Both of these cars were competing in the same class, but the difference is the solar array. All of the Nuna cars have had multijuction arrays made of III-V semiconductor cells. Such cells are very expensive, and are only used in satellite or concentrator applications, but they can have efficiencies up to 35%. Enough cells to cover a solar car can easily be more that a million US dollars. Sunswift IV, on the other hand, has an array composed of Sunpower C-60 solar cells, which have a theoretical efficiency of 22%. The record they set was for a solar car with silicon solar cells, and as someone mentioned, they did the run without a battery pack. The wikipedia quote for theoretical top speed is running solely off battery, which doesn't make it much of a SOLAR car.

If you're willing to spend a little more money and effort, you can get a used Parker 51 on ebay for less than $50. It's by far the best pen I've ever used and has lasted me five years so far. Parker Vectors ($10) are a good way to see if you like fountain pens, I think.

I agree with your conclusion, but you CAN have a war on a noun. For example, Japan.

There are around 6 billion humans, so this would make the total water consumption equal to around 16 times the human population.

I started out writing this post on how this number is way too low, but it's actually more reasonable than you might think. I do think it's off by an order of magnitude (but I thought it was off by at least 5).

Assume water consumption is roughly proportional to mass. The rest is based on wiki Biomass article. From the "domestic biomass = 700 mill tonnes = 1% of earths biomass" estimate a total biomass of 7*10^10 tonnes. Say a third of that is from animals (there are a lot of small critters in the sea, so maybe), so we have around 2*10^10 tonnes animal biomass. 6 billion humans = 1*10^8 tonnes biomass, so we're talking the equivalent of 1200 billion humans, ie, around 10 times your estimate. So that would make the two quantities even closer.

The interleaving is assumed random. I believe that the deck is randomly divided into 2 stacks, and then we sequentially take cards from each of the stacks, choosing the stacks with probability in proportion to the number of cards in each. I would actually take a look at the paper if you're interested - the result is not that esoteric. What's nice about the proof is the very clever idea (something about certain sequences of cards), which makes the proof itself pretty accessible.

Great idea, memorize two three digit numbers that can give you 6 places of pi. Now all you need is a calculator to divide them ;)