Yep, I did this math with a toxicologist roommate many years ago. We concluded it was very hard to do with brewed coffee. But if you're eating chocolate-covered espresso beans, you might have a shot at it! It's only like a pound of beans you have to eat! It would be a totally suck way to go, though.

Eh. Better than the table, but I don't love it. I'd put in a vote for a small-multiples graph instead of overlays. I'd also suggest replacing the lines with two boxes, one below the line and one above. (That is, fill the lines you've already drawn.) That would lead to an interpretation where the area of each box would be the total increase or decrease from each quantile.

Hi Kaiser. Might the discrepancy be people vs. households? There are 300 million people in the US, but more like, what, 100 million households? Most households share a subscription (although my wife and I do not!)....

Very cool. I wonder if you could use a mild-flavored apple as a natural source of pectin in something like this? I recently tried to make a carrot leather, using some apple as a binder. (5:1 ratio, microwaved to cook, then food-milled, sweetened with simple syrup, a bit of salt and nutmeg.) Sadly/luckily, I planned to ghetto dehydrate it on a silpat in the oven for a few hours at 170, then let the oven cool on its own overnight, which has worked well for other leathers in the past. But this time, I started too late in the evening, set an oven timer for 2am, and completely slept through that alarm, giving me carrot crackers instead of leather in the morning. Had to reorganize my dessert as a result. That said, as with your honeydew paper, it had a really great texture and flavor, and is something I'm going to do on purpose in the future! Maybe a bit more savory -- would go really well with pork... Incidentally, I love the bubble-wrap look in your paper. Very playful. What's that from?

Direct-to-consumer SNP tests such as 23AndMe provide increased/decreased risk estimates as just that, posterior probabilities of risk given the customer's DNA. It's not correct to say 'any test, including these DNA test kits, can be tuned to allow more of the false positive errors, or more of the false negative errors. ' That's only true if the tests were saying "you have cancer" or "you won't get Alzheimer's." That is, if the risk estimates were being collapsed to binary predictions. But they're not saying that. This isn't necessarily an argument that these tests should be sold to the public, or that the Times reported the story well. But your criticism is misleading in this particular respect.

Interesting. I think it would work on me, if I were in the market for a Renault! The reason why is that I don't really care about $50 one way or the other. That rebate makes no difference to my standard of living at all, it's just rounding error in the 5-10 years I'll own the car. But 11,000 Euro is a lot of money, and winning would noticeably improve my quality of life! This is an argument why, if you have to buy lottery tickets, you should buy the ones where the jackpot is millions, not just a few dollars or even a few hundred dollars. The expected return of buying a weekly lottery ticket is -$52/year, but that's also the most you can lose, while the best-case upside is unlikely, but quite large! Winning would seriously change your life! (And if you're responsible with your money, it might even change it for the better!)

Oh come on. I actually really like this one! Many of the issues you raised have been addressed by others. Regarding the ordering within a column, it looks quasi-geographical to me. Asia at the bottom, then the US, then European powers. I think your only actually valid point is that the legend should be vertical and should align with the country data.