#sensor basics

[kaiserouo]

(prev | next | first)

nom

[twentyfivemiceinatrenchcoat]

hellooooo ari!

question for you!! how long do you think suguru’s s/o could keep a big secret from him? or maybe a better question is how long do you think his s/o would last before he sniffs it out?

HELLOOOOO ANON …… hm hm hm . well he’s just very observant LOL you would have to be very good at lying to hide just about Anything from him bc his bf sensor works overtime . he Will notice a nervous tick, a glance away, a purse of your lips and while he won’t be confronting you immediately he stores it away in a little mind cabinet and waits for you to bring it up yourself; if you don’t, and if it’s visibly bothering you, that’s when he’ll take action …… but he gives you a chance to Be Good LMAO. in general he doesn’t like you keeping secrets from him, not because he doesn’t trust you but because it makes him feel as if you can’t trust him — and that bothers him a lot more than what he shows at the surface!!

[lucabyte]

hey do i have any followers who are seasoned xenofiction fans . im trying to figure out an animal cast for something and want to know what the vibes are for "wouldn't raise an eyebrow in a setting" vs "would raise some questions". Like, as a speaking role having creature. So i don't mean a cast entirely comprised of just that species, I mean alongside other animals.

this is entirely vibes based i'm just in like. a decision-making pre-production stage rn

(no need to rb by the way i dont need a very big sample size for this)

[sentientstump]

Happy New Year stump! o/

Got any fun plans for the new year?

i genuinely thought about this for the first few days of this new year and couldn't come up with anything that fun? 😭

gonna live, gonna enjoy small moments, big moments if they happen, something like that👍🏼

[malwarechips]

tip for bug videos on youtube never fucking look in the comments you will get pissed

[bobblestheninja]

Plants have a rudimentary system for stimulus and response, as well as flows of nutrients and glucose and biological energy.

You don't need a brain, you just need something that will translate those signals. It's the same as making a robot arm that responds to pushing buttons, or to changes in water flow, or certain electrical signals.

It's a "brain" in the same way as a plant has a "brain" which does things like grows roots to seek out water and anchor itself, or turns its leaves to face the sun. It's not conscious thought, it's signals and chemicals.

Think of it more like an infant that's just learning to scoot and roll over being put on a Dance Dance Revolution pad. No they won't be controlling what comes up on the screen, or intentionally pressing buttons, but when they roll on certain bits the button will send a signal just the same as if they were actively playing the game.

Heck, you could make a plant a "control" system for a robot just by using a panel of light sensors. When the leaves turn towards the light they cover and uncover different bits of the panel and that can be translated to do things like "make a little car that scoots across the floor to try and center the shadow in the middle, to try and make sure the plant is getting the ideal amount of sunlight"

We've developed enough sensors for chemical changes, light changes, pressure/weight changes, moisture changes, and a billion other tiny things to be able to use that (and coding of course) to make robots that react to tiny environmental changes from the plant or the surrounding environment itself in order to do random things.

Living plants have more than just shadows and environmental changes. They also have ion currents, and water currents to transport nutrients and water and that, and something called axin, which can tell one part of the plant to grow more than the other side and make leaves turn.

You can use lemons to make a (really weak and shitty) battery because of things like ion concentrations and that.

So basically there's enough chemical signals in a plant to make something that can be read. Honestly you could do the same thing with a fish tank. Add some motion sensors in various places and a small school of fish. Make motion in front of different sensors correspond to different movements/different bits of the arm activating, and you can get a robot arm that's controlled by fish swimming.

This doesn't mean it's halo-esque integration of a neural net, or high level science translating a plant's "thoughts" to actions, it's just that plants do "react" to stimulus in order to grow, move petals, emerge from seeds, and turn towards light, and our technology is at a point where we can tap into random chemical changes and make things happen.

Also if you look at the article you can see that they're not even going the extremely fancy way I'm suggesting, they're using pretty basic tech. Basically everything that has ions moving makes background electrical "noise," so what they've got is electrical sensors (literally the EKG pads they use for electrocardiograms, I should know, I've had to have EKGs before, and those are the exact stickers) on the leaves, and that's translating these random tiny electric potentials/currents into signals for the robot arm.

The plant (though it will be producing signals in response to the patches stuck to its leaves and blocking photosynthesis in those areas) has no idea that that it's connected to a robot arm, or that there's a machete, and it's not actively controlling anything. It's just being read by the systems in the arm and that's being translated to signals for motion.

Interestingly enough you could probably do the same thing with sound sensors in the right conditions, those conditions being with forced rhubarb. Set up some microphones around the perimeter, and one in the middle, then use the direction from which the sound comes from as the basis of your movement.

You could probably use this setup to "teach" rhubarb to "draw" by setting up different pitches as different colours, and different volumes/distances from the centre microphone as different amounts of movement. But it's really just getting a computer to respond to a specific stimulus by doing a specific action.

There's also a certain amount of error to analysis, and the more sensors you have the more chances of error, which can make results come from "nothing" or very little actual input. One doesn't have to look further than the dead salmon FMRI study for evidence of that. If you don't filter your data correctly you can determine that dead salmon have neural activity in response to pictures of humans... which...they don't.

Anyway thank you for coming to my unhinged 8AM TED talk on plants, sensors, computer input, errors, and dead salmon.

dark science beyond belief

[meme-loving-stuck]

I ordered a Suri toothbrush because they seem to be a certified B-corp now which is also at least 2 years old and the reviews seem truly mixed. I am so excited about having a fancy electric toothbrush WITH the mail back program because honestly even if the brush heads arent biodegradable i like the idea of not trashing the entire toothbrush after three months

[theskyexists]

The thing is, people are like 'use ai! Use chatgpt!' and last night I was like ok, I'll do one very specific prompt to see what it says. And it gave me nothing as in, it did not answer my question and I think it didn't even give me correct information

[emotionalsupportdman]

me: yeah my car’s been acting up and making this noise and i’m pretty sure it’s this thing

my dad: huh. sounds like it’s actually this [ten times more expensive] thing

me: ……….

me: [redacted]

[wyrmzone]

Recommended reading:

this sucks so bad i need to (remembers suicide jokes are unhealthy) spread the skulk