'Publish or perish' is a very real thing. Sometimes scientists know their results aren't amazing, but are desperate to publish something because they've just spent a year's worth of grant money on it.

That pretty much works in Starcraft, too

Yeah, you make a good point, this study didn't even claim find any correlation between age and skill-level; the headline is wrong. They claim to have found a correlation between click speed (reaction after the screen changes) and age, but even that claim is tenuous.

Even then it's hard. For example, one pro-gamer hasn't been winning as much as his prime, but he's said he hasn't been practicing as hard (his teammate just won the championship in Korea, and he definitely practiced hard).

Of course, we can't necessarily trust his self-assessment, but it shows that even people who have a financial interest can get burned out and lose interest in the game.

Yeah, I remember the good old days, when there weren't many demands on developers, and release schedules were easy, and if you said to your boss, "can I have more time?" He said, "Sure, no problem my good friend, have a raise too!"

Look at the graph. Data often clarifies things.

How would you improve the interface? I'm not sure there's any way of handling 60 different units easily....

They calculated the mean time between switching to a new screen and then clicking on something on that screen.

Here is the data they collected. Look at it and see if you can figure out where it peaks. What are the things that strike you most about that data? The primary correlation is between skill-level and mean time, if age matters at all it is a far weaker variable.

Looking at the actual data, I would say they've found the age when people stop playing Starcraft; it's a fairly sharp drop-off. And the change in mean-switching-time is not a real effect, merely an artifact of the suddenly smaller data they have around that age. This paper is probably relevant (suggesting scientists often need to improve their statistics).

Furthermore, if you read the actual paper, you have this quote: "A second analysis of dual-task performance finds no evidence of a corresponding age-related decline." So I'm going to say there's not a story here.

Oh yeah, that's exactly what I look for in a job.

Python guru Jeff Knupp should go find a job where he can program, and not worry about ops. Simple solution.

That's a good point

This is exactly why I refuse to give them any of my money. I do it by hand, every year. Bonus I get to annoy IRS agents by making them read hand-written forms.

As long as you're getting a refund, there's no problem. If you made some mathematical mistakes, then the IRS will correct them for you. If you later realize you didn't take a deduction, then you can file a correction.

The real problem comes when you owe money, and you didn't pay it in time. That's when the IRS starts charging 1% interest per day (up to ~30% or something). Also, if you claim tax credits that you don't qualify for, you could have to pay them back.

This is why I do it by hand every year. To spite the IRS, to avoid sending cash to Intuit or some other heavy-lobbying company, and also it helps me understand taxes better.

That's cool, but the problem in SF is that there's just not enough housing. We can set price controls, but there will still be the problem of not enough housing.