A "Critical Mass" Mechanism

Hi stranger, welcome to Snowdrift.
We host lots of cool projects that are looking for funding. You choose whatever amount you want, and it’s donated automatically every month to the project. That’s your “base pledge”, and it helps the project keep existing, so I suggest you put it as high as the project is currently worth to you. But again, you can set it to whatever you want.

Looking among all the project’s other patrons, you can see there are crowds forming. They’re forming around particular goals. Goals are like things on the community’s wishlist. If you see a wish you love, a wish that would mean a better future for the project, a wish that would make it worth that much more to you – you can join that crowd!

All you have to do is say how much higher you’d be willing to raise your base pledge to see that goal come true. And again, it can be any amount you want. But the more you pledge, the closer you bring the goal to coming true!

Some of the crowds grow fast, and some don’t. You can join as many as you like – it’s all hypothetical money you’re pledging. You can even leave a crowd whenever you like. Or change your mind and give more! If you’re lucky, you may even push one over the edge!

As soon as a crowd is big (and generous) enough to afford the goal – bam! Everyone in the crowd starts giving what they pledged. They’re held accountable to it, too! For a certain amount of time, everyone in the winning crowd has their base pledge “locked in” at the new total, and can only be set higher. But you know what just got unlocked? The crowd’s wish coming true!

The projects are held accountable too. Goals are concrete and achievable by design. For a community wish to become a goal, they must lay out exactly what additional resources they’d need to get it. A new full-time developer maybe? A new full-time drummer? Project leaders must form not just a “this might be enough” list, but a “this is definitely enough” list.

Then from the “definitely enough” list, they also come up with a “definitely enough” budget. We want to guarantee success – if the estimate is a range, the goal becomes the maximum of the range. All of this is public and can be rigorously refined by the community. If you want, you can audit it before pledging. Point is, by pledging to a goal, you’re not just backing a wish - you’re backing a whole thought-out, publicly audited plan to make it come true!

This system has the almost magical property that everyone’s always giving the maximum they’d feel comfortable with giving at that time. The “comfortable” is good for the patrons, the “maximum” is good for the projects. Not some people, everyone. Not some portion of what they’re willing to give – all of it, always. And even if they’re willing to give much more for a certain brighter future – this system doesn’t take a dime more from them until that brighter future is already coming true.
– Dalt Wisney

3 Appreciations

Would you accept a free lottery ticket?

I don’t know about you, but I rarely participate in the lottery. The tickets may be cheap, but the chances are winning are so slim that even huge jackpots are are simply not worth it. (That said, the thrill is real, and the psychology is proven: since its inception, lottery tickets are purchased every day by multitudes!) So, if you’re like me, you would normally turn down an offer to purchase lottery tickets.

But what if I introduced you to a special lottery, with a drastically more realistic chance of winning - and offered to sell you tickets at just $1 a piece. Not only is that super cheap by default, but you can also buy as many tickets as you want, and each one puts you that much closer to winning. So if the chances of winning with one ticket was 1 out of, say, 1000, with a spare $20 bill you could push that up to 20 out of 1000.

By now, participating in this game is starting to sound a lot more appealing. But your rational mind still has reason to hold you back. Sure, maybe you can spare a few dollars, but you you’ll probably just lose them - you get nothing in return.

BUT what if I told you that if you lose, you pay NOTHING? :astonished:

Woah! You’re rational mind can’t hold you back anymore: suddenly, it’s a “no-brainer”! Not only do you have nothing to lose, but assuming you value the prize much higher, you might as well dig as deep into your pockets as you can to really maximize your chances of winning. If you don’t win, it’s like nothing ever happened, no biggie. And if you do (“this huge prize for only twenty bucks!?”), you’re getting a crazy good deal no matter what!

Since we’re on a roll, let’s make the deal even sweeter. Going after a jackpot like this seems pretty selfish, so let’s make it altruistic as well. What if I told you… that “if you win, EVERYONE wins.”!? :exploding_head:

You heard that right - if you happen to win this lottery, not only will you get 100% of the prize, but we’ll give the same one to everyone else as well. Heck, we’ll even include all the people that didn’t even participate!

Well, that’s Game Over. Not only is this a “no-brainer”, but one could argue it’s even moved into the realm of “moral obligation” if you have any money to spare.

But what if you don’t have much money to spare? Simple - since everyone wins when one person wins, you already know that anyone’s increased chances of winning translate to your increased chances of winning. In other words, instead of (or in addition to!) buying more tickets, you can increase your chances of winning by recruiting more people! In this special lottery, there is no competition - the more people you get involved, and the more tickets you get them to buy, the more likely everyone wins!

Overcoming The Snowdrift Dilemma

Applying The Analogy Using the psychology described above, I believe we can solve the Snowdrift Dilemma *without crowdmatching*!

In the analogy, the “magic lottery” is a project’s goal. The “chance of winning” is the crowd’s progress toward the goal. And the “tickets purchased” is the patron’s zero-risk pledge towards the goal.

Bringing it down to Earth:

  • The “prize” is your favorite FLO project unlocking a bunch more resources, so they can improve for you.
    (I know, not quite as motivating as personally winning a large sum, but as long as they value the improvements more than their pledge, the psychology should still work!)
  • The pledge would not be a one-time contribution, but a monthly one.
    (Again not quite as motivating to pledge a lot, but you still only pay if you win.)
  • In this mechanism, patrons also sustain a project as-is with a monthly any-amount contribution. If a goal succeeds, the pledge is actually applied by permanently adding it to that monthly contribution.

But besides these three points, everything about the magic lottery really does apply. All that motivational goodness! I believe we can use this to overcome the Snowdrift Dilemma.

A Community Clearing A Snowdrift In case it wasn't clear, this plan clears "snowdrifts" by addressing them all-at-once or not at all. Using the classic analogy, two examples:
  • A snowdrift blocks the road into a community. People want it cleared - but not badly enough to do it all by themselves, nor enough to do it with a group if the group is too small to accomplish it with less than X hours of work per person. Recognizing this dilemma, one leader makes a conservative (high) estimate of the minimum number of shovelers required to clear it in X hours of work, and sets up an event on <community’s social media page>. If not enough people RSVP, the event doesn’t happen. If there’s enough RSVPs, the shovelers come out at event time, and the snowdrift is gone in two hours!
  • A snowdrift blocks the road into a community. People want it cleared - but not badly enough to spend more than a few dollars on it. Recognizing this dilemma, one leader calculates how much TNT is required to blow the snowdrift to oblivion. (:stuck_out_tongue_winking_eye:) He then makes a local Kickstarter/GoFundMe campaign to raise the money to buy it. If it fails, the snowdrift remains. If it succeeds, the snowdrift is blown up and the spectacle is enjoyed by all.
1 Appreciation

A Day In The Life Of A Patron

You’re a local adjunct college professor, and you make enough money to live, though not by much. In your spare time, you study weather systems, and develop models that will help early warning systems save lives from natural disasters.

You arrive home one day and it’s time to review your account statement. Settling in, you sit on your futon and skim over the money that left your account last month.

reading bill

$47.97 Broadband Internet
$16.73 Telephone
$75.17 Grocery delivery service
$95.00 Maple math tool license

$48.00 Electricity
$55.00 Cable TV

Woah! Wait a minute, $55? “Last time it was $50”, you grumble. You knew a price increase was inevitable, and there’s not much you can do about it. “A 10% increase? Seriously? Did my cable service just get 10% better somehow? Am I suddenly getting 10% more value than I used to!?” you quip to yourself. But no, it’s the same old service.

But then you see it

As you finish scanning your expenses, at the very bottom you see an unusually high charge - one which used to be only $5 a month. Your eyes grow wide.


$67.45 Car insurance
$15.00 Donation to FSF
$20.00 Snowdrift Contribution to “Sage Math Tool”

:astonished: Woah!! $20? :money_with_wings:

But you know what this means, and it turns out, you’re not upset - you’re EXCITED!
Why??
Because that $20 means your favorite FLO math tool is finally getting its own FULL-TIME DEVELOPER!

That struggling project is now getting a huge upgrade, ensuring support and improvements for years to come. The best part? Not only does the developer get a full, competitive salary like commercial software, but you’re paying for that salary FOR A MERE $20 BUCKS A MONTH!

Actually, you SAVE money

But it gets better. Why are you so excited for this project to get a full-time developer? Because you follow the project closely and you know that in no more than a year, that special treatment will improve the software so much that you can DROP the proprietary tool you’ve been using FOR GOOD, and when that day comes, you can drop the hefty LICENSE FEE along with it! It costs you $1140 a year. As you can see on the bill above, that’s $95 each month.

So by next year, you’ll be paying $75 LESS each month, saving you loads of money, all while using software that respects your freedoms!

It’s all thanks to the GUARANTEE

Since you don’t have a large income, you can’t afford to donate as much as you’d like to volunteer projects - especially projects you can’t rely on for your everyday work life. You can hardly afford that ridiculous $95/mo fee for Maple, but you needed it for your research since Sage was underdeveloped. With the spare change you had left, $5/month was about all you could donate to the cause, which is what you’ve been contributing up until now.

But thanks to the Snowdrift.coop guarantee policy, you had asked yourself “what if I KNEW my money could get Sage where I want it to be? How much would I contribute to Sage if I could get rid of Maple?”, and realized you could pledge MUCH more - money that was only freed up by getting rid of Maple!

If you weren’t sure that you’d be able to replace Maple, it would be way too risky for you to give away that much money. But since the project knew exactly how much a developer would cost, and Snowdrift.coop will only take your pledge money if the crowd can afford it - it wasn’t much of a gamble at all!

Mockup of pledge page

1 Appreciation

Some Virtues

Straightforward mechanism summary

From the patron’s perspective, it’s profoundly simple. In one sentence:

You choose how much you wanna give each month, and then for each goal, you pledge how much more you’d give monthly if the pledges in that crowd ever total enough to meet the goal!

Or, in three brief steps:

1. Patrons select how much they're willing to give to sustain a project as it currently exists.
  • This is the “sustaining” or “base” pledge and is transferred monthly, starting right away.
  • They can select a different amount any time they like, taking effect next payout.
  • Basically Patreon.
(How this money gets used)
  • Each month, the project receives the crowd’s contributions and uses this lump sum the best way they can, spending it on new and existing resources. At the same time, the project also keeps track of the resources it doesn’t have and forms a wishlist, prioritizing it via feedback from supporters and researching the exact cost associated with acquiring those resources.
  • When the project confidently establishes a hypothetical additional income that is just big enough to guarantee that they can afford one of the changes on the wishlist, the project sets the next “goal” with this amount as the target. The goal includes a concrete implementation plan.
2. Patrons look at the promises of the project's next goal, and select how much they would be willing to increase their monthly pledge by iff they were to get the promised changes.
  • Again, pledgers may drop out or raise/lower this amount whenever they want!
  • Thanks to the way goals are defined, pledgers can rest easy knowing the things they’re pledging for are GUARANTEED. It’s like you’re just paying for it yourself, but at a fraction of the cost!
3. Iff the day comes when the crowd surpasses the goal's target, the goal is achieved, and those who pledged to it have their monthly contributions jump up by the amount they pledged.
  • The patron’s monthly contribution is locked there for a while. No take-backs!
  • They may always increase it, however!
(Actions taken by the project upon success)
  • The project accepts the money.
    • This is a manual process (button click) so that projects get every chance they need to make sure their guarantee is still rock solid. (Costs haven’t increased, etc.)
    • They can even wait a few days if they want, and allow the crowd to possibly grow even more.
  • The project swiftly moves to hire the developers, pay for the hosting service, upgrade the server’s bandwidth plan, start constructing the public infrastructure, or whatever else was exactly promised in the goal’s description!
  • They then publish a post on the new changes, arriving just as people start noticing the difference in their transaction history.
  • Again thanks to the way goals are defined, the patrons should see and feel the results immediately. This will help with the “sting” of their new monthly contribution, if it’s much higher.

…that’s it!

Let me know what you think :slight_smile:

What are goals?

Though we haven’t talked much about standards for goal-setting, we will have to codify it eventually, no matter what mechanism is chosen. I’m proactively defining goals here, because this opinionated definition will greatly help people to learn to trust Snowdrift (like they generally do with e.g. Kickstarter - though I think we can disappoint even less people than that!) and thus reinforce the idea that you get what you paid for. This mechanism proposal more or less relies on that promise: the stated outcome should be guaranteed.

If you’re a project on Snowdrift, trying to define a goal:

  • Focus on what resources you’d be able to immediately acquire with the money, not what output you’ll eventually produce.
    • Software example: Instead of “We will add features X Y and Z!”, prefer “We will hire a full-time developer!”
    • Artist example: Instead of “I’ll start writing/release a new album!”, prefer “I’ll get access to a professional recording studio for 15 hours a month!”
  • Base the target on actual costs, and actual quotes (if contractor labor is involved) that you can refer to when defending your target.
    • Example project with website: Instead of “We want some money for website hosting”, prefer “We want to subscribe to ExampleHost, which costs us $17 per month”
  • If you only have a cost range, always set the goal to the upper bound of that range. “Better be safe than sorry” should be taken as a rule, even it will take longer for your goal to be met.
  • Have an implementation plan for your goal, that prefers taking easily provable actions - e.g. a transaction that generates a receipt.
"It's just Patreon plus Kickstarter!" In fact, one of the perks of the simplicity of this mechanism is that it can be quickly pitched to any layman familiar with Patreon and Kickstarter.

Example Pitch to the Crowdfunding-Aware Layman:

It’s like Patreon crossed with Kickstarter!
First, You automatically donate a fixed amount to a project every month. Just like Patreon! (But with our system, you can choose whatever amount you think it’s worth!) Second, in addition, you look at the project’s goals and pledge money to get those “perks”, knowing that you pay none of it unless one day everybody’s pledges total enough to meet the goal - which is when everyone pays out and everyone gets the benefits. Just like Kickstarter! (BUT with our system, you again get to choose any dollar amount you want!)
So what’s the twist, you ask? Instead of pledging a one-time amount toward the goal, you actually pledge how much you’re willing to increase your monthly donation by if the project makes the improvements you asked for! :exploding_head:

Smaller crowds

Sometimes a particular project is not widely popular. But if Snowdrift was ideally designed, it would make every dream come true where there’s enough people on earth willing to give enough money (total) to come true. Even the most expensive projects with the smallest fanbases should be able to reach critical mass if it exists. This system is the only one to guarantee that if a critical mass exists, it will be reached.

What is critical mass?

If the whole world combined is willing to give $X to make project/goal Y a reality, and $X is more than enough to make Y a reality… the ideal platform should make Y a reality.

If it takes any longer than that, or if it just manages to never happen despite a critical mass existing, there is wasted potential. In my opinion, the quality of the algorithm for bringing funding to projects should be judged harshly by that one major criterion above all else: how much potential is wasted? Maybe we can even call this the mechanism’s efficiency .

Framing is trivial

Instead of debating over “I will give $10 at $10 000” vs. “I will give $10 at 1000 people”, and worrying about what incorrect implications some people might take away, we can just say this about our monthly pledge:

“I will give $10 to reach the $10 000 goal”.

There’s no confusion, in fact it’s literally true: if you ever end up giving the $10, the $10 000 goal is reached. It’s like you just multiplied your money. That’s a powerful feeling that people will love.

It’s also easier to say than

Patrons always know what critical mass really is

In this system we always talk in real dollar output, and projects show exactly why they chose that amount – as transparent as it gets.

Why this is the most evidence-based mechanism proposal so far

Crowdmatching theory assumes not only that people will give much more than they would have otherwise in instances where their donations are matched or multiplied by someone else (which is widely seen to be true) – but also that this mental phenomenon works on repeat (every month, not just in a particular moment) and with a constantly changing, not necessarily predictable match rate.

This theory is still largely untested in the wild. Just because people reliably give more when they know they’ll be matched at a fixed rate, doesn’t mean they’ll be willing to “automatically” give more due to some future match rate that keeps changing.

“Critical Mass Theory”, on the other hand, is happening every day on Kickstarter, Massdrop, and hundreds of other sites. This system also relies on the assumption that the mentality works on repeat, but not with a constantly changing match rate.

Instant Gratification

To appeal to our primitive desires (for a good cause!), it’s clear to see how “your snowdrift bill never changes, and if it does change, it’s because you’re getting results!” and the “guaranteed” nature of the way goals are defined, leads to a rapid stimulus-response pairing. You contribute the money, you get results. You’re immediately rewarded. Imagine how fast that will build the visitor’s trust in our system!

No half-assed funding, no discomfort zone

See the thread describing the issue. This is the only current mechanism proposal that avoids the described problem entirely.

No Patrons vs. Dollars debate

There is less need to debate on the “patron based” or “dollar based” framing for this mechanism, since it would allow (and even encourage!) a group of 10 dedicated fans to meet a $1000 goal by contributing a massive $100 each, just as much as 1000 people meeting the same goal by contributing $1 each. The patron count is irrelevant to the goal’s financial support.

Native support for multiple goals

We can even let some of them run simultaneously. Let’s face it: we will always want goals to be more flexible eventually, if we start with limits like “goals must be stacked on top of each other” or “only one goal at a time”. Why not start with the most general solution we’ll eventually move to, in a mechanism where it’s actually easy to optimize for?

No one gets “kicked out”

This finally becomes a non-issue! Dealing with patrons getting automatically (or manually) kicked out potentially breaks many of the promises of the other systems and leads to the perception of unfairness. In this system, you’re made well aware of your financial commitment, just like in bidding in an auction. When it doesn’t matter, you can change your mind. When it does matter, everyone is locked in to what they committed to. It’s much easier to commit to fixed amounts!




There are many benefits to the system, so I tried to list a few that are unique among mechanism proposals so far. Let me know what you think would be a good nickname for this mechanism.

4 Appreciations

I appreciate the prose and explanation, and thought that went into all of this! TBH I’d love to see it boiled down into the most basic functions/mathmatical aspects of the mechanism: because a lot of the ‘pitch’ aspects could apply to other mechanisms as well (as @wolftune has talked about - what are the hard levers that make this different so it makes it easier to compare?). I think this would be really useful to have by tomorrow because there’s a lot of the above to read that’s not essential to read (point being I know it’s pertinent explanation but I’d want to see all the ‘levers’ consolidated at the top and the explanation/context below, etc). I’m envisioning a spreadsheet on which one could compare which features are similar or different such as:
(for this mechanism if understood correctly

  • Can you change your pledge:
    • Yes
  • How many donation amounts are there for each pledge?
    • 2 at minimum, but allows for additional patron-set increases along the way
  • When do donation amounts change:
    • there is one (patron-set) flat change once the threshold is hit

Clarifying questions I’m wanting to confirm:
There’s only two different donation amounts (selected by the patrons): the base amount while the threshold is not reached and ??? the full amount (I’ll call it a target or max base pledge) when the threshold is reached? (and, if the patron chooses, they can change their base pledge along their way?

This is what needs specifying if I misunderstood

This would mean there’s only one pledge that directly changes all other pledges, yes?

Then there’s this

But people’s finances change and I don’t personally agree with not allowing people to lessen their pledge:
(that’s kinda like kicking people out for not making a good first estimation of their ability to pledge imo)
I get the idea though, having some sort of accountability to keep people in for more than a day, but maybe I’m misunderstanding when that certain amount of time kicks in: when they make their initial pledge or after the goal is reached?

Those are some initial thoughts - I wanna make sure I understand the proposed mechanism properly: thanks for contemplating this and posting!

3 Appreciations

Thanks for the many hours and thoughts that went into these many (many) characters! I look forward to discussing this tomorrow, along with the other mechanisms. I will second @alignwaivers’s hesitation regarding the “lock in”. It is resolving one of the biggest risks of some of these petition situations, pledgers claiming they will donate, but not having a credit card on file or taking whatever action they said they would. That said, the only way to really enforce it would be to collect the first month’s pledge ahead of time, putting us in the position of holding money, a thing we’ve been trying to avoid. Anyhoo, just some initial thoughts, thanks again and see you tomorrow!

Thanks a lot! It did take a lot of work.

Looks like I wasn’t clear enough on this point:
This mechanism supports multiple goals.

All of your other points seem to assume there’s only a base pledge and one goal, but in reality I think patrons would, after setting their sustain rate, sign up for multiple goals. They’re all somewhat a “stretch”, so you hope that at least one of them passes the threshold.

Projects can have only one goal, or even zero. When I said “crowds are forming around particular goals”, I meant any number! The goals are like separate projects and do not affect each other.

… up until that goal succeeds.

If by “each pledge” you mean each person, then there is only one amount (the base pledge) required. Subscriptions to any goals is totally optional. But you could sign up for a bunch!

Would you mind looking at the fake screenshot I built in Inkscape? I think it pretty well summarizes the experience from the perspective of a visitor.

“What if a patron changes their mind?”

Good point!

  • We say your pledge is locked in, and the buttons for lower pledges become disabled. However, if a person’s financial situation has changed, they can just contact us and we will take them off. We just don’t want to advertise this possibility, for obvious reasons.
  • The goal will include a timeline. Let’s say they want to hire a full-time developer for at least one year - you know in advance that you will be paying your chosen amount every month at least 12 times in a row. Therefore, you can know exactly how much you’re signing up for! (Of course, the hope is that most people won’t reduce their contributions even after a year is up, because they got what they asked for.)
  • When you pledge to a project on kickstarter, you can probably bail out a little while later, right? The only time you can’t is if the project actually hits the goal. Then you are locked in, because your contribution was critical to crossing the line!
  • Presumably pledges will not land “perfectly” on the target line, so the crowd total for a goal will almost always exceed the target in the end. This extra cash can help make up for any patrons that request a bailout.
  • If a user realizes they are getting uncomfortably close to their max spending limit across the site, due to a goal being achieved, they can simply and harmlessly eliminate their commitments to other projects and goals right in their dashboard. The priority is to fund achieved goals, not the potential ones. If the patron can’t afford to spend any more right now, they simply hold off on making any more promises.
  • If you participate in an auction on ebay, and come in second place, you lose nothing! But if you happen to come in first place, with the highest bid as the timer hits zero, you have to pay for the item. There’s no take-backs. You get used to that pretty quickly though, because you learn to only place bets that you’re willing to pay if you win. Ebay seems to do just fine with this, so I think we can too. They probably have systems in place for edge-case/rare exceptions though, so we could just copy them!
  • @Salt If a user repeatedly makes commitments and fails to pay, we can place limitations on their commitments going forward. Basically, anti-abuse controls!
  • @Salt After gaining some experience with the system, we learn that, say 3% of a goal’s budget (on average) tends to disappear due to pledgers bailing out. Perfect! From now on, goals succeed not when the pledges exceed the target, but when they exceed it by at least 3%. Problem solved!

Interactive Demo

Woof, that’s a short timespan to build an app, but I’ll try it!

2 Appreciations

I already pointed @tannerd at this mechanism conversation, so let’s not double up efforts, rather combine forces. That said, I’d rather not see either of you burn brain cells staying up late if the trade-off is not being as sharp tomorrow.

I appreciate the thought that went into this, and it’s viable in that it is a possible mechanism that could be used and doesn’t fundamentally fail in some way.

What you are proposing is Patreon + (sustaining version of) Kickstarter. The idea that funding can be a “choose your number” instead of a few tiers is minor detail. It’s just threshold campaign that pays out monthly instead of one-off.

Threshold campaigns do work, and adding the sustaining approach resolves some core problems with one-off accountability. I like your emphasis on actions (“hire a developer”) rather than speculative outcomes (“achieve feature X”).

Some details though:

  • You’re mistaken that this resolves the crowd vs dollars goal question. You just are choosing dollars.
  • The all-or-nothing style of thresholds still has some of the problems described at https://wiki.snowdrift.coop/about/threshold-systems:
    • The challenge of setting the threshold! It’s more consequential if it’s all-or-nothing. This puts a different level of pressure on the goal-setting decision than if the funding of partial-goals is happening.
    • The freeriding issue described there partly remains. Since people are already pledged at all, it’s less of an issue. But I could still think “eh, they will (or won’t) hit the threshold without me (or even with me), so I’m not going to bother pledging, I’ll see how it goes with others first.” Whereas with flexible crowdmatching, this doesn’t happen. If they goal won’t be hit, my pledge still has some impact.
  • But you have resolved many of the issues!
    • Not reaching goal no longer brands the project as a failure! For one, they have their foundational patronage anyway. Also, they don’t need a time-limit, so there’s no “failure” point.
    • Recruiting new patrons without reaching the threshold still adds some core foundational support for the project instead of nothing, so promotion of the funding is less risky (less chance of wasted effort).
    • The time-limit pressure is mostly fixed because the fact that all patrons are giving something means they (and their payment method) are active in the system. So, it’s more okay for the threshold goal to take a long time to succeed.

I’m mostly skeptical about the all-or-nothing thresholds because I think that goal setting is so speculative by nature. But that goes with my own resistance to doing some of the goal-setting work. Still, the idea of asking for an estimate, making the goal the max… projects that don’t want to risk setting high goals and not getting them will just claim lower (more frugally optimistic) estimates anyway.

I worry that tying the thresholds to specific actions could bind the projects too much. What if they decide that it makes more sense to have a half-time position instead of full time and use more funds to pay a different half-time project-manager? Can they adjust what they later want to do with their income? What a huge risk, if changing budgets results in losing a huge chunk of their funding.

What happens if they hit critical mass and then a small portion of funders drop? Suddenly they lose their funding and can no longer pay the new hire? Or they are supposed to be not having that employee anyway, since the crowd is no longer supporting that particular goal?

I’d much rather just add your idea of a unilateral foundation donation to crowdmatching

Your idea about a “set your base non-matched amount” is interesting. It could fully resolve some issues: the concerns about mixing donations across platforms, the concern about “half-assed” funding, the concern about patrons doing anti-matching (bumping their donation at first, and pulling it down as it gets matched more).

If we just add that foundation as a base underneath crowdmatching, it would have most of the important impact you are suggesting.

As for the goals, we could simply offer both all-or-nothing and flexible. The math would be the same for both, but the “all-or-nothing” goals would simply not have the extra funding turned-on until the goal is reached fully. That’s a concept I’ve suggested as possible for crowdmatching anyway.

That would be effectively Patreon + (sustaining version of) IndieGogo (instead of Kickstarter).

I’m much more into this idea because I think it’s actually a big burden and issue for projects to nail down exact goals and visions. I think flexibility is a big deal. The flexibility to get a bit more or a bit less and make the best dynamic decisions. For many projects, I’d rather they just keep working as best they can with whatever resources they get, and boxing them into overly concrete goals can be counterproductive. But I also think there really is something to strong concrete goals! So, I’d rather offer both.

In summary: I like the idea of base foundation of support. Add both all-or-nothing and flexible crowdmatching as options to add to the foundation. This still does not necessarily resolve the dollars vs crowd question because we could even be flexible enough to offer both types of goals.

One other concern: the snowdrift dilemma / freeriding at the foundation

The motivations for crowdmatching also includes the low-risk to current freeriders. The question is how to convert the most freeriders into patrons. Under your idea, such freeriders might still hesitate. It feels crappy emotionally to pledge 10¢ or something (I suppose a foundational amount less than worth charging amounts to something like $6/year instead of 50¢/month). So, imagine I’m a stingy freerider who is motivated by the idea that I don’t want to give unilaterally, but I’m willing to give if the other folks are too (especially if they give because I do). I don’t want to jump in and pledge a unilateral $5/mo foundation (especially not for a lot of projects, that adds up!), and it feels too shitty to give pennies as a foundation (I don’t want to say the project is only worth pennies). So, I hesitate and don’t pledge. I still sit and wait to see what others will do. And… you’ve lost me as a patron.

In short: the base foundation has the snowdrift dilemma and nothing to get over it. It becomes a barrier to making the crowdmatching pledge. Many people who are hesitant to donate until others do will pledge in plain crowdmatching but not if they have to put in a unilateral ongoing donation as a fee to access crowdmatching.

We could consider the foundational unilateral donation as an option without requiring it. There’s a degree to which I like pushing it. But I’m worried that it could drastically reduce the patron pool. At worst, it would reduce our current hope for orders of magnitude larger crowds and just drop it down to the miniscule crowds of current patrons (people who already give at Liberapay and Patreon etc).

I don’t think the 135 patrons currently pledged to Snowdrift would have all given foundational substantial pledges at this point. And I suspect that when we really announce, we could easily get thousands of patrons. I worry that would not happen with your system.

I worry that your system is aimed at targeting the existing community of donors, working to get them to give more, whereas our main emphasis otherwise has been on recruiting the latent donors, the many people who because of freeriding dilemmas do not pledge at all (like me for most of the projects I might want to support).

I want to pledge to crowdmatching for Inkscape. I do not actually want to donate even a monthly $1 to it unilaterally because I’m not up for that for every project I use and appreciate to the degree I do with Inkscape (hundreds of projects). I don’t want to say Inkscape isn’t worth $1 to me, that seems harsh. But somehow it is what it is without my $1, and it all just seems discouraging to even think about adding my $1. It won’t make a noticeable difference.

Put another way: I’m okay with giving pennies and seeing no noticeable difference. I’m not okay giving dollars and seeing no noticeable difference. But I dunno. I’d consider maybe an optional foundational $1 along with crowdmatching. I think the optional foundation might reach the widest crowd (since people have varying motivations). And we do have a goal of growing the widest possible crowds.

My position tonight: I’m sympathetic toward adding both unilateral foundation donations and all-or-nothing thresholds as options to crowdmatching. With those simple tweaks, we get all the benefits and flexibility of all the proposals.

I still see a lot of hypothesizing all around, thus a need for more real-world evidence about how the projects and patrons will actually behave.

2 Appreciations

Good point Salt.
I’m going to bed, but I got something working!

Here’s what I have so far.

That was the fastest I’ve ever built a working program of that depth, actually. Woo!

2 Appreciations

Yes! I said exactly that, above.
The fact that it can be seen as a combination of existing, tried and tested systems, is an advantage for both efficacy and for explaining to people!

I didn’t claim the question is “resolved”, but yes I chose dollars and I explained why dollars makes more sense for this approach. This system is all about concrete numbers where possible.

I actually deliberately address every one of the points on that page :slight_smile: I did lots of reading before proposing this lol

This can a challenge, sure. I think that is worth the lack of discomfort zone!

True, it still has some impact… and if there’s no growth for many months, it still has that same some impact… over and over. But if you knew for sure it would never reach the goal, would you still pledge that amount? No, you’d probably give the full amount you want to contribute (possible in this system via base pledge) or you would decide not to give to that goal at all. No one would say “I want to give 46% of what I think is the right amount… indefinitely.”, which is why I called that the half-assed funding problem.

Yes, I’m very excited about that!

Hmm, I wouldn’t call “not getting them” (achieving a goal) a “risk”. Or at least it’s nothing compared to the risk of taking lots of money from your patrons with expectations - and then letting them down. That’s why super small stepping-stone goals are encouraged!

You’re right, but we’d be in good company (aka along with most of the other crowdfunding sites) AND you must consider that monthly costs have the potential to be far less speculative than one-time projects.

  • If the goal was not yet achieved, they can just update it to the new plan!
  • Or they can add that as a competing goal, and let the community decide!
  • If the goal was already achieved, why was this a good idea yesterday but suddenly it’s not? Best to do what you said you were gonna do.
  • That said, projects manage their reputation on their own, so if they convince the community that they made the right decision, who cares!

Yup! Pre-lockin, anything can change. In fact the project itself is in charge of triggering the lock in. So I see very little room for excuses.

Good objection, but it was already raised above, see my reply where I tag Salt. We can have a safety buffer, above the goal amount.

Well, as soon as you put partial funding back in, half-assed funding is possible again.

Interesting suggestions!

Let me be clear - projects are free to have wishy-washy goals! I just predict those goals are a harder sell. But there is no reason you can’t have a goal that says “for whatever we decide” if you want.
But again, if you’d “rather they just keep working as best they can with whatever resources they get”, then the base sustaining pledge very well may be able to do the trick without goals!

Would love some concrete examples of that. Perhaps it’s more common than I think. Perhaps it’s less common than you think.

I think for this system we can actually stick with whole-dollar amounts, since there’s no fancy math going on anyway. Easier for folks to talk about than cents.

Simple, that’s the sort of person who would participate in the goal games and not the base pledge. They get to freeride while the project stays as-is, which they were doing anyway. But if one of their wishes comes true, they’re no longer freeriding! And they can’t complain because they’re getting more than they were getting when they were freeriding. By exactly the amount they think that difference is worth!

Not sure why you’ve boxed this guy in to only using the base pledge. The proliferation of wishful goals is very much a free-rider conversion strategy!

…The base pledge can be optional. Otherwise not sure what you’re saying here.

There is no snowdrift in the road blocking the project from existing as is - it already exists as is. Not sure what barrier you mean.

That’s fine - it can stay at $0 until your first goal succeeds and bumps it up!

That doesn’t seem logical, even if a $1 minimum was required. But maybe I’m missing something.

No one said they have to be substantial! Also, this system trades insubstantial amounts, for substantial amounts of hypothetical money (with strong rewards attached). That’s the incentive. (See the lottery story, second post in this thread.)

I don’t see that at all. We’re not even necessarily trying to get existing donors to give more - if they paid $5/mo on Liberapay they can come here, set their base to $5/mo and be done with it. I am, however, trying to get freeriders to give more!

You seem to have a conflict between your favorite audiences: Are you targeting freeriders or not? Because you’re mainly talking about the “minimizing losses” contributions here, and that is not the kind of contribution that freeriders would suddenly want to give. But there is not much of a snowdrift dilemma at that stage, since every little bit helps.

I thought you were a fan of minimizing losses wherever possible. Hmm.

In other words: dollars is your discomfort zone!! You hit the nail on the head.

1 Appreciation

Thanks for the proposal!

Some thoughts:

  1. You can define flexible goals on Patreon. So the difference is that this has all-or-nothing goals like on Kickstarter. Does that really solve the issue?
  2. Locking people in reminds me of the internet contracts in germany that are mostly for 2 or 1 year. You have to pay even when you moved and can’t use it at the new address. That’s very inflexible. I want to be able to decide every month if i want to cancel a service.
  3. When funding is limited to say 1 year, it’s also bad for the developers you want to hire. They want to have a perspective to work for as long as they want, like in a normal company. You shouldn’t have to create a new goal for that every year and hope people support it again. Imagine projects at the scale of Mozilla.
  4. Having clear goals would be perfect, but i guess it’s too much to ask from many projects. On the other hand, “hire another developer” is a simple and easy goal that projects (like Godot Engine) already use. And that’s what most software projects use funds for.
  5. I also want a solution to be simple, so everyone can understand it. Having a base pledge, multiple flexible and all-or-nothing goals and maybe also crowdmatching is a lot to (potentially not) understand.
3 Appreciations

I fundamentally disagree with this entire premise. I do not have a conscious set amount I think is “right” or that I’m willing to give. I don’t know what’s fair. I find deciding this to be a source of hesitation that causes me to not donate. I don’t care if it’s 46% of a goal that is somewhat arbitrary anyway. I want to know that I am giving with others and that my pledge is a matching offer to others to encourage them. I’m happy to give more because others are joining and not happy to give more without that.

The half-assed issue is not a problem for me. If it’s half-assed because others aren’t joining, then it’s not costing me much anyway. I could give up on the project or not (and I don’t mind staying half-assed, I really don’t), but I don’t feel like just donating more on my own.

The question is whether there actually is a threshold where money less than that isn’t helpful. In that case, yes, getting partial funding is not good. But if some money is positive and will help, then there just isn’t this threshold point. The stress of giving results exists at the threshold anyway. People don’t expect full results with partial funding anyway.

Mixed feelings here. This is good in some ways, but it veers toward bounties and restrictions on pools of money only available for certain budgets. I think this risks micromanaging / limitations. I prefer the general idea that patrons can just express priorities but it’s not a hard restriction on money for specific directions.

Sure, but then it’s weird to awkwardly build flexibility that the all-or-nothing is taking away, i.e. just multiple whatever goal points.

The point has been missed here: pledging $0 is like a public statement registering as a freerider! It’s very different than passive freeriding, and the vast majority of hesitant freeriders will, I suspect, feel uncomfortable doing this. They will keep passively freeriding while a portion decide to put in a substantial amount.

Again, this is misunderstanding. I don’t have such a fixed comfort zone individually. I have it socially. My comfort is related to my sense that I’m part of a crowd encouraging others. My comfort donating increases with my sense that I have a crowd with me and matching me. My discomfort with dollars is the unilateral part of it. And I don’t find the all-or-nothing approach inspiring either. I want to just tell the world that I’m in and that my comfort level grows with the crowd, and specifically when the crowd is all doing crowdmatching (so I know my donation is part of that and I can’t just freeride and have the rest of the crowd keep giving the same).

There’s a fundamentally social dynamic here that the emphasis you’ve given here is somewhat lacking. It misses the deeper social dynamic which is present in flexible crowdmatching.

3 Appreciations

@Adroit Supposing you seriously consider this proposal, I’d love see this mechanism outlined like the proposal found in the mechanism repo: Snowdrift.coop / Mechanism · GitLab.

I know it takes a lot of time to condense everything to just its moving parts, but it would be helpful to have a single place store (and maintain) and compare mechanism objectively and without discussion.

1 Appreciation