Hacker News Reader

Comment by jf | Hacker News Reader

I recently learned about a similar use of animals for assisting humans:

In about 1915, he tried a more natural approach to control the street lighting in Brightwater – chicken power. They didn't have time switches in those days, so he connected a switch to the perches in his chicken house. When the chickens started to roost at night the weight would turn the lighting on and in the morning when they got down off their perches, the spring switch would turn the generator and lighting off.

Via: https://www.theprow.org.nz/people/robert-ellis/

View on HN

Some designs for nuclear bombs included live chickens which would produce enough heat to keep the electronics active:

http://news.bbc.co.uk/1/hi/uk/3588465.stm

That is one of the most weirdest (and messed up) thing I have read in a while.

And they say it is not a fools day story despite it surfaced on 1.April, but I am not convinced.

But they also don't say how serious that plan was. It was never more than a theoretical concept it seems, after all.

There was a fringe-y program in WW2 run by Skinner to try to train pigeons to act as primitive guidance computers for glide bombs. It only got minor funding and AFAIK was never tested with actual birds in actual bombs but they were able to train the birds and it seems like it could have worked.

https://en.wikipedia.org/wiki/Project_Pigeon

Let me trump that with the bat (fire)bomb!

https://en.wikipedia.org/wiki/Bat_bomb

Also, trojan rats:

> During WWII, British Special Forces considered stuffing dead rats with explosives and dispersing them over Germany. They hoped that the Germans would gather the rats and dispose of them in industrial furnaces, producing tremendous explosions that would cause catastrophic boiler failures.

https://www.theguardian.com/uk/1999/oct/27/richardnortontayl...

Which inadvertently proved it's own effectiveness when some of the buildings at the base being used for testing were burnt down because some bats flew outside the intended test range and roosted in random buildings. Pretty grizzly story and I'm glad it didn't go to wide spread adoption but that part has always been hilarious to me.

I've always figured the birds would probably lose focus once in freefall. They'd surely know they were.

It was for glide bombs which don't free fall. They're, as the name implies, gliders so fly further from the release point and don't fall ballistically. No idea if they got as far as actually strapping the guidance device onto an actual bomb though to see if they would work under stress.

Fun fact, this is the same Skinner that eventually wrote the book on modern autism treatment, aka Applied Behavior Analysis.

FYI ABA, even modified for modern times, is still heavily criticised in most autism circles.

https://en.wikipedia.org/wiki/Applied_behavior_analysis#Crit...

Yeah, I've heard that ABA is for the parents, not for the children with autism.

Birds and bombs are a popular combination:

https://www.military-history.org/feature/pigeon-guided-missi...

It's interesting how we're sorta circling back to these concepts with drones today

https://en.wikipedia.org/wiki/Anti-tank_dog

Bats too

https://en.wikipedia.org/wiki/Bat_bomb

- "Tests were used to determine how much napalm an individual bat could carry, determining that a 14 g (0.5 oz) bat could carry a payload of 15–18 g (0.53–0.63 oz)..."

Similar ideas were (perhaps apocryphally) used in the medieval ages,

https://en.wikipedia.org/wiki/Olga_of_Kiev#Drevlian_Uprising ("Olga_of_Kiev#Drevlian_Uprising")

- "Olga then instructed her army to attach a piece of sulphur bound with small pieces of cloth to each bird. At nightfall, Olga told her soldiers to set the pieces aflame and release the birds. They returned to their nests within the city, which subsequently set the city ablaze. As the Primary Chronicle tells it: "There was not a house that was not consumed, and it was impossible to extinguish the flames, because all the houses caught fire at once."[22]"

Birds, bombs, fire, and vengeful Kievans as well:

<https://en.wikipedia.org/wiki/Olga_of_Kiev#Drevlian_Uprising>

I believe it's still the case that the system where TSA swabs things and puts those swabs into a detector uses live bees in cartridges to sense explosives[1]

[1] https://www.homelandsecuritynewswire.com/new-device-uses-sni...

Sounds like bullshit. Time switches would be trivial as long as you had a clock, and clocks existed long before electricity. The idea that rigging a switch on a perch in a chicken coop would be easier or more effective than using a clock seems ridiculous.

"But I don't want to adjust/wind/sync the clock as the days get shorter and longer!"

Time switches can only roughly accommodate the changing time of dusk and dawn, and can't handle dark weather at all. Said otherwise: if you're trying to approximate a light sensor, a clock is a worse approximation than a live chicken.

Or to give a specific counterexample to "The idea that [a chicken could be] more effective than using a clock seems ridiculous": during a solar eclipse, a chicken may go in to roost, but it would have to be a pretty sophisticated (or broken) clock to handle that case. (I'm not saying that solving for the solar eclipse case is necessary or useful, just that it demonstrates the point.)

At that point, look out the window and flip the switch yourself. Then you don't have to worry about whether the clock is wound, the chicken is sick or dead, or any other anomalies

Presumably the consensus of a number of chickens could be more consistent than one person (who could get sick or forget).

Thinking about it, nearly everything about computing is fundamentally about how to translate between analog and digital.

Reply on news.ycombinator.com