Lots of statistical skulduggery around, regarding the threshold income for 'poverty' and the resulting trend. But we can sidestep binary gerrymandering entirely by plotting the distribution over time. Max Roser:


I am not a Swedish corporate lawyer but:

Does IKEA own itself?

There are some human beneficiaries, so it's not the full-blown Gibsonian nightmare... (Beneficiaries besides the customers and employees I mean.)


Goodreads has a nice culture. Shame about the owner, but what you gonna do?

Saw this t-shirt on the way to work today. I'm not even mad, just confused

Bold new direction from O2, targeting the lucrative pro-drowning demographic.

Schools, philosophies, factions:

* Bayesians
* Frequentists^
* Likelihoodists

To what extent is ML a fourth school, as opposed to a mere toolkit, shut up and descend the gradient?

(Image: Greg Ganderberger)

Most intellectuals focus on what's wrong - which is good - but then conclude that the world is gettting worse - which is mostly wrong. Pinker & Roser are odd because they push back, & with dramatic data.

One thing even they neglect is the sheer intellectual wealth of our time. It's not visible unless you look; hidden in Springer yellow books, in unprecedented huge open datasets, in the vast public domain of Gutenberg and Github. Another source of joy and hope. Never a better time to think.

Can any Dutch people explain what's lost in translation with "biological" here?:

Discussing our taxonomy of Markov decision processes, my friend noted that some philosophers (the finitists) don't believe in infinities; maybe we should leave it out.

I noted in response that some philosophers (Eleatics, Zuseans) don't believe in continuousness; many don't believe in randomness (determinists - these days superdeterminists); & that, come to think of it, some (e.g. Parmenideans) don't believe in moving things.

So to be fully uncontroversial, we'd have to prune the whole tree.

Show more

A Mastodon instance for maths people. The kind of people who make \(\pi z^2 \times a\) jokes.

Use \( and \) for inline LaTeX, and \[ and \] for display mode.