Question about getting cut out at budget limit

Hi,

I just watched the video (it’s beautiful) and clicked around a bit. There’s one point that surprised me in the presentation: the graphic seems to show that once the donation level for the project I am supporting reaches my limit, I am “bowled out”. Is that true? Surely if someone has been paying regularly every month, you want to keep that person preciously and gratify them?

1 Appreciation

tl;dr: Yes, the system suspends a pledge when you hit your limit. But it’s just a control-point. We hope you’ll feel it’s worth it to increase your limit. If you don’t, maybe it’s fine to just let others donate without you anyway. Regardless, this is far off. Our goal today is to get to where patrons hitting limits is the problem!

We do need to keep working to present the budget context better. The sole purpose is to give patrons a guaranteed control point to not spend more money than they are willing.

In some sense, the ideal would be that you never hit your limit — not because the crowdmatching doesn’t grow but because you keep increasing the limit as you feel more and more that it’s worth it to stay in.

When would you not increase your limit? One case is that you don’t feel some projects need or deserve that much funding, in which case it makes sense to drop out anyway. Another case is that you just truly can’t afford to donate any more. In that case, we indeed do not like the idea of you getting cut out.

If we let you stay in but have no more increase from crowdmatching, it would undermine the mutual assurance. It would be dishonest to say to new patrons that everyone else is helping match them if actually many patrons were just staying at a limit and not matching.

Some of many potential ways we can keep everyone matching but also let you could stay on when you can’t afford to donate any more:

  • Project splits into sub-projects so you can support a lower-budget smaller sub-project
  • We implement an option such as less-frequent donating (bimonthly, quarterly, annually)

But you might also just be happy that other people are donating so much (it must be happening for you to have hit a limit), and you get to enjoy the results without spending any more money yourself.

See more discussion of this issue at https://wiki.snowdrift.coop/about/limits

Incidentally, we could potentially offer an alternative funding approach. As a different option from authorizing a monthly budget, we could let patrons authorize a one-time larger amount (e.g. $50) and allow monthly crowdmatching to keep going until they’ve donated that total, at which point they’d have to make another authorization to continue. In some sense, the monthly budget is like that but where it renews automatically and doesn’t roll-over the remainder each month.

1 Appreciation

A post was split to a new topic: Crowdmatching implementations, charging up-front?

This concerns me also. If most people have a $10 dollar limit, and then someone pledges over that, suddenly most of the patrons are not allowed to give anymore. Oops. But wait now they can give again because the pledge is back down to a dollar. So on and so forth. This is not a good loop. The people with a limit should be kept at their limit. You can still advertise “if you pledge so much others will pledge this much more”, it just won’t be everybody that will be able to match.

I don’t think this is a minor problem. There needs to be a real answer here about what’s going to happen. Just dropping people is a big mistake.

Welcome qrpnxz! This was 3 years ago, and most of our current visions for the crowdmatching algorithm do not work quite like the video anymore.

The loop you mention is indeed a potential race condition if the algorithm was implemented in a “drop everyone at once” fashion. However, if we have a method for prioritizing who to drop first (perhaps those pledged to the most other projects), the dropping can iterate down the list until the matching formula no longer has any remaining patrons coming up short. Then it would stop there - no unstable or looping behavior necessary. If, say, all 100 people have a $10 limit, and we’ve hit that limit, dropping a single patron would bring the matchers to $0.99, under their limit. No need to drop everyone.

More realistically there would be a variety of different limits, with no particular limit being dominant in the crowd.

They can give again, but nothing would happen automatically, a new patron would need to join for anything to change.

That’s a good idea too; I think the concern was that it wouldn’t be true matching sometimes, and we want to be able to confidently say that crowd size always increases pledge, guaranteed.

I invite you to take a look at the myriad other threads on this forum about other proposed algorithms - many of them don’t involve dropping at all.

Yes, to reiterate what @Adroit said, we’re not going ahead with the pledge-dropping approach any more. The new approach is simply to have the project have a threshold point at which matching is turned off for everyone (but pledges are not dropped), and then we can have the option of people further pledging toward stretch goals and so on, understanding that any matching at that point is limited to the portion of the crowd that opts-in.

This new approach solves the whole issue. Nobody gets dropped, budgets exist, the premise of matching holds up clearly and transparently, and there’s still the option to encourage further donations and matching as appropriate.

It’s not worth further discussion about why pledge-dropping might not be as bad as you suggest (we wanted to encourage the budget as a failsafe, get people to think of it that way, and get people to choose to increase their budget because they think the further growth in the project is worth it and so on). But the whole point was that we were stuck with some definite downsides and common misunderstandings or concerns like yours, and it just was not worth keeping that budgeting method. A successful approach has to gel with people’s understanding, and we accepted that the angle we originally wanted people to take to the budgeting wasn’t something we could reliably achieve. The updated approach is just cleaner, better, and more socially easy to accept and explain.

Oooh, it’s project set now. That’s why I cannot change it. Yeah, I like it. It sets a concrete goal to reach that matchers can commit to, then there can be optional traditional contribution on top of that. I see.

You should add the budget of the project on the page of the project. On the dashboard it’s also displayed as my budget (is it the sum of all the pledged projects’ budgets?) rather being associated with any project. I think that would clear things up unless I have misunderstood again how it works.

Although we’re not absolutely settled on the various approaches, the idea isn’t traditional contribution, the idea is to have continued crowdmatching be opt-in (such as to a new higher threshold). We still want crowdmatching but don’t want confusion about who is in the crowd matching each other. So, this opt-in would mean that those who aren’t opted-in past the original threshold are not doing further matching. Of course, there’s no reason for further matching if the project actually needs no further funding, but that’s not likely the case most times.

Yes, that makes sense. I’ll see about fixing that ASAP. Unfortunately, fixing the site in general has been almost on hold since we made a decision to change to a new front-end to make development easier and yet that hasn’t been completed and in practice development is stalled. This situation is no good and is a real threat to our goals and to understanding. I hope this gets resolved soon. Meanwhile, it’s possible for me to tweak the existing site still.

2 Appreciations