Topic: [Bounty] Update: 100BTC for 2 Working Crow Vending Machines (Read 3362 times)
The stages could be automatically set based upon the weight of the hopper (peanuts removed) and the frequency of coin drops.
I think it would have to necessarily require user intervention, since it would be different timing for any given set of birds.
I just want a working unit, I'll handle the bird training 
Well I guess what he is pointing out is that the machine is supposed to handle the training, and that part of it might be annoying to create. However, an easy one would be to just make the machine work on step 4.Or maybe the machine could handle all 4 modes, but would require manual intervention to switch between modes?
My initial brainstorm would require user intervention to switch between modes since we don't really have an idea of how long each stage needs to operate for proper training. I suppose the stage timing could be automated (with adjustable time periods)
Interesting, you guys really are serious. I'm interested in building this, but I'm don't have the time/not interested in trying to train the black bastards to actually use it. If you guys just want a working unit as far as the mechanics and electronics go I could start piecing it together and let you know if it's feasible (cost wise).
Because there are some intricacies, like whether the sound of the machine scares the birds, whether the coin slot can avoid clogs from being outside for a period, etc., I'd like a tested machine. If you build one for Inaba and he can train them, I'll send him the 20BTC for the video, and buy another from you for 40 (assuming Josh Klein doesn't get back to me first).
If I end up buying directly from him, I'll still honor half my video bounty, of 10BTC for anyone else who can demonstrate a working crow vending machine with a bitcointalk sign in the video.
One day, I hope to see a world where man and crow engage in mutually beneficial commerce in hundreds of cities around the world. :-)
I guess my point is if a bunch of squirrels eat all the nuts before you get a crow, experiment over. Either that or constant babysitting.
Squirrels will only be a problem for the first few stages (where you leave food by the machine to get them to come over to it), but Josh found that the squirrels weren't able to learn to deposit coins. So presumably, once you get the first few crows trained, squirrels will no longer be a problem.
I notice that Josh says a kit might be available soon: http://www.josh.is/crow-machine/#comment-1544
Awesome! Just emailed him about the kit. I'll keep this thread posted.
I notice that Josh says a kit might be available soon: http://www.josh.is/crow-machine/#comment-1544
I just want a working unit, I'll handle the bird training
Well I guess what he is pointing out is that the machine is supposed to handle the training, and that part of it might be annoying to create. However, an easy one would be to just make the machine work on step 4.Or maybe the machine could handle all 4 modes, but would require manual intervention to switch between modes?
I just want a working unit, I'll handle the bird training
Interesting, you guys really are serious. I'm interested in building this, but I'm don't have the time/not interested in trying to train the black bastards to actually use it. If you guys just want a working unit as far as the mechanics and electronics go I could start piecing it together and let you know if it's feasible (cost wise).
I still think a rooftop is the best location. Squirrels don't often climb buildings and from what I've seen, crows are pretty good at discouraging other birds away from a food source. Even if there are power lines or some other easy access for squirrels, a fence made of the proper materials would probably keep them out.
Yeah, there will have to be constant babysitting in the beginning. I don't think there's a good method to keep out other animals that would increase the complexity drastically. Though I'd certainly not mind the possibility if someone has a brainstorm.
I guess my point is if a bunch of squirrels eat all the nuts before you get a crow, experiment over. Either that or constant babysitting.
Wow you are really serious about this! I might draw out some prototypes to see how difficult it would be. Another issue is placing the device in an area where it will be used by crows, but not invaded by all other sorts of creatures. How does the device differentiate between a crow and a hawk, or say a crow and a squirrel? You could solve the squirrel problem with a weight gauge, but what about a hawk of comparable size? Smaller birds?
necessary waste during the first 3 stages. During the 4th stage if an animal isnt smart enough to drop a coin in the slot he wont get a peanut, eventually those animals who cant figure it out would stop coming by.An area away from humans is more important. If people are constantly walking by it and touching it less animals will drop by.
Wow you are really serious about this! I might draw out some prototypes to see how difficult it would be. Another issue is placing the device in an area where it will be used by crows, but not invaded by all other sorts of creatures. How does the device differentiate between a crow and a hawk, or say a crow and a squirrel? You could solve the squirrel problem with a weight gauge, but what about a hawk of comparable size? Smaller birds?
It was my understanding that the "finished" product was all there was to it, and you stage the crows around the machine. I don't think the machine is any different in the stages, you just scatter some peanuts around the machine... then when the crows are comfortable with htat, you scatter some coins and put some peanuts on the machine.
Then put a coin on the machine or some such and a peanut in the hopper... then when they figure out a coin going in the machine yields a peanut, there you go.
Then put a coin on the machine or some such and a peanut in the hopper... then when they figure out a coin going in the machine yields a peanut, there you go.
Klein built his with a 4-learning-stage shaping procedure, but if those stages aren't necessary, even better. I'll pay whoever can successfully train crows to use it, however that is. (Could be totally different design)
It was my understanding that the "finished" product was all there was to it, and you stage the crows around the machine. I don't think the machine is any different in the stages, you just scatter some peanuts around the machine... then when the crows are comfortable with htat, you scatter some coins and put some peanuts on the machine.
Then put a coin on the machine or some such and a peanut in the hopper... then when they figure out a coin going in the machine yields a peanut, there you go.
Then put a coin on the machine or some such and a peanut in the hopper... then when they figure out a coin going in the machine yields a peanut, there you go.
OP updated
The stages are key: they are what teaches the crows to use the machine.
The reason why I am offering such a high bounty for the video (20BTC!) is because there is now doubt about whether this can be done. I don't expect this to be a "quick buck", but rather a parts plus a small stipend for an interested hobbyist.
Hmmm, well seems like this project may be harder than I initially thought.
Alternatively, if someone can create a working prototype and record a video of a crow successfully exchanging a coin for a peanut, I'll send them 20BTC. Please include your username and "bitcointalk" on a sign in the video, just to be sure its actually you.
My previous offer still stands (after seeing the good video, I'll buy it for an additional 40BTC). I would like you to include plans.
it would take a while for the crows to become a custom to this. Id say more then a years worth of time to go through all 4 stages. IF you want a machine with just the last stage thats not so difficult.Alternatively, if someone can create a working prototype and record a video of a crow successfully exchanging a coin for a peanut, I'll send them 20BTC. Please include your username and "bitcointalk" on a sign in the video, just to be sure its actually you.
My previous offer still stands (after seeing the good video, I'll buy it for an additional 40BTC). I would like you to include plans.
The stages are key: they are what teaches the crows to use the machine.
The reason why I am offering such a high bounty for the video (20BTC!) is because there is now doubt about whether this can be done. I don't expect this to be a "quick buck", but rather a parts plus a small stipend for an interested hobbyist.
Hmmm, well seems like this project may be harder than I initially thought.
Alternatively, if someone can create a working prototype and record a video of a crow successfully exchanging a coin for a peanut, I'll send them 20BTC. Please include your username and "bitcointalk" on a sign in the video, just to be sure its actually you.
My previous offer still stands (after seeing the good video, I'll buy it for an additional 40BTC). I would like you to include plans.
it would take a while for the crows to become a custom to this. Id say more then a years worth of time to go through all 4 stages. IF you want a machine with just the last stage thats not so difficult. Alternatively, if someone can create a working prototype and record a video of a crow successfully exchanging a coin for a peanut, I'll send them 20BTC. Please include your username and "bitcointalk" on a sign in the video, just to be sure its actually you.
My previous offer still stands (after seeing the good video, I'll buy it for an additional 40BTC). I would like you to include plans.
I would pay for one as well, just want to reiterate that. So that's two guaranteed sales of a working CVM.
Jump to: