally warrant no attention, it does raise a point that is worth noting:

Because the from address is invariably forged, you do nothing with a bounce. In fact, it's worse than nothing, because you create backscatter. I have suffered from backscatter and it is a pain - it just multiplies the spam problem. So, could I request that you just stop it!

Consider also that most researchers run more than 20 regressions when testing their data. That means that the 95% significance level is grossly overstated.

The key is that the 95% level applies only if you don't data snoop beforehand and only if you run the regression once and only once. The true significance levels of many studies that claim a 95% level is likely to be 50% when you consider all the pretesting and data snooping that goes on in reality - not the rather idealised setup that is reported in the journal publication.

Spotlight.

When you have a search function that works - really works - organisation isn't so important.

For example, the difference between my use of Mail at home under OS X and Outlook at work is mindboggling. With Mail, I type in a few key words and by the time I've finished typing there is a list of relevant emails, usually containing the one I'm looking for. At work, I have yet to find anything in Outlook using the search function - half the time it doesn't even finish searching before I give up and try something else.

There is a perfectly good name for the thing that some people call "vegetarian chilli con carne" - it is chilli. It reflects the problems that occur when words cross languages and people don't understand what they mean ("vegetarian chilli with meat" - really?). This is not the same as a fake Rolex. But if you think vegetarian chilli con carne is acceptable then so too should be "vegetarian chilli-con-carne with meat" and so on to any order or recursion you can stand.

Very, very wrong.

Rolling resistance is not related to the coefficient of friction of the rubber. It is primarily related to the suppleness of the tyre casing allowing it to roll over minor variations in the road surface without giving up lots of energy in heat associated with the deformation of the tyre. (It is also related to inflation pressure - which is one reason why car manufacturers recommend you maintain your tyres at optimum pressure.)

These sorts of tyres are well know for bicycles and the effects can be dramatic and noticeable when it is your own legs supplying the power rather than an engine. Indeed, the tyres with the lowest rolling resistance also tend to have the stickiest rubber. When I ungraded my bike tyres I got: lower rolling resistance, better cornering, and better wet weather performance. On a bike at least, they also give a more comfortable ride because you are not bouncing over the micro-contours of the road nearly as much.

Here are some additional details for those of you so inclined.

Consider a simple binary choice question. This is easily modelled by the binomial distribution which has well understood distributions. (Other distrbutions may be relevant but the principles remain pretty constant across them all.) The standard deviation is given by sqrt[np(1-p)] where n is the sample size and p is the probability of the observation you are interested in (the mean is np so in what follows I will be dividing by n to talk about percentages if you are taking notes). For example, are you male? If the true p is, say, 75% then you need a sample size of approximately 833 to get a 95% confidence interval (2 s.d.) of +/- 3%.

You might also note that the closer the true p is to 50%, the larger the sample size needed. If the true p is 50% you need a sample size of approximately 1100 for the same confidence interval. Furthermore, if you want to get it within 1%, the sample size goes up dramatically - to 10,000.

The population size is pretty much irrelevant. The population matters for ensuring that your sampling is truly random, but political pollsters can use the same sample sizes in Australia (pop ~20 million) as in the US (pop ~300 million) for similar accuracy. (Sampling bias is the reason that political polls can be out by so much - if you call households during work hours you are going to get a very different sample of people than if you call at dinner time.)

Because they didn't accept foreign drivers licenses and insisted on passports from foreigners. They had a book of all US state licenses and wouldn't accept anything not in the book (because it could be a forgery and they'd never know). Some bars also rejected the Massachusetts official non-drivers license IDs for some bizarre reason. Some bars accepted foreign licenses some of the time (depending on whether they had been busted in the past 3 months or not), so it was kind of a crap shoot. And it is no fun to go out for a drink with friends and have to sit there watching them drink while you sip on a Coke.

I spent a number of years in Boston as an Australian.

The only time it was necessary to carry a passport was when I wanted to have a drink at a bar. They card everyone. Unfortunately, carrying your passport and getting drunk are not circumstances that generally lead to a happy conclusion.

No. Apple displays competitive behaviour. 'Comptition' is pretty cut-throat and there is never any love lost between competitiors. Such behaviour only becomes 'anti-competitive' (i.e. contrary to the Sherman Act or similar) when you have a monopoly. For example, a new startup wants to get their product out there so they give away free samples; fine if you are a startup with no market power, but not if you are a monopoly who is thereby foreclosing competition.

Apple also displays control-freak behaviour. Being a control freak and being a monopolist are two very distinct things. Not all control-freaks are monopolists and not all monopolists are control freaks.