Bounties that take into account the time cost
Bounties are like "if you find the solution to my problem, I'll give you $Y."
But every time I think about writing a bounty, I worry it's not going to be very helpful. Because people often don't vet their solutions that much. It takes a lot of time to decide if the potential solution is an actual solution.
e.g. I say "$30 if you find me a place that will give me a dental appointment in the next week." 3 people suggest dentists they like; they each say "try calling this person!" No one actually called the dentists to ask for availability. I take what I can get; I call them myself. Dentist 2 has availability next week. Oof, this didn't save me that much time; I could've just googled for local dentists. Am I supposed to pay out now?
e.g. I say "$500 to fix my nighttime breathing problems," and then I get these suggestions: buteyko, Wim Hof, sleep study. But then I spend 2h reading about buteyko and learn it would take me 1 month of regular practice before I should expect to see results. (And maybe it just won't fix my problems at all.) It's going to take me more than $500 of additional time/money to evaluate each potential solution, so it'd be great if people only made suggestions with a high probability of success -- but how do I make that happen?
Even if verifying a solution takes just 5min on my end, if 20 people make suggestions, then that's going to take me a lot of time.
I wish I could say "if you make my problem disappear with as-little-work-as-possible on my end, I'll give you $Y." But it seems hard to operationalize for different problems.
like this
Ben Weinstein-Raun, Ben Millwood, Adam Scherlis, Sam FM and Satvik like this.
Adam Scherlis
in reply to kip • • •like this
Ben Weinstein-Raun, Sam FM, kip and Raemon like this.
Alex Altair
in reply to kip • • •Alex Altair
in reply to Alex Altair • • •Ben Millwood likes this.
kip
in reply to Alex Altair • •Ben Weinstein-Raun
in reply to kip • • •kip
in reply to Ben Weinstein-Raun • •Ben Weinstein-Raun likes this.
Ben Weinstein-Raun doesn't like this.
Sam FM
in reply to kip • • •This makes me think of how in normal society there are a variety of expensive-to-verify tasks that have their quality/correctness guaranteed by the threat of reputational damage, or even malpractice/liability lawsuits.
Say a civil engineer submits a design for a bridge, and then it collapses on day #2 after the expensive construction is complete. Even if he can't get sued to cover the construction company's loss, he'll have trouble finding work afterwards.
But with internet bounties, generally people aren't laying their expertise and credibility on the line with their suggestions, so there isn't much of an incentive to hold back from offering dubious solutions.
kip likes this.
Satvik
in reply to kip • • •like this
kip and Sam FM like this.