keenerd

I would love to merge that into Arch's wtf, but it is just a collection of scanned images. No text.

I maintain my own set of modern internet acronyms for it. Speaking of which, I need to merge a pull request for that.

And how much horsepower do you need to shell into that cluster?

I'm one of those people.

> not possible to make today with with unlimited budgets

The Lofstrom Loop is possible today for a few billion dollars. Though "catching" an incoming capsule with a Loop is going to take a while to human-certify.

On the other hand... if you wanted to quickly identify incompetent individuals, give them rope and let them cause damage.

Ignoring the BS about coriolis, I think there is something legit here.

Of course there isn't any power to be gained from the vortex. What you would gain is a lot of RPMs from very little head. It makes the system be high speed and low torque. The dynamos would be higher voltage (lower amperage), the wiring and inverters are cheaper with lower amperage, and the structure doesn't have to be as strong since the torques are lower. It might even automatically safe during floods, since too much water entering it would distrupt the vortex and halt the wheel.

It is more like a clever hydraulic gearbox that potentially makes all the other components less expensive. Or am I completely off base? The water is spinning around faster than it would be flowing out, correct?

What is the net weight of a package? (And why isn't it on the packaging?)

Areas of low contrast become whiter and areas of high contrast become more saturated.

It is also more robust than k-means. The author's algo will only work on scanned images. Photographed pages from a book will often have a slight shadow on half the page from the curvature. Blur-divide will clean this up. K-means will think you've used a lot of gray and not figure out that there are multiple background colors.

A simpler way of achieving the same thing is to duplicate the layer, blur the top layer heavily, and then set it to "divide".

It was faster to do this than reading the content the way it was intended. Because they were using random selection you'd have to click 370 times to have a 90% chance of seeing all 59. It is nearly impossible (like 7 nines impossible) to see all of them in under 120 clicks.

How did we both do this at exactly the same minute? You botched up something with the encoding though.