Money is already all over the open-source world. Many smaller projects will not involve it much, but it matters in the larger ones. E.g., much of the code contributed to the Linux kernel comes from people who write it as part of their jobs. That means that feature additions & bug fixes in Linux will be largely those that matter to (say) Red Hat. This idea offers a way to get work I want done prioritized without needing to know how to do it myself.
It also offers the possibility of software development midway between the start-up model that groups like YC and Kickstarter are pushing, and the no-money-involved-ever model. Surely somewhere between the two we might find something workable.
> Also, when it comes to many projects, the organization that offers services around them has a negative incentive to contribute clear, quality code.
A good point. I don't have an answer for that one.
I'm probably being naive, but in my version of this idea, I was thinking that contributing well-written, tested, and documented code back upstream from the company's account would be a good way to have the work that's done contribute to the marketing.
Part of the reason why I haven't moved past just talking about this idea is the possibility of it having a negative impact in terms of the original creators/maintainers, so I'm definitely interested in hearing more about this perspective on it.