Hit it on the third shot though... it's hard to make it out on the fencepost at 200' with just the steel sights... eyesight ain't what it used to be.

... an algorithm was something which reliably produced results when processing the same input. NN/AI people keep using that word, "algorithm", I do not think it means what they think it means...

Professor Matthew Walker, Director of UC Berkeley’s Sleep and Neuroimaging Lab, published "Why We Sleep" last month. http://www.simonandschuster.co... I devoured the book; it's good science and remarkably well written. It covers a lot of the current work, and that work would go a long way to explaining this effect. There is so much going on when we sleep that is key to mental and physical health, it's not just "downtime", and to a degree I had not imagined. Sadly, I now realize I must consume less of my beloved coffee. Gladly, my sleep habits were already pretty good and the kids in my house have an early bedtime, no tech in the bedroom rule.

I think that Nikola Tesla might call this out as an old idea (http://teslaradio.com/pages/tesla.htm,https://www.damninteresting.com/teslas-tower-of-power/) although his designs were a little more grand. If he'd only pulled it off we'd have been wireless for a century now. :-)

Seems "AI" hasn't come much further than the chat bot we had at Waterloo about 2005. We fed it the Star Wars scripts, the Bible, the Complete Works of Shakespeare and a few scholarly articles about the Iraq war. It was occasionally funny. It was about as coherent as that script.

I'm reminded of the remark that a dancing dog is interesting because it has been taught to dance, not because it dances well.

I did my Master's at a top Canadian university in 1992. I was T.A. for the 4th-year graphics course, one of the "big three" at that school.

Each term we had about 25 students, 5 or 6 of whom were really good.

I went to industry for a decade, then decided my life was too easy.

I finished my Ph.D. at the same school in 2007. I was T.A. again for the 4th-year graphics course.

Each term we had about 50 students, 5 or 6 of whom were really good.

The "gene pool" of good, motivated students didn't change from 1992 to 2007, the profs were every bit as good both in research and teaching skill. The only "improvement" was the pool of fee-paying, half-subsidized students taking the course. The population of the province sure didn't double during the time, but the 1990's drive to get "more technical people" sure doubled the intake. The 5th year of high school was also trimmed back.

I knew faculty who quit for industry rather than teach remedial high-school math.

"Computer Science: because your Mom told you to!" was the headline of a poster seen outside the student society door... hits the nail on the head.

Corporations are legal entities comparable in law to people and these fines are little worse than traffic tickets. I like the approach of Japan's regulator to these frauds: ban the companies involved from any activity related to LIBOR based instruments. That cuts off the source of the motivation to fraud. It would kill the investment banking arm of these institutions and leave them the quiet, boring banks that we thought they were. Write your MP/Congressperson....