robryk @robryk@qoto.org

@ruarl @grimalkina

I was nearly never taught the history of any discovery in school. When I was (e.g. nonexistence of aether or phlogiston) the whole thing was presented backward, so that often it was hard to realize that these weird hypotheses that turned out false were actually reasonable. I consider this a significant failing of the education system I've experienced and I think it's related to the problem you're pointing out (because it leaves people with less of an idea how discoveries happened).

@logicalerror that depends. It's hard to make a trike that will be stable in the kinds of turns you can easily make on a bike.

@b0rk the caret notation for parent of a commit (very useful for interactive rebase or when you want to be explicit about what you are diffing)

@0xabad1dea

An article I've found on the topic: https://nltimes.nl/2024/04/15/ns-trains-halt-brief-protest-conductor-viciously-beaten

Not that the poor gal was beaten by a _group_.

@ZachWeinersmith

Ah, and in an area where this becomes somewhat more confusing -- probability -- I found that describing everything in terms of stochastic "experiments" and then talking about random variables corresponding to some values observed during the experiment is the least confusing approach.

The area where IMO we don't have good terminology for this is partial differential equations. There, we very often talk about curves through some space and functions applies to e.g. the point the curve provides and its tangent. We sadly create a ton of confusion by not naming the arguments to that function (rather using `curve(t)` and `curve'(t)` in their place). I think this is an area where talking about (in)dependent variables is a reasonable crutch for the ambiguity that's typically introduced.

@ZachWeinersmith

I find this way of thinking about functions very confusing. The way I imagined functions is that they're a "computational" primitive that takes some inputs and gives some outputs back. This model admits natural composition, reasonably natural argmax&al and inverses. In it the notion of dependence of variables is, depending on your pov, either trivial or not really a thing.

@grimalkina

Predictive value for what kinds of questions?

@whitequark

Do you also want not to get a stream of audience comments/responses/reactions? I'd guess so, and if so this is an additional important difference.

@ProfKinyon

Thanks, this is a proof of this fact I haven't seen.

(The way I've been thinking about this is that if \sqrt{N}=a/b, then \sqrt{Nb^2}=a and multiplying a perfect square with a nonperfect square yields a nonperfect square (which can be easily seen by looking at prime decomposition of both).)

@Michael

There are places that have smallish companies operating as a food delivery intermediary (e.g. mosi.ch claims to employ 100 deliverers -- likely many of them part time). I would expect such companies to be somewhat close to the range of models you are considering, so I expect them to have stumbled into many of the issues that I wouldn't even imagine that would arise.

I would be afraid of running into some unsolvable tradeoffs between privacy of customers and ability to detect some kinds of antisocial behaviour from various parties (e.g. how would we deal with areas that no deliverer wishes to serve for whatever reason? if we consider that fine (or consider much higher delivery prices there fine), this is IMO too close to extremely laissez faire capitalism to net the benefits I think you expect. the only ways to prevent that that I can see rely on social pressure, which requires some transparency into delivery orders accepted/rejects).

@textfiles

When I look at wiki page of VCF, I can't tell how related the different editions are (in particular, how related are the European ones to the USAian ones). How much are they all organized by the same organisation?

@timorl

Have you looked at fly.io? They're doing infra that others use, so it's kinda hard to estimate value to society. Everything else seems true of them.

@LukaszOlejnik is there a description of what direct means somewhere? I don't see a very clear line that makes this direct and e.g. ensuring delivery of tap water to a military base not direct.

@LukaszOlejnik

How should I interpret direct causation? If I use my intuitive understanding, I would conclude that setting up an automated observation post does not satisfy the wording, which would surprise me.

@christmastree @fasterthanlime what about the stations that are in the zone that doesn't support standard oyster because the zone number is too large?

@grimalkina

To me (with my worldview currently skewed by working in infosec bordering systems reliability) estimates by default mean probability estimates or frequency estimates.