A Word on Statistics

I rather liked this one:

A Word on Statistics

Out of every hundred people,

those who always know better:
fifty-two.

Unsure of every step:
almost all the rest.

Ready to help,
if it doesn't take long:
forty-nine.

Always good,
because they cannot be otherwise:
four - well, maybe five.

Able to admire without envy:
eighteen.

Led to error
by youth (which passes):
sixty, plus or minus.

Those not to be messed with:
four-and-forty.

Living in constant fear
of someone or something:
seventy-seven.

Capable of happiness:
twenty-some-odd at most.


Harmless alone,
turning savage in crowds:
more than half, for sure.

Cruel
when forced by circumstances:
it's better not to know,
not even approximately.

Wise in hindsight:
not many more
than wise in foresight.

Getting nothing out of life except things:
thirty
(though I would like to be wrong).

Balled up in pain
and without a flashlight in the dark:
eighty-three, sooner or later.

Those who are just:
quite a few, thirty-five.

But if it takes effort to understand:
three.

Worthy of empathy:
ninety-nine.

Mortal:
one hundred out of one hundred -
a figure that has never varied yet.

- Wislawa Szymborska, translated by Joanna Trzeciak

+1 Mythos Knowledge

Also, forgot to mention: my Raven Guard gained his first Insanity point. They grow up so fast!

(turns out making psychic contact with the Tyranid hive-mind is a Bad Thing. Who knew?)

Back to school

At the end of last year I did the Stanford online course on machine learning; this year I'm doing algorithm design and game theory. (I was signed up for one on information theory as well, but it got cancelled, probably good for my workload.)

Aside: My boss^4 has a story that he tells to illustrate his view on theoretical seminar presentations. Old monk, young monk; old monk gives young monk a sieve and tells him to fetch water with it. Young monk spends all day trying to pick up water in the sieve without success, and returns saying "Why did you give me this job? I could never hope to bring back water this way."

"Ah, but look how clean your sieve is!"

Frank's view is that some of these seminars are there to clean our sieves: nobody is expected to understand the content, but they're meant to sharpen our game by showing us what theoretical rigour looks like and remind us that we could be doing better.

I was a little comforted to hear this - I get hit by self-doubt when I can't keep up with a lecturer - but I'm not sold on the approach. To me, the point of a lecture is to let other people understand what it is that you're talking about.

So one of the things I've been really enjoying about these courses is that they're pitched at about the right level. I could probably cope with something 20-30% more demanding, but as it is, I can multi-task while skimming the lectures and I can get warm fuzzies by helping out those of the students who aren't having as easy a time of it.

The ML material is likely to be very relevant for my work; the other stuff not so much, but still Relevant To My Interests.

So, yeah, if you're interested in geeky subjects, the free online stuff at Stanford is worth a look.