search

gratipay / inside.gratipay.com

Here lieth a pioneer in open source sustainability. RIP
https://gratipay.news/the-end-cbfba8f50981
57 stars 38 forks source link

Distribute funds #1162

Closed chadwhitacre closed 6 years ago

chadwhitacre commented 6 years ago

Placeholder for conversation about how to distribute funds gathered via the new homepage, once we have funds to distribute. 😬

So far I've got a conversation going with the JS Foundation (and the Linux Foundation by extension). Should reach out to PSF as well?

chadwhitacre commented 6 years ago

Thinking about various methods as tied to goals at https://github.com/gratipay/inside.gratipay.com/issues/1177#issuecomment-333268514.

I think we should try distributing the first $100,000 using Libraries.io data.

chadwhitacre commented 6 years ago

https://libraries.io/experiments/unseen-infrastructure

or https://libraries.io/search?q= which is by Source Rank — top 1,000 that respond within 30 days, weighted by Source Rank?

Would need to come up with an email contact list based on either set and mail them to see who wants to participate.

chadwhitacre commented 6 years ago

In the video #1155 we are heading towards using footage that talks about "partnering with Libraries.io to fund unseen infrastructure."

andrew commented 6 years ago

👍 happy to help mine to email addresses from the top 1000 projects repositories

chadwhitacre commented 6 years ago

👍 happy to help mine to email addresses from the top 1000 projects repositories

Love it, thank you! Also bringing in @BenJam's comments from https://github.com/gratipay/inside.gratipay.com/issues/1160#issuecomment-333520144. Can we reach a decision on unseen-infra vs. digital-infra vs. top-1000?

gets split however Libraries.io sez we should split it.

^ easier said than done. Certainly there's a function of splitting the money by

  • the amount the thing is used
  • the amount of support the thing already has
  • the amount of people already working on it
  • the amount of work the project looks like it already has

I think the /unseen-infrastructure list is a better starting point than the /digital-infrastructure list as there are good indicators that those projects need more support, but it won't solve all your problems.

@andrew and I have been thinking about collecting data on a project's financial support. We'll spike that out tomorrow together: https://github.com/librariesio/libraries.io/issues/1744

chadwhitacre commented 6 years ago

Proposal

For first $100k, will decide beyond that as it becomes clear that we'll need to ... ☺️

andrew commented 6 years ago

I wonder what percentage of /unseen-infra have no active person behind them, compared to /digi-infra 🤔

Another potential would be to put the offer up in public for work on the project if the project is active enough to be confident there will be someone to merge prs? Although that only really covers code contributions, much more tricky to manage some kinds of non-code contributions.

BenJam commented 6 years ago

Proof of work is a difficult problem to solve. its easiest if you set the task first but then the problem you are likely trying to solve is paying for time to set the tasks.

If you have pure donation on one side and contractor relationships on the other is it still a donation? What are you promising anyone who donates? If there are no promises then I would try to keep away from the difficult problems :)

mattbk commented 6 years ago

IMHO the funds should be untied from specific work. That's been the Gratipay model thus far, as opposed to the Bountysource model (pay $ for specific work) or the Patreon model (pay $ for general work but you get unrelated perks back).

chadwhitacre commented 6 years ago

Another potential would be to put the offer up in public for work on the project if the project is active enough to be confident there will be someone to merge prs?

If it's active enough to accept funds then it's active enough to farm out to others w/o us getting in the way, no? We give the money to the project, what they do from there is their problem, I think. Could use Bountysource, OC, etc.

If you have pure donation on one side and contractor relationships on the other is it still a donation?

It's certainly not a tax-deductible charitable donation, if that's what you mean (at least not for this first iteration ... perhaps down the line). It's a payment to Gratipay, LLC for open source software. Our main promise is a follow-up report on the impact of their "investment" (not technically an investment either 😆 ... the weird middle-ground of crowdfunding).

chadwhitacre commented 6 years ago

Follow-up report (#1172) will include:

rohitpaulk commented 6 years ago

IMHO the funds should be untied from specific work. That's been the Gratipay model thus far, as opposed to the Bountysource model (pay $ for specific work) or the Patreon model (pay $ for general work but you get unrelated perks back).

I agree 👍

Rather than phrasing this as "We'll give you money to meet the following expectations", we could phrase it as a cycle of "Here's money for the great job you've done in the past" -> "More work/progress to show => Likelihood of more money in the future".

BenJam commented 6 years ago

^ Interesting. With a wink and a nudge a windfall payment for work already completed on the assumption that it could be 'reinvested' by the individual if needed. I like it.

BenJam commented 6 years ago

This leave the door open to the individual taking as much as they like without blowback from the community. If the community deems the maintainer exorbitant they will stop contributing. It's self-regulating. Only problem is split between maintainer groups rather than individuals. As I say, the problem is already hard enough :)