allow for one-off tips

[chadwhitacre]

The essence of Gittip is recurring tips, but once we solidify that I'm open to mixing in one-offs.

The reason recurring tips are important is because my mortgage is recurring and Gittip is about paying my mortgage.

There is a $80 open bounty on this issue. Add to the bounty at Bountysource.

[chadwhitacre]

@isaacs

So, basically, your answer is that yes, the goal of gittip is to be a bank, even though it is not currently, and yes, gittip will be investing cash holdings and operating with a smaller cash holding than their current liability. (Where do the profits of said investments go? Who eats the losses? Will you be FDIC insured?)

These are undecided. Along with #938, #279 is still open, and I've cross-posted your questions there. ;-)

And, since gittip will offer a 0% interest rate, explain to me how this is not just a much worse checking account?

Sorry ... where has it been decided that Gittip will offer a 0% interest rate?

This decision makes me trust gittip a lot less.

From what I've been able to tell, you didn't really trust it in the first place. :-)

[chadwhitacre]

@isaacs Are you right now in the habit of sending $1,000 payments to developers you appreciate as rewards for exceptional performance? I'm trying to gauge where you're coming from.

[isaacs]

@whit537 I complained on twitter immediately after trying to do this, failing, and then sending the developer a check for $1000, yes. I got some money from a contract to deliver some functionality, and he was instrumental in helping get it done, so I thought it fair to split the proceeds (since we both have regular jobs, and so the money is just a bonus anyway). It's not the first time I've done such a thing, and I'm sure it won't be the last.

As I've said before, if you want to actually replace developer paychecks, you can't be scared off by large transactional payments. They're a part of life.

[chadwhitacre]

@isaacs

But still, the answer seems clear to me: Gittip is served, and they are served by increasing their cash holdings without liability. I'm not arguing that it's just a bad idea. I'm arguing that it's duplicitous and morally wrong. If you're holding money in escrow in order to invest it and pay the maintainers, fine. Say that.

It sounds like we have a trust problem, and that you are suspicious of me and Gittip, like, "Chad is trying to fool everyone into escrowing money with him so he can secretly profit off of it!" :-(

[isaacs]

And, since gittip will offer a 0% interest rate, explain to me how this is not just a much worse checking account?

Sorry ... where has it been decided that Gittip will offer a 0% interest rate?

Fair. That is an assumption I've made. What will the interest rate be on the money that gittip holds in escrow? If it's not out-performing my own investments, then I'd still prefer to just get the money out as soon as possible.

[dstufft]

@whit537 I don't think it's so much thinking that you're trying to do that, but where money and other high value things are concerned minimizing risks (besides those taken to attempt to gain profit) is just good sense.

[isaacs]

It sounds like we have a trust problem, and that you are suspicious of me and Gittip, like, "Chad is trying to fool everyone into escrowing money with him so he can secretly profit off of it!" :-(

Well, @whit537, I trust you personally, though I don't know you that well. You seem like a really stand-up guy. I just think that you're suggesting setting up a bad system, and it's a mistake that will cause harm eventually.

In the long run, I don't trust any for-profit corporation to do anything but maximize profits. I also don't trust any corporation to be smart forever. So, if you set up a system where the incentives screw us, and our only protection is your upstanding morality and wise insight, then when gittip is purchased by BigDumbEvilCorp, we're all screwed.

If gittip is a not for profit foundation with a specific mission statement or whatever, then that's a bit different. Not for profits are still not necessarily trustworthy, and you definitely have to take a hard look at their mission and how they are run in order to guess what they'll do. In that way, for-profit corps can be much simpler to interact with, since their behavior is easier to predict.

As much as you might say so, and it might actually be true right now, gittip is not Chad Whitacre, and it's not "all of us". We're users, gittip is a company, and you are its leader (for now). If you want to Do Good here, plan for what happens when you turn evil.

[chadwhitacre]

I complained on twitter immediately after trying to do this, failing, and then sending the developer a check for $1000, yes. I got some money from a contract to deliver some functionality, and he was instrumental in helping get it done, so I thought it fair to split the proceeds (since we both have regular jobs, and so the money is just a bonus anyway). It's not the first time I've done such a thing, and I'm sure it won't be the last.

Okay, cool. Thanks for the backstory. :-)

As I've said before, if you want to actually replace developer paychecks, you can't be scared off by large transactional payments. They're a part of life.

I'm fine with large payments, but transactional relationships are not what Gittip is fundamentally about. Gittip is about on-going, no-strings-attached gifts to people and teams you believe in. Gittip is more than a "payment broker." Use PayPal for that, or write a check.

What will the interest rate be on the money that gittip holds in escrow? If it's not out-performing my own investments, then I'd still prefer to just get the money out as soon as possible.

This is a valid point, and I don't have an answer right now.

[chadwhitacre]

We're users, gittip is a company, and you are its leader (for now). If you want to Do Good here, plan for what happens when you turn evil.

The plan is to fork.

[dstufft]

Forking a community is hard work and almost never leaves the situation better except after a lot of pain. I hope there's a better plan.

[chadwhitacre]

I haven't felt a huge need to make Gittip a non-profit because you're right: it ultimately does come down to trustworthy leadership, and that can't be gotten through regulation, only through reputation. To expand on my previous comment, the plan is to keep Gittip as open as possible, so that when it goes sour (not if), it will be as easy as possible to fork under new, trustworthy leadership.

[chadwhitacre]

@dstufft The situation @isaacs described is the worst-case scenario, when Gittip rots from the inside out. Of course let's try to run it as a good project for as long as we can.

[dstufft]

I understand that it's the worst case scenario, and it's possible that forking is the only thing at that point, but that doesn't mean you can't strive to limit risk otherwise. So I guess my statement is better reflected in "I hope there's more to the plan than just Fork if it goes sour".

[chadwhitacre]

@dstufft The plan is to lead Gittip as best I can, build a strong community, and eventually chose a successor, unless we revert to some form of democracy sooner rather than later.

[mvdkleijn]

wow... I'm away for a short time and this exploded... I'm thinking I should stay away from this discussion after this last comment. This is starting to feel more like an argument rather than a discussion.

My position is clear (I think) and I believe I've answered @isaacs questions on why a receiver would want to opt for a spread out payment. I'd be such a receiver and I would, for my own reasons, opt for spread payouts. Whether anyone else feels my reasons are valid or even sane, I don't care. They're my reasons. Just like you don't want Gittip to hold people's hands, I don't want you to hold or in this case force my hand.

I've already proposed a solution that would allow RECEIVERS to make the choice themselves if they want spread payouts or single lump sums. I do believe the ’bank rush’ example of @whit537 is not a valid scenario and I believe we should not use a fixed term of 10 weeks (see my earlier proposal) but I do believe optional spread payouts are a good thing. Not everyone is in the USA, so circumstances differ as does what makes economic sense.

So to recap, my proposal / view is:

• allow one time donations
• allow RECEIVERS to choose if they prefer one time lump sums or spread payouts

What Gittip does with any money left in the RECEIVER’s balance is another discussion. (provide interest, etc)

[chadwhitacre]

+1 from @daclements via Twitter.

[3Phen]

For what it's worth, I just discovered Gittip thusly:

There's a project I found about a month ago, I fell in love with it, started interacting with it frequently. I discovered that its developer has other, bigger projects, and the thing I love is usually just a fun side-thing taking him away from his more important work. Out of appreciation, I feel like donating to support the project. I'm a college student without much income, I try to avoid recurring expenses whenever possible to leave my budget with wiggle room -- what I had in mind was giving 15-30 dollars, and then maybe month from now deciding whether I wanted to donate again. I ask, "Will there eventually be a way to donate to this project?" he answers essentially "There's this thing called Gittip, I haven't used it before but we can try it." So I head to gittip.com, and discover that I can't donate money without dedicating myself to weekly payment thing. So now I'm back to square one, and I'm feeling kind of awkward going back to the guy and saying "hey, well, I'd still like to donate but this site you mentioned is only for weekly things, and I don't want to give you money weekly".

It may be that my intended use just clashes entirely with what gittip wants to be. But it feels close enough that it's odd to just turn back -- I ended up reading this entire thread trying to wrap my head around the misalignment. If gittip even offered a monthly option instead of weekly, I might use it.

I suppose I could divide my intended payment in four, set it up, and hope I remember to come back eventually so I don't look at my bank's website half a year from now and realize I'm still spending money where I shouldn't be. But that still feels weird. Idunno. I hope this testimonial was useful to you.

[waldyrious]

The above comment by @3Phen shows clearly (to me) that as Gittip is trying to fit the needs of the receiving end (i.e. a stable income flux) it ends up alienating a lot of the giving end, due to a somewhat inflexible model that doesn't allow a lot of variation in the profile of the giver, both in material conditions and in motivations for donating.

My question is, must we be fundamentalists (in the literal sense; no offense intended) in the "ideology" Gittip will stand for, or should we admit some compromise if it advances the cause of providing more income to developers and increasing the popularity of donation-based income?

[mvdkleijn]

@3Phen @waldir Guys, I don't believe you've read through the discussion so far. There is no fundamentalism going on here.

If you read through it, I believe you'll find that after a healthy discussion on pros and cons, most people already agree that one-time donations are a good thing (including me) and should be allowed.

The only part that was still under discussion was exactly how to implement it in such a way that both the funders AND receivers wishes are respected. See my proposal earlier in the discussion.

(on which I have received no feedback from the other participants by the way :tongue: )

[3Phen]

@mvdkleijn I really did read through this entire page! I was under the impression that most people agree that one-time donations are a good thing, but that @whit537 himself is still kinda reluctant, because he sees them as antiethical to Gittip's purpose. If I'm reading his posts correctly, it looks like one-time donations might only be available to people who already have a significant amount of money flowing through Gittip weekly.

It strikes me that he kinda fears that if one-time donations are allowed, they'll end up supplanting recurring payments entirely, negating the entire purpose of the site. This honestly is a reasonable concern! It's a matter of how narrowly focused Whitacre wants this service to be, which I think is what @waldir meant by "fundamentalism" – how far will Gittip stray from its fundamental concept?

I was just posting from the perspective of someone who came here to do something and was told by the interface (then later, the FAQ) that he couldn't. My intention was to give @whit537 an illustration of one sort of person who'd be using this feature (sort of on the opposite end from the person looking to make a $1k transfer).

[mvdkleijn]

Ok, understood. The concern about recurring payments being supplanted by one-time donations is a very real one indeed. In fact, if that would occur we might as well not spend any time on Gittip at all. The whole reason for existence of Gittip is the recurrent aspect. If we lose that aspect, Gittip is not much better than say PayPal, Flattr, or any other one-time donation platform.

That is why the way we implement one-time donations is so important.

[waldyrious]

I knew I shouldn't have used that word :P even with the disclaimer, it rubbed the wrong way. Indeed I have read every single comment in this thread, and I'm happy to confirm that I too think the discussion has been a healthy one full of good points for both sides. Thanks @3Phen for clarifying what I mean by "fundamentalism".

I don't claim that a more flexible approach would definitely work for Gittip; there's a risk involved, of course, but I don't think we should avoid that risk at all cost. Especially the cost of enabling Gittip to eventually "make it". My concern is that by narrowing the focus of Gittip we may be keeping a lot of people out of the loop and potentially undermining the very ability of Gittip's model to get critical mass.

Note that Flattr does support recurrent donations and in fact I have several of those set up. The core difference Gittip offers isn't the recurrence, but rather the focus on the receiver (a stable monthly income, as opposed to a stable budget for the givers' donations), the focus on the person rather than their work, and the openness of the platform.

It might be pertinent to query Flattr regarding the consistency (or lack thereof) of the donations. That will likely enable a more informed discussion about whether non-recurring donations (which were the only kind Flattr initially supported, actually) really do undermine the ability of a system to provide stable income through donations, rather than relying on opinions, predictions, hopes and fears.

[rgieseke]

Why not try adding another selector after the amount with options for duration?

[$0,25]/week [duration]

• until you cancel or change
• for 1 year
• 6 month
• 12 weeks
• once

(Or freely selectable like the amount).

Many charities have similar options, so it might work. In a way this could also increase stability of donations.

[balupton]

One thing about a selector after is that it means that we can only do one or the other, recurrent or one-off. Which eliminates the use case of me already supporting someone I like, but then they do something super awesome, and I want to give them a bit more once-off. It may also deter people away who only want to do a one off payment, but it is hidden away in a menu.

[jdorfman]

I like the idea @rgieseke had. I think it should be available to those who give over a certain amount recurring, to encourage more users to give back more money. At the same time, being in the web service industry I can play devils advocate and understand that this might cause more fraud, headaches and technical debt that Chad wants to deal with at this stage. At the end of the day it is his decision, and I stand by him 100%.

[mvdkleijn]

Whatever the final decision will become and whoever makes it, I believe respecting the receiver's wishes is important. In the end, the funder is there for the receiver. I understand that the funder wants to have options, wants to feel good, etc. That's fine, but let's not do that at the cost of choice for the receiver who you're trying to give a monetary "thank you".

[balupton]

Just received a $5 once-off thankyou via paypal, and was shocked to see this:

Total amount: $5.00 AUD
Fee amount: -$0.47 AUD
Net amount: $4.53 AUD

That's a fair fee amount that paypal places. I'd imagine that Gittip would have far less fees, as the way they earn money is from donations to Gittip, rather than commissions? In which case this would be a huge win for one off donations.

Reading https://www.gittip.com/about/ it seems to imply that no give money fees are done, only take money fees. However there is a $10 min, I think that should be fine for one off donations, it would help them give to other people with their remaining dough, explore gittip more, and perhaps setup a recurring donation of a smaller amount - that could even be in the workflow of a one off donation "great, I'm sure they'll appreciate that donation, now that's out of the way, what about donating a smaller amount each week to help them get by? What about 25c? Cheers!"

[ghost]

An example of excellent UX for accommodating one-time and recurring donations, as well as organizational support: https://secure.actblue.com/contribute/page/sciortinomain

[chadwhitacre]

+1 from @Sarcanon via Twitter.

[chadwhitacre]

+1 from @tarekziade on #1641.

[aausch]

I'm under the impression that the vast majority of open source contributors do one off, or rare contributions. I support being able to do one off tips, to reflect that.

[chadwhitacre]

+1 from David Chartier via support@gittip.com.

[chadwhitacre]

+1 from @alimac on Twitter.

[nicksergeant]

Homebrew considered using Gittip for donations but didn't because of no one-off tipping:

https://github.com/Homebrew/homebrew/issues/20482#issuecomment-19738328

[chadwhitacre]

+1 from @ckdarby on Twitter.

[mgifford]

Related discussion about incorporation of GitTip on drupal.org -> https://drupal.org/node/2138397

[chadwhitacre]

+1 from @zspencer on Twitter.

[zspencer]

First and foremost, I agree gittip's primary design focus of strongly encouraged recurring tips is awesome. I see it as an incredibly powerful force for good in the tech community and has enabled people to reach financial independence who would otherwise have to work a day job to do what they love.

I look forward to seeing how you design a solution that allows individual tips without hurting the core statement

Zee

[gJigsaw]

Between the false dichotomy of allow or omit one-offs, I choose omit. (-1) If we want one-offs, I'd +1 a function that minimized income variance by altering dispersion time - rather than a blanket time of x weeks. Lastly, I'd +1 giving each recipient full control over their own escrow function.

[aknuds1]

+1

[syrsly]

+1, however...

...I want options. Lots of options. I want to choose between "Allow all payment types", "Allow only once-only payments", and "Allow only recurring payments". I also want the end user, the person giving the tips to be able to choose between giving recurring payments and giving a one-time tip.

[abnor]

+1 @joshmaines

I like options. I don't see how letting the user(s) decide how they want to receive their tips does not interfere with the core concept of sustainable tips. Since people can discontinue tipping other people anyway, and international users don't yet even have a recurring payment method, I don't see allowing one-off tips, as an opt-in, disrupting the core concept. The core concept remains, and peripheral options are included.

[abnor]

There are (51) +1's on here and only (5) -1's on here.

[alimac]

@whit537 already +1'd this on my behalf, but I'd like to add my reasons. At present I have a job and I am comfortable with recurring payments, subscriptions, etc. But I remember a time when income was uncertain and any kind of recurring payment scared the crap out of me. It's true you can set it up and cancel it in a month or whatnot, but for some people this is an extra step that they would have to remember and there is the potential that they forget, which might discourage them. One time payment would put me at a peace of mind.

[syrsly]

@alimac +1, because I try to avoid recurring payments even when I have a steady income, which is admittedly not often. I don't even mind going back to GitTip each month to make new tips, because it means I have greater control over my funds. I know I can cancel subscriptions and all that shiz, but why should I have to mess with recurring payments at all? What if I lose my internet connection and am unable to cancel? What if I don't realize it's time for the payment and forget to cancel? So many things can go wrong with recurring payments that I would prefer to avoid them entirely.

[pluma]

@whit537 I think for me personally having the option for one-off tips would result in me making larger tips more frequently to more authors. As a freelance dev myself I already don't have a stable regular income and therefore am excessively careful about setting up recurring tips -- the overhead of having to adjust or cancel the tips is just too great (i.e. I'm just too damn lazy).

As it is, I currently only tip a tiny handful of authors who I know actively publish new open source projects semi-regularly and I only set up a tip that is low enough that I am confident I won't notice it if I don't have any income for a couple of weeks (or months).

I'm aware that Gittip aims for long-term support of open source authors, but if you think one-off tipping will lead to an unreliable income I would say that is already the case with the ability to adjust and cancel regular tips at a whim (which IMO also sends the wrong message: there's no way to tell whether a tip is intended as a one-time thing or as long-term support). The only way around that would be to force tippers to commit to give N recurring tips of value X (like a long-term subscription model).

I don't see how a "less stable" but more realistic flow of income would be detrimental to the idea of creating a sustainable income for open source developers. Anyone who's doing contract work or gaining revenue from side projects is already used to working with unstable income flows.

It's been two years and an order of magnitude more +1's than -1's. I'd say the community is largely in favour of this idea.

Besides, it may be about paying your mortgage for you, but I think most open source devs would be happy if it was merely a small supplement to their regular income (allowing them to take time off to work on open source projects without sacrificing income from their day job or taking time away that they could spend with their family). The full 100% "open source stipend" is likely only something a minute fraction of the community will have a chance of experiencing (either because they have so many supporters or because their cost of living is that low) -- it won't be the default case for a long while, no matter how much we would want it to be.

[chadwhitacre]

+1 from Mike Borek in private email.

[chadwhitacre]

+1 from @calendee on Twitter.

[rtyler]

I'll echo some of @pluma's statements. one-off tipping would be a great way to do a sort of nod of appreciation to a number of hackers. IMHO Flattr has a reasonable model for this, but has more or less failed to embrace the GitHub community the way Gittip does.

[carthik]

Would be nice to be able to tip someone a lumpsum amount and have it be disbursed over a few weeks. Some folks who want to give might have incomes that vary drastically, and also recurring payments make me inherently uncomfortable. I would totally get an option to fun one's donor account one-time and continue to pay out from it as long as it lasts, without linking it to a funding source, too.