If you really want to use this method to calculate pi, here's how to actually go about it. What you need is a hundred yards or so of string, four stakes, a stick and something that's a reasonable approximation to a right-angle (perhaps a piece of a cardboard box salvaged from the apocalypse). If you're really stuck for a right angle you can construct one with three stakes and a piece of string by putting two stakes in the ground and using the string to mark a straight line between them, then tying one end of the string to one of the stakes and tying the third stake to the string, so that length of string between them is a bit over half the distance between the stakes in the ground. Mark out a circle using this. Then mark out a second circle with the other stake in the ground as the centre. These two circles will intersect at two places - use the string to mark a straight line between them. The two straight lines you have marked will be at right angles.

Now put two stakes in the ground, about 20 yards apart. Stretch string between them. Put your right-angled thing with one side against the string and the right-angle corner at one of the stakes. Measure another piece of string to be the same length as the piece stretched between the two stakes. Tie it to a third stake and stretch it out so that it runs along the other side of the right-angled thing. You've now marked out two sides of a square with string. Repeat to form the other two sides.

Take your stick and break it down to about a foot long. Use it to mark out on the ground equally-spaced marks along each side of the square. Get two people to hold each end of a fifth piece of string across the square so that you can mark straight lines on the ground, dividing the square into a grid.

Cut your fifth piece of string to be the same length as one side of the square. Tie one end to one of the stakes. Now use the other end to mark out an arc from one corner of the square to the opposite corner.

Count the number of squares that are inside the arc and the total number of squares. Take the ratio of these two numbers and multiply it by 4. Here is your approximation to pi.

This method has many advantages over the one proposed: With the dimensions given above, it gives a considerably better answer, correct to four significant figures (3.141). It is easy to scale for better accuracy - make the square 100 yards and the stick four inches and you get six correct digits (3.141590123). You don't need to correct for uneven shot pattern. And, crucially I'd say in an apocalypse, you don't need a shotgun or ammunition and, if you do happen to have them, you can use them for useful things like fending off the zombies or hunting.

I thought the engineer's answer was, "More than three and probably less than twenty."

Oh, no, wait, that's the answer to "What's 2+2?"

Last time I checked, actually, it is. It's just not in the USA.

A staggering number of people commenting here appear not to understand English, let alone French or Italian.

Well colour me puzzled. Surely the expression "whatever the French call la dolce vita" demonstrates that, whatever the French do call it, they don't call it la dolce vita? So he knows it's not a French expression, he just doesn't know what the equivalent expression in French is.

Well done for supplying the French equivalent.

Yes and no. In the UK the price has roughly quadrupled in that time, but the real increase (ie increase over RPI) is only 33%[1]. Assuming that overall rate continues (very unlikely, but there you go) it takes over 70 years for the price to double.

[1]http://www.speedlimit.org.uk/petrolprices.html

Well, yes, but the vast majority of the difference is in tax, not in production costs; $3/gal supplied to the US Navy is probably something like £9 per litre in Crawley.

300,000 in five hours? God forbid!

Depends on where in the supply chain that $3/gal is. $3/gal supplied to the US Navy is probably more like $7 or $8 at the pump for putting in your car - not so viable.

Once electric cars become prevalent, the charging time doesn't really matter for the supply and HV distribution side of the grid - each car sucks either 10.2MW for 30s or 10.2kW for a bit over eight hours (30,000s). Once there are enough that the spikes in charging smooth out, the demand increase is the same whichever charging rate you use. The only problem really comes at the edge of the grid, with the connection to individual houses currently being sized about three orders of magnitude wrong for this use. At this point, it's probably not too unreasonable to ask homeowners to pay to have their grid connection upgraded to give them the privilege of a 30-second charge for their car.

Let's see, a 4,700mAh 5V battery has a capacity of 23.5 VAh or 84.6kJ. To charge that in 30s, you'll need a 2.82kW charger output. So whether it's feasible or not probably depends on what jurisdiction you're in - a British 240V 13A socket will give you 3.12kW, so as long as your losses are below 10% you'll just get it. An Australian 240V 10A socket will give you 2.4kW, so allowing for 90% efficiency of the charger you'll get about 40s to charge. A US 110V 15A socket will give you 1.65kW, requiring about 57s at 90% efficiency to deliver a full charge.

Of course, I'm sure now I'll be buried in crap telling me either a) how crap Wordpress is and how much better blogging-platform-XYZ is or b) how useless the advice is because Bennett will never follow it and how my metrics for assessing the advice I give out is all wrong.

Seriously, Bennett? You don't think maybe researchers who devote their lives to figuring out good advice on health, diet and exercise know just a teeny bit more about experimental design than you? Sorry, I forgot, teenagers know everything.

Uh huh. Can someone tell Bennett about Wordpress so he can go get a real blog, please?

In Soviet Russia, dash cams are compulsory!

Hmmm, that didn't quite work out like I wanted.

I dunno; put it in an acid bath and see what happens.