robryk @robryk@qoto.org
I enjoy things around information theory (and data compression), complexity theory (and cryptography), read hard scifi, currently work on weird ML (we'll see how it goes), am somewhat literal minded and have approximate knowledge of random things. I like when statements have truth values, and when things can be described simply (which is not exactly the same as shortly) and yet have interesting properties.
I live in the largest city of Switzerland (and yet have cow and sheep pastures and a swimmable lake within a few hundred meters of my place :)). I speak Polish, English, German, and can understand simple Swiss German and French.
If in doubt, please err on the side of being direct with me. I very much appreciate when people tell me that I'm being inaccurate. I think that satisfying people's curiosity is the most important thing I could be doing (and usually enjoy doing it). I am normally terse in my writing and would appreciate requests to verbosify.
I appreciate it if my grammar or style is corrected (in any of the languages I use here).
Joined Nov 2020
@kravietz Does is really use a reverse-driven screw pump to generate energy? (I've never seen such an approach.)
@eoaiuastwg Example 2 (there's something weird going on with the font for the comments, but the code one seems monospace)
But what's the score?
How would you define that for things that you don't normally make at home? (I'd guess cheese/tofu could be one example.)
Do you have examples of solutions that are hardest to come up with using just the understanding how things work? I wonder if these are mostly things where Yet Another tool is helpful, where a tool has a somewhat hidden feature/option, or something else.
@b0rk BTW I think the parent post didn't federate to your instance (not sure why; length? happenstance? some problems caused by your question having been edited?) and I think no one else mentioned the /proc/N/root trap.
Absence of evidence is (usually weak) evidence of absence though. It's not the case only in situations where we're not looking for evidence in the empirical sense, but in a mathematical proof sense.
Some of these problems were actually useful for me, because they caused me to realise I misunderstand something (e.g. the behaviour of current directories that contain symlinks in their path, or how the commandline is parsed (see the "how do I remove a file with a name that start with a dash" problem), ...). For that to work, one needs to have resources to learn what's going on that can be used in the time one has available to spare (so it works way better for same resources for tinkering teenagers than someone just trying to use the system to get this particular thing done today).
Does putting them in the fridge significantly lengthen the time they remain edible? (I remember an onion expiring in a _different_ way when cooled and haven't actually experimented, so am unsure.)
PS. I hope/wish you have an uncomplicated (in any way) recovery.
"Fake" symlinks in /proc. For example, /proc/PID/root/some/path is not necessarily the same as $(readlink /proc/PID/root)/some/path (there might be no other valid path that points at /proc/PID/root, when you have mount namespaces and e.g. yours have that directory shadowed by a mount).
`mkdir -p` failing with `No such file or directory` due to a dangling symlink in the requested path.
Potential effects of `cd .`: filesystem gets asked to resolve the path again, because shell actually does sth like `chdir(getenv("PWD"))` in that case.
Related: when you move a directory that's above some process's working directory, that process's working directory moves, but if it keeps track of its working directory path (like shell does) that will become desynced.
Writing to a mountpoint before the filesystem gets mounted, ending up writing "behind" the mountpoint, and being very confused where the files ended up (in rootfs in the mount directory, but its contents is normally invisible when something is mounted on top).
Don't they sometimes mean "but it (or its future replacement) is not guaranteed to continue to have properties we'd rely on"?
Shell builtins vs identically-named commands in PATH.
Ending up with screen-inside-screen and having to figure out how to detach the inner one.
Having a program leave the terminal in some weird state when killed (either without echo, or in that weird alternate charset mode that can be fixed by _outputting_ ^O).
Not having job control at all (i.e. can't kill or suspend anything that I start in foreground) in some kinds of emergency shells (probably due to something like http://lists.busybox.net/pipermail/busybox/2002-June/006397.html).
Problems with terminals that are misconfigured so that control sequences are misinterpreted (most recently I had the issue when sshing to something that didn't understand `TERM=rxvt-unicode`).
Heh. This reminded me of the annoyance with the Polish signaler simulator (www.isdr.pl), where you can't really reroute a train. The only way to get it to do something other than in its schedule is IIRC to use shunting instructions to start the deviation.
Where would you put the line that distinguishes this kind of work and work of e.g. a water pump (which I assume you don't object to on similar grounds)? On "can be commanded using nontrivial expressions of natural language" maybe?
TBF Factorio's signaling automation encompasses both signaling and decision making that's normally done by the signaler and/or dispatcher.
Actually, @gregeganSF in case he finds this topic interesting. tl;dr how could a world look like if "legal positions" were quantized to a grid at a scale that's macroscopic (or nearly macroscopic) to its inhabitants.
Hm~ I then don't understand what level of self-consistency you want (nit: it's not a biological detail, because anything that processes information will have such problems). Another thing to think of in such a world is the question of straight lines: are all directions permissible in contexts where something spreads along a line? What's the distance in e.g. electrostatic attraction, or what are the directions a light ray can travel in (I vaguely remember that restricted wave propagation directions is something we do encounter in our world in some crystal structures btw.)?
If you haven't already seen anything by Greg Egan, you might enjoy his novels and stories. Many of them posit worlds with different physics (and two even with, after a fashion, different mathematics) or different ~biology (with the source of the difference unspecified) and try to extrapolate into them. Depending on what is the difference in a given story, the extrapolation is either very detailed (e.g. Clockwork Rocket), very vague (e.g. Teranesia), or somewhere inbetween (e.g. Morphotropic).
I enjoy things around information theory (and data compression), complexity theory (and cryptography), read hard scifi, currently work on weird ML (we'll see how it goes), am somewhat literal minded and have approximate knowledge of random things. I like when statements have truth values, and when things can be described simply (which is not exactly the same as shortly) and yet have interesting properties.
I live in the largest city of Switzerland (and yet have cow and sheep pastures and a swimmable lake within a few hundred meters of my place :)). I speak Polish, English, German, and can understand simple Swiss German and French.
If in doubt, please err on the side of being direct with me. I very much appreciate when people tell me that I'm being inaccurate. I think that satisfying people's curiosity is the most important thing I could be doing (and usually enjoy doing it). I am normally terse in my writing and would appreciate requests to verbosify.
I appreciate it if my grammar or style is corrected (in any of the languages I use here).
Joined Nov 2020
