A more common use-case would be a kickstarter-like situation, where some number of people have to be on board for anything to happen, but unlike regular kickstarter, the outcome is also in question.
I’ve also been interested in this use case for assurance contracts. I suspect an assurance contract website would have to have this functionality, for exactly these sorts of cases where the desired outcome isn’t obvious.
This is also useful for something I’ve thought about, which is crowdfunded Bounties, where people gather funds for an idea they’re interested in in the hopes that it will motivate someone to do it.
Let’s say for example I want an HPMOR video game made. In regular crowdfunding I would open a campaign to fund my own project, and get funds from people to help me make a video game. Problem is, I don’t know how to make video games.
In bounty crowdfunding I’ll open a campaign that explains why this is a good idea that should be funded, people who also want an HPMOR video game will support it, and then anyone can make the game and reap the funds as if it was his own crowdfunding campaign.
The problem is, how do you make sure someone worthy gets the funds? In regular crowdfunding you see the offer before you give it money. Here you need some way to verify worthiness.
So I think something like a no-confidence vote can be used. When someone makes a claim to the funds, a vote is held to choose who if any of the bidders get the funds, and if no one gets it then the campaign stays open.
What I’m not sure how could be done well, but is more a technical problem than a theoretical problem, is the commitment mechanism for pledging money. In Kickstarter this isn’t as much a problem for two reasons, one is the shorter time frame (campaigns usually run for 30 days), and second that someone has already put in some effort beforehand with the belief people will pay them for it.
Here, you have lots of people saying they will pay for something, so you might be motivated to work on it, but how can you be sure they will really pay? If you solve that by sending money to a shared fund held by the platform, then there’s a different problem where people might be sending their money to do nothing for years until someone comes and takes on the bounty.
Yeah, this seems really interesting but really tricky. Someone puts a lot of effort into an HPMOR game but it’s just not quite up to snuff and gets no-confidence from the community of funders? Makes it seem like an unappealing prospect even for people who are fairly confident in their skills—which in turn means the potential reward needs to be higher to compensate for the risk, making it a more expensive funding method. Let’s say there’s a 5% risk of no-confidence (as estimated subjectively by a potential programmer). That does not just mean a 5% markup, because for small firms / individuals, 5% chance of no compensation is a high risk (ie, they’re not very risk tolerant—they Kelly bet, or realistically, are even less risk-tolerant than that; so a 5% chance of no compensation is effectively a big downside risk).
So you also need something resembling “no-confidence insurance” built into the platform. This could take the form of some kind of mechanism for investing in potential programmers: people can vet the candidates and extend funds immediately, with some returns for them in the case that their candidate succeeds.
But, all this complexity also probably makes the funding model a harder sell, especially for early adopters.
The way I thought about it was that someone gives a proposal, and if they win then they get the funds and are now obligated to create the thing (like with kickstarter), or perhaps it could be a half now half later scheme. Anyway the point is that the most important thing for a potential developer is how much effort they need to put in before they secure/get the funds. I guess that changes mainly depending on reputation. If someone is known for successfully taking on projects after getting funded, people will be more likely to accept their proposal with less initial investment than other people put in, and other people might have to put in a lot of effort and resources into a proposal to stand out and prove themselves.
There’s a balance here of who takes the risk—the funders who aren’t sure beforehand if the developer is going to do a good job, or the developer who doesn’t know beforehand if they’ll be accepted by the funders. Maybe the balance could be different between different projects.
(I’m sure they figured it out already in dath ilan, haha)
I’m reminded of the common idea in the startup-funding community, that you’re investing in the team, not the idea. The idea is an important foundation, but projects can and should pivot. So, to invest, you need to have confidence in where it could go—which depends on the people who will take it there.
The whole “crowdfunding bounties” idea is a bit like “kickstarter, but decouple who proposes a project from who works on it”, right? So, this might result in a big mess if it’s not done very carefully. That’s one of the main motives of my previous reply—worry about the chesterton fence of investing-in-people-not-ideas.
Not to say it is an insurmountable obstacle; just to say that’s one area where I’d poke at the idea a bunch.
The whole “crowdfunding bounties” idea is a bit like “kickstarter, but decouple who proposes a project from who works on it”, right?
Yes, that’s a very good way to put it.
And I agree that investing-in-the-team is important and shouldn’t disappear with this model.
So to reiterate the idea:
Someone would come up with an idea, which would be far more general/bare bones than the usual kickstarter project. Something more like “An HPMOR video game” than “An HPMOR video game, with these mechanics, and these graphics, and this story, etc...”.
And people would commit money as a show of interest in an idea, sort of like an investor saying “I’m interested in startups working on VR, send me your pitch”.
Then if enough money has been committed to attract teams/individuals to the project, they would have to pitch themselves and their vision to the crowd (with varying degrees of investment before the pitch, that part doesn’t change from kickstarter/startups), and the crowd would have to decide together whether to accept that pitch based both on the team/individual and their vision.
If they do, the promised funds have to be transferred, though not necessarily in one go. It could be dependent on certain milestones, for example (I think this is also common in the startup world?).
-
But even with lots of talking and figuring it out theoretically, it would probably need a lot of actual testing to get right.
A simple first test of this could be to crowdfund a simple a bounty with kickstarter (perhaps for LessWrong posts), create some voting system to let the funders pick who wins the bounty, and see what we learn from it.
Ah, ok, fair enough, so the initial “commitment” of money isn’t really much of a commitment, it is basically a collective signal of interest in kickstarting projects of a specific sort. And then, specific teams basically run a regular kickstarter “under that banner”, hoping to attract those specific investors. Plus, perhaps, some additional flexibility on when funds actually get transferred.
So it sounds to me like the simplest version is to make a website “parallel” to kickstarter, ie, where kickstarter investors can indicate how much $ they think they would invest in various projects (and where kickstarter projects can submit their kickstarters as supposedly meeting those requirements, and investors get notified, and have an easy interface to approve funds).
I did mean an actual commitment, that’s why I said it’s a technical problem in my original comment. I don’t imagine It’s a commitment you can’t walk back on at all, but there should be terms on that (say, a week/month notice, perhaps changing depending on the project), so potential developers* can be confidant the amount shown is the amount they’ll get (That’s really the important bit). Otherwise they have to guess whether people are still interested, and they have to get them to then actually pledge money, so it would end up not actually helping them that much—and if it doesn’t help the developer then it also doesn’t help the investors.
So what you suggested can work as a precursor without the commitment mechanism, but I think the commitment mechanism is important enough that it wouldn’t actually tell us much (so it wouldn’t really be an MVP).
*What’s the right word for the person/team that takes on a projects? executors? because they execute the project? Sounds a bit dark though, lol.
I’ve also been interested in this use case for assurance contracts. I suspect an assurance contract website would have to have this functionality, for exactly these sorts of cases where the desired outcome isn’t obvious.
This is also useful for something I’ve thought about, which is crowdfunded Bounties, where people gather funds for an idea they’re interested in in the hopes that it will motivate someone to do it.
Let’s say for example I want an HPMOR video game made. In regular crowdfunding I would open a campaign to fund my own project, and get funds from people to help me make a video game. Problem is, I don’t know how to make video games.
In bounty crowdfunding I’ll open a campaign that explains why this is a good idea that should be funded, people who also want an HPMOR video game will support it, and then anyone can make the game and reap the funds as if it was his own crowdfunding campaign.
The problem is, how do you make sure someone worthy gets the funds? In regular crowdfunding you see the offer before you give it money. Here you need some way to verify worthiness.
So I think something like a no-confidence vote can be used. When someone makes a claim to the funds, a vote is held to choose who if any of the bidders get the funds, and if no one gets it then the campaign stays open.
What I’m not sure how could be done well, but is more a technical problem than a theoretical problem, is the commitment mechanism for pledging money. In Kickstarter this isn’t as much a problem for two reasons, one is the shorter time frame (campaigns usually run for 30 days), and second that someone has already put in some effort beforehand with the belief people will pay them for it.
Here, you have lots of people saying they will pay for something, so you might be motivated to work on it, but how can you be sure they will really pay? If you solve that by sending money to a shared fund held by the platform, then there’s a different problem where people might be sending their money to do nothing for years until someone comes and takes on the bounty.
Yeah, this seems really interesting but really tricky. Someone puts a lot of effort into an HPMOR game but it’s just not quite up to snuff and gets no-confidence from the community of funders? Makes it seem like an unappealing prospect even for people who are fairly confident in their skills—which in turn means the potential reward needs to be higher to compensate for the risk, making it a more expensive funding method. Let’s say there’s a 5% risk of no-confidence (as estimated subjectively by a potential programmer). That does not just mean a 5% markup, because for small firms / individuals, 5% chance of no compensation is a high risk (ie, they’re not very risk tolerant—they Kelly bet, or realistically, are even less risk-tolerant than that; so a 5% chance of no compensation is effectively a big downside risk).
So you also need something resembling “no-confidence insurance” built into the platform. This could take the form of some kind of mechanism for investing in potential programmers: people can vet the candidates and extend funds immediately, with some returns for them in the case that their candidate succeeds.
But, all this complexity also probably makes the funding model a harder sell, especially for early adopters.
The way I thought about it was that someone gives a proposal, and if they win then they get the funds and are now obligated to create the thing (like with kickstarter), or perhaps it could be a half now half later scheme. Anyway the point is that the most important thing for a potential developer is how much effort they need to put in before they secure/get the funds. I guess that changes mainly depending on reputation. If someone is known for successfully taking on projects after getting funded, people will be more likely to accept their proposal with less initial investment than other people put in, and other people might have to put in a lot of effort and resources into a proposal to stand out and prove themselves.
There’s a balance here of who takes the risk—the funders who aren’t sure beforehand if the developer is going to do a good job, or the developer who doesn’t know beforehand if they’ll be accepted by the funders. Maybe the balance could be different between different projects.
(I’m sure they figured it out already in dath ilan, haha)
I’m reminded of the common idea in the startup-funding community, that you’re investing in the team, not the idea. The idea is an important foundation, but projects can and should pivot. So, to invest, you need to have confidence in where it could go—which depends on the people who will take it there.
The whole “crowdfunding bounties” idea is a bit like “kickstarter, but decouple who proposes a project from who works on it”, right? So, this might result in a big mess if it’s not done very carefully. That’s one of the main motives of my previous reply—worry about the chesterton fence of investing-in-people-not-ideas.
Not to say it is an insurmountable obstacle; just to say that’s one area where I’d poke at the idea a bunch.
Yes, that’s a very good way to put it.
And I agree that investing-in-the-team is important and shouldn’t disappear with this model.
So to reiterate the idea:
Someone would come up with an idea, which would be far more general/bare bones than the usual kickstarter project. Something more like “An HPMOR video game” than “An HPMOR video game, with these mechanics, and these graphics, and this story, etc...”.
And people would commit money as a show of interest in an idea, sort of like an investor saying “I’m interested in startups working on VR, send me your pitch”.
Then if enough money has been committed to attract teams/individuals to the project, they would have to pitch themselves and their vision to the crowd (with varying degrees of investment before the pitch, that part doesn’t change from kickstarter/startups), and the crowd would have to decide together whether to accept that pitch based both on the team/individual and their vision.
If they do, the promised funds have to be transferred, though not necessarily in one go. It could be dependent on certain milestones, for example (I think this is also common in the startup world?).
-
But even with lots of talking and figuring it out theoretically, it would probably need a lot of actual testing to get right.
A simple first test of this could be to crowdfund a simple a bounty with kickstarter (perhaps for LessWrong posts), create some voting system to let the funders pick who wins the bounty, and see what we learn from it.
Ah, ok, fair enough, so the initial “commitment” of money isn’t really much of a commitment, it is basically a collective signal of interest in kickstarting projects of a specific sort. And then, specific teams basically run a regular kickstarter “under that banner”, hoping to attract those specific investors. Plus, perhaps, some additional flexibility on when funds actually get transferred.
So it sounds to me like the simplest version is to make a website “parallel” to kickstarter, ie, where kickstarter investors can indicate how much $ they think they would invest in various projects (and where kickstarter projects can submit their kickstarters as supposedly meeting those requirements, and investors get notified, and have an easy interface to approve funds).
I did mean an actual commitment, that’s why I said it’s a technical problem in my original comment. I don’t imagine It’s a commitment you can’t walk back on at all, but there should be terms on that (say, a week/month notice, perhaps changing depending on the project), so potential developers* can be confidant the amount shown is the amount they’ll get (That’s really the important bit). Otherwise they have to guess whether people are still interested, and they have to get them to then actually pledge money, so it would end up not actually helping them that much—and if it doesn’t help the developer then it also doesn’t help the investors.
So what you suggested can work as a precursor without the commitment mechanism, but I think the commitment mechanism is important enough that it wouldn’t actually tell us much (so it wouldn’t really be an MVP).
*What’s the right word for the person/team that takes on a projects? executors? because they execute the project? Sounds a bit dark though, lol.