Takeaways from the first Snowdrift.coop retreat

A couple of weeks ago, @team members came together for the first Snowdrift.coop (virtual) retreat. After ~6 hours of large and small group discussions, we were able to overcome some snow drifts of our own, and anticipate a renewed vigor in pushing the project forward!

This thread is my post to the team, and to the broader community who has been supporting our efforts.

First, I’d like to thank you all for participating, it is wonderful that everyone is still working together towards clearing the launch path! I really feel as though we made headway in terms of unblocking things. That said, now that they are unblocked, the ball needs to start rolling.

Please take your time and review the notes below. If anything is unclear, or if something was uncaptured, please mention it.

This snow pile may look large, but together we’ll clear the path to a better world!


  • We outlined a roadmap for the next iteration of the mechanism in alpha
  • The mechanism we launch with will NOT be the final iteration
  • Snowdrift.coop is not solving the problem of funding FLOSS, it is solving the snowdrift dilemma
    • …which we hope will eventually lead to solving the problem of FLOSS funding
  • We all agree that we’re eliminating the “kicking people out” part of the mechanism
  • We are charting a new direction in UI/UX development which includes:
    • The addition of a frontend lead role, responsible for:
      • Development of the steps outlined below
      • Communication with ui/ux leads before mockups or prototyping
    • Splitting off the frontend into a separate service from the mechanism backend
    • A mechanism-agnostic website, with the potential of multiple mechanisms running at the same time
    • This is a first step towards establishing a better process for resolving frictions in prototyping

NEXT STEPS (adroit):

  • Explore what we currently have in the prototype
  • Explore converting the prototype to Elm
  • Research something like elm-pages for noscript patrons
  • Work with mray to select one page from the prototype

NEXT STEPS (adroit, mray):

  • Select one page from the prototype that has unimplemented changes
  • Have adroit implement the page as is in Elm
  • Have mray make the changes, with guidance
  • Note any pain points and see how feasible they are to resolve


  • [ ] Create an epic in GitLab, make sure these issues are all captured and connected to “launch”
  • [ ] Express pledges as $6 toward a $36,000 / 6,000 patrons goal
    • shows the proportional concept
  • [ ] Express the meaning of $36,000 for Snowdrift.coop in terms of real progress
  • [ ] Express value of partial progress towards the full goal
    • consider multiple points along the way
  • [ ] Express the mechanism version clearly and consistently
    • e.g. 1.0, “alpha”, etc.
  • [ ] Write statement acknowledging (primary) shortcomings/issues to be improved on in future iterations
    • [ ] decide where this statement appears (likely multiple places)
    • see below
  • [ ] Implement patron dashboard view
  • [ ] Implement project page
  • [ ] Update video to reflect changes
  • [ ] Update how-it-works page to reflect changes

ISSUES NOT ADDRESSED BY ALPHA (to be publicly acknowledged):

  • Varying pledge options
    • Different patrons have different wealth / willingness
    • Different patrons have different relation / enthusiasm for project
  • Goal of crowd size vs dollar-amount
    • Related to varying pledge options
  • Different project scales/ambitions
    • How project goals are set
    • Adjusting a single goal
      • partially addressed, in theory
    • Multiple goals
      • concurrent or sequential/stretch
    • Site wide vs per project goals
  • Holding off funding until a threshold is hit
    • Patron vs Project decided thresholds
Group Photo

Invisible folk: msiep, iko, and bnagle17

Silly Version

3 Appreciations

Put another way: we believe that the snowdrift dilemma is a core obstacle to a particular part of FLO software funding: voluntary donations (mainly from individuals). To the extent that we are right and that we succeed at addressing the dilemma, we will have resolved a portion of that part of the funding issues.

Or shorter: FLO funding and public goods support in general has a snowdrift-dilemma problem. We’re aiming to solve that. That’s our focus. So, we’re not aiming to necessarily solve other issues related to public goods (at least not now).

because it was too uncomfortable, not intuitive enough, and costly to explain. But the reason for it remains: for a matching pledge to be real, it can’t have some people counting in the crowd as matchers but not actually matching (which is what would happen if people could set their own budget cap and stay in the crowd at that point without continuing to match).

Our shift is toward setting target goals for projects at which point all patrons will hit their cap at once, matching is essentially turned off at that point. Perhaps there will be opt-in stretch goals to continue matching further. This way, the crowd is all still pledging and matching together, and patrons still have a clear budget limit.

This is something we’ve planned all along, but we’re bringing it up as priority. We will state at Snowdrift.coop Project | Snowdrift.coop what we estimate to be able to accomplish with increased funding.

We also did not have consensus about how and when these issues will be addressed. Particularly, we have questions about how much research to do (modeling, surveys, outreach to projects and patrons) and how much to trial live with multiple models versus to keep the functioning system simple and separate the messy kitchen from the end product so to speak. There are many ways to make progress on these questions before and during alpha operations.

As mentioned (and something we’ve long been aware of), we clearly have a severe diversity problem. Superficially, it looks bad, and practically we are missing perspectives. On that note, I’m opening a new topic: Recruiting for diversity

1 Appreciation