[brilliantcode]

> I would be super interested in a Kickstarter for open source, though I wonder if it would suffer from the same issues that Kickstarter itself suffers from (lots of ambitious projects, most not being delivered on time).

I'm actually working on that, please checkout my profile for link or this https://news.ycombinator.com/item?id=13604703

The problem you illustrated is a challenging one, and I'm still thinking about how to solve it, would love to hear ideas.

[Eridrus]

I signed up for your site already, I'm interested in seeing where this goes. I'm particularly interested in funding libraries, rather than finished products.

Kickstarter still does fine, despite this problem, but it has lead to people using Kickstarter as the final fundraising for manufacturing, which doesn't make a tonne of sense for software.

One option is a milestone-based system where the project author splits the work up into various chunks, and the funding gets released as they are done (as voted on by contributors within a few days of completion?) where if things start diverging too much from expectations the funding stops?

One thin that may be interesting besides developer-driven proposals, but community-driven requests. Sort of how bounties exist already, but potentially on a different scale. People could put their money where their mouth is and pre-commit to their dream open source self-hosted whatever. Maybe even pre-committing money to specific features.

Something like Patreon could exist for software people if people want to fund a maintainer, though that's less clear.

[brilliantcode]

Thanks for signing up! You've raised more questions than I can answer but it gives things to think about.

I've thought about the milestone system (it was the first thing that came to my mind) but do consensus based software have unchanging requirements? More importantly how is consensus reached? I think that for smaller scope like libraries or plug-ins it makes sense where there is one clear purpose. ex) write a scraper plugin that returns data.

Community driven requests would most likely be function or feature requests. It's exactly how you've described where you can put up a bounty for specific features where a developer will come along and implement it. If it's not worth their time (amount of work is far more than the money raised) than a few things could happen:

- another developer could do it for the purpose of building a portfolio or gathering resumes etc.

- the requester/backer can advertise it to his peers or companies, if it gets enough backing it would entice more developers to work on it.

Finally, the Patreon model where a lone dev or a team will work on maintaining an open source project on an ongoing basis.

It gets me super excited just thinking about the possibilities but I'm grossly overlooking the pitfalls and challenges of bringing people together in the first place.

I'm going to do some brainstorming and maybe lay it out in a blog post this weekend.