Well, auto-remove feature looks like a bit too much for me, especially with no clear write-up on rules it follows - getting unexpected automated false positive bans often sucks a lot emotionally; good thing you've announced it at least.

I have an opinion (just a personal vision on the big picture, I have no desire to instill it in others) that whenever we start needing automated tools like these in Fediverse, it's a sign that decentralization concept starts failing locally and it's time to federate more: make more servers with smaller governance.

It's nesting season now. Keeping cats outside pretty much means easy and fun hunting on fledgelings.

Now, altruistic feelings aside, in addition to somewhat endangered tits and not really endangered sparrows, there are birds that are totally lethal for cats. Some of those, like owls, are often really hard to spot for an untrained human, for they are very fast and make no noise. Others, like crows and seagulls, are just very fast and devious. And they would totally try to do anything in their power to prevent the hunting for their babies, from instilling fear, to physical damage of various severity (that they could also inflict onto humans, I guess rarely lethal in Europe, but eye traumas happen; I know of conflicts between humans and crows where government had to resort to inviting professional military sniper to settle it. Hunters know not to mess with owls lest they are fine losing their face, literally.).

So this is just another theory that does not require cat predators and quite fits the setup described.

My neighbors keep cats outside; those look scared to death most of time this season - for a good reason, there is an extremely hungry owl family learning to fly here (that only I and people on Fediverse who saw my shitty smartphone pictures know about). Those are an addition to everpresent lynxes, a bear, and at least one wolverine, yet it's never this scary for them as now, it seems.

No it's not, I'm saying kinda the opposite.

Yeah, I just compare every lowtech to it, it's like a baseline anyone can make. Anything simpler is awesome, anything harder is useless usually.

A Stirling engine is symmetric device that could work both ways.

But I admit, after staring at this for a bit longer, it doesn't even really need any kind of special or even similar magnets. It is simpler indeed.

nice, but seems more complicated than Stirling, unless you have no pistons and lots of copper wire magnets. (edit: distracted writing, I'm watching it lol)

Yeah, like, let's improve the super reliable stuff that works perfectly with gambleware!

Also, this irresponsible behavior is now heresy or something? Like non-opensource code, too.

One of the most prominent improvements in Opus 4.8 is its honesty, ruthless efficiency and an almost fanatical devotion to the Pope

ah it's this way.

ok, just consider buying bare sensors and attaching them to some MCU boards you like - with good connectivity and all - some distance away from the boards (working MCU alters readings significantly unless you bump it into ultra low power, which makers of these simple adafruuit probes often do not care about). It's way cheaper and easier to integrate in most cases, also you get to choose sensors just for the task, often you can get insane performance (like, glass-clad Pt thermal probes are often surprisingly cheap, for example)

This task needs more specifications; humidity is such a wide spectrum.