this post was submitted on 25 Dec 2025
540 points (96.9% liked)

Programmer Humor

28043 readers
994 users here now

Welcome to Programmer Humor!

This is a place where you can post jokes, memes, humor, etc. related to programming!

For sharing awful code theres also Programming Horror.

Rules

founded 2 years ago
MODERATORS
 

Post:

You have three switches in one room and a single light bulb in another room. You are allowed to visit the room with the light bulb only once. How do you figure out which switch controls the bulb? Write your answer in the comments before looking at other answers.


Comment:

If this were an interview question, the correct response would be "Do you have any relevant questions for me? Because have a long list of things that more deserving of my precious time than to think about this!

you are viewing a single comment's thread
view the rest of the comments
[–] sga@piefed.social 14 points 2 days ago (1 children)

After reading the incandescent bulb solution, and problems regarding touching the bulb, i would switch first switch on for a appreciably long time, such that bulb has hit maximum luminousity (they heat up as they run, the hotter they get, the brighter they are), then turn switch off, and turn second switch on and quicky run to other room. we are trying to observe change in luminousity as time elapses. if it reduces, it was first (we ran it for a long time, there would be some residual glow, from my irl observations from when i was small suggest roughly 1 min period where i can still tell, but bulb wattage, contrat with background and distance matter). if increasing or max luminous, then second, if nothing then third.

but it was a stupid question. my naive guess was it can not be done, because with just 1 binary observation, you can not tell from 3 switches (you need atleast 2, which the solution assumes as temp and light state, i substitute heat with light state in transition). but still stupid. my natural assumption was leds, even when i head incandascent bulbs in my house somewhere for nearly half of my life. it is also stupid, because when you allow me to do something i was mentioned in question to do, i could just bend my way to do anything. like punch/drill through wall, or hack surveillance systems, or just pull out my handy multimeter that i always have on me, open switch box and see which switch is live, which is dead, or see voltage/current/wattage change across the loop, or measure resistance and guess what thing is there, or like blackmail the interviewer to extract the answer.

[–] Pulptastic@midwest.social 6 points 1 day ago (1 children)

You’re almost there. Turn the first one on for a while. Then then it off and turn the second one on and run to the room. There are theee possible scenarios. If the bulb is on, switch 2 controls it. If the bulb is hot but off switch 1 controls it. If the bulb is cold and off, switch 3 controls it.

[–] sga@piefed.social 5 points 1 day ago (1 children)

i know this solution, but many problems are there, what if it is in a ceiling fixture, or enclosed in something such that i can not know the temp (hot or cold), then i can still observe luminosity changes. if you feel your eyes do a bad job, get a camera properly color and temp caliberated, and just focus on filament (now auto exposure or temperature adjustment).

tl:dr i am still trying to poke holes in this thermocol wall of defence.

[–] xthexder@l.sw0.com 2 points 1 day ago* (last edited 1 day ago) (1 children)

If it's an LED or flourencent bulb they usually have a small amount of glow after turning them off from the phosphor coating. You might be able to catch that instead of the residual heat, but generally it dissipates pretty quickly, and it might be hard to see with one of the other lights on.

[–] Tlaloc_Temporal@lemmy.ca 1 points 1 day ago (1 children)

Worse, if the LED is wired to the hot side it will just barely glow at all times.

[–] xthexder@l.sw0.com 1 points 1 day ago (1 children)

I think your house might be wired wrong if this is happening... The only thing I can think of is maybe if you've got some smart switch and no neutral, so the wifi in the switch has to power itself by leaking current through the light, which is a pretty unusual setup. I don't see how this could ever happen on a regular dumb switch.

[–] Tlaloc_Temporal@lemmy.ca 3 points 1 day ago (1 children)

LEDs are so efficient that even microamps can power them. If your LED driver is cheap, it'll run on basically nothing, or charge up enough to start for a fraction of a second.

The microamps come from a hot wire running next to a switched wire behaving as a capacitor when carrying AC voltage, letting microamps leak through. (It's not required that the light is on the hot side of the switch as I said previously, my bad).

This can happen if the switch box is a terminal box with hot and switched wires in the same cable, which is rather common. Probably some other configurations too.

[–] xthexder@l.sw0.com 1 points 1 day ago (1 children)

Well, I can't say I've ever seen it happen, but I could see how it could happen in certain scenarios, especially if the LED has some weird driver in it. Maybe the capacitors in the driver would be allowed to charge up in some designs before getting dissipated through the LED in a flash?
The simplest form of LED light (just a rectifier and a bunch of LEDs in series for a 120V diode drop), idk if you'd ever see any glow or flashes, since LEDs don't turn on until a certain voltage, and if you're getting like 50V on an open circuit that seems to me like you've accidentally built a transformer in your walls.

[–] Tlaloc_Temporal@lemmy.ca 2 points 17 hours ago (1 children)

A very low current transformer, more of kess yeah.

Some lights will charge op and flicker, others have a constant glow. The speed/brightness depends on how long the wire is, so most residential lamps are unnoticeable even when it happens, but large rooms and weird wiring can make it more obvious.

[–] xthexder@l.sw0.com 2 points 17 hours ago

I guess that makes sense it'd happen more in big buildings. The runs in most houses wouldn't be long enough to have a noticeable induced current without the electrician adding a few extra loops for fun :)

Thanks for humoring my skepticism, it's been interesting to think about how this would happen.