Both mean and median can be misleading here. Mode may be somewhat more useful, but could be flawed as well (while probably somewhere in the 70's for quite some time, it's very possible that there have been times when the modal life expectancy was 1 year, which misses just as much, if not more, important information about the statistic.)

I guess what I'm trying to say is that by boiling down a deep rich quantity of information to a single number, be it mean, median, or mode, you lose a lot of the usefulness that existed in the source data, and that leads to misunderstandings of what it means. So you get people baffled by why average life expectancy keeps going up but people aren't living to 150, because they're used to averages being used with nice normal distributions and don't realize that we've been thinking of the children for so long that they hardly ever die these days. Not that there's anything wrong with that, but it doesn't help their great grandparents live much longer.

I'd say the answer here is fairly simple, we haven't put much effort into keeping 100+ year olds alive, relative to the amount of effort to keep, for instance, 5 year olds alive. As I understand it, a huge amount of the gains in average life length have come from squeezing the bottom of the graph, not extending the top of it. Here's an interesting, though somewhat morbid, exercise. Go to a very old graveyard and look at the stones on the family plots. You'll often see a family with 12 children, half of whom died in childhood, and the other half lived to their 90's. So in that family the average life length was around 50, but that doesn't mean that a 50 year old should be looking for the grim reaper around the corner, quite the opposite in fact. As I understand it, the life expectancy of a 25-year old has been fairly stable for a fairly long time. Once you've survived the fragility of youth and the stupidity of adolescence, the following decades are a cake-walk, morbidity-wise.

Well, I wouldn't say that it means absolutely nothing... to me it meant it was a fantastic day to take my boys on a four mile hike through some beautiful mountains. Black Friday may possibly have meant that there were fewer people on the trails, but that difference, if extant, was probably the difference between seeing 10 people instead of 11 on the trip.

Does it worry anyone else that this satellite shares a name with a unit of ionizing radiation?

I mean, sure, probably the sat was named after WIlhelm just like the unit, but this is definitely a case where absorbing just one roentgen would almost certainly be fatal.

Assuming that the cure/infestation could spread, then sure, it's a serious threat, and I wouldn't want to take it lightly. Even a single zombie should be taken seriously (/wave Clairvius). But if you take the worst case scenario that a few million cancer patients suddenly become infectious zombies, then yes, it's a serious threat, a huge disaster, and the premise for an interesting sci-fi story.

On the other hand, if you take the best case scenario of a non-transferable zombification resulting as a side effect from a preventative measure, then you're likely to have a handful of the world's poor, remote, and those with a religious or cultural bias against modern medicine all engulfed in a sea of billions of the walking dead. That's converting your story/movie from a sci-fi horror disaster movie into a post-apocalyptic survival movie.

A subtle distinction between this and the whatever-it-was in the I Am Legend movie is that the drug in the movie was a preventative measure, while this is a curative measure. It's important because preventative measures need to be applied to everyone who might ever get cancer (which would be everyone, unless it's possible to predict cancer, which would be its own major breakthrough), while curative measures are applied to those who already have cancer.

From a societal standpoint, if everyone who has cancer suddenly turns into brains-craving zombies, that's a relatively small fraction of the population and it's a threat that can be dealt with. If, on the other hand, everyone who doesn't want to get cancer turns into a zombie, that's pretty much it for the human race.

From a personal standpoint, a 1E-9% chance of turning into a zombie and being detonated by a crack government anti-zombie squad is fairly easily offset by an alternative of near-certain death by cancer. It's at least a little bit harder to swallow if, like most of the US, your chances of getting cancer and dying of it are more like 25%.

Either that, or it's an ADD-less app that isn't easily distracted.

Or maybe it *does* add, but it adds less.

Ya know, that was my first thought too. I'm right now going through the heaviest effects I expect to see from Irene (which is to say, occasional moderate winds), but my generator is fueled up and tested. Not because of "irene", or even "hurricane season", but because it always is, in preparation for whatever happens.

Hurricanes are interesting beasts, as they're a disaster you can know about days before they hit. So many other types give much less warning. Makes me happy to be living in a time where there are weather satellites.

Another thing that I would love to see is this: take the time that the yellow lasts, and multiply it by the speed limit to get a distance. Then put a marker on the road that distance back from the end of the intersection. That way, if you're going the speed limit and you've passed that mark when the light turns yellow, you know you can just keep going. And if you haven't passed the mark, you know you need to stop. There are too many places where topography and different sized and timed street lights make it hard to judge if you should stop or go.

here's my idea for a solution to this: Make a home movie. Play it on the TV. Sue the TV manufacturer for violating your copyright to your home movie by creating and distributing a derivative work. Profit.