First: I really like the idea of Gittip.com. Congratulations on getting this far.
But I strongly object to the tip amounts. Without going into too much detail, $0.25/week is not something I'd really consider a tip for a barista making my coffee, my dry cleaner, or any of the other myriad service folks I interact with on a regular basis. $0.25 will get you about 10 minutes of on-street parking in many cities; it might take longer to get your coffee. I'd guess that my attorney, physician, or my lawn guy would not consider $0.25/week a genuine "thank you" gesture as much as "hey, I have this quarter in my pocket." Personally, I think programmers are worth a lot more than most programmers think they themselves are worth.
Every other day, there's a HN post about some clueless business guy looking for technical co-founders. The HN community rages about how business people always undervalue their technical contributions. And then here's a project where programmers will tell you that a good project is worth $0.25/week. It that's true, maybe the "clueless" business guys are right to devalue technical work.
So while I really like your idea, I really do wish you'd reconsider the worth of programmers who contribute to these projects. I'm trying to be as constructive as possible here, because I think good things can come from what you're doing if you don't advance the notion that good programming in any meaningful quantity is worth $0.25.
Note that we already raised the rates significantly (max used to be $1.28 instead of $24), and I'm personally looking for $2,000 per week. So I don't think Gittip understates the value of good programmers.
Here's the thinking behind the current tip amounts:
I should add that I did read that before posting. And that's part of my point: you seem to have developed the tip amount framework without any regard to the larger market. My point is that the fact that many people gladly tip their bartenders > $1/week is relevant to your project. In the real world, $0.25 is not a tip, it's pocket change. (Here's a list of suggested tips, note the conspicuous lack of tips under a dollar: http://www.getrichslowly.org/blog/2006/10/12/basic-tips-on-t...)
The other gripe I have is that you match expected tips for a "top open-source programmer" with an average salary for a "mid-career engineer". Why shouldn't a top open-source programmer be able to make top engineer money (read: a lot more that $160k) with Gittip? And why is there a max tip amount? What if my business wants to "tip" someone more than $100/mo? (I know, it's insane to think that code would be worth that much but to some it may be.)
So I guess I would ask you to reconsider:
1) The signaling issues around the idea that a quarter or a dollar is a reasonable amount to contribute for good programming. How do you explain to non-techies why your work is worth so much more than that?
2) The expected top end of the range for super-successful devs probably shouldn't be in the range of what recent grads are making at top firms.
Again, I'm trying to be constructive, and I think that if you want to meet your goals of helping programmers find a living building open source software, you could do a lot worse than asking folks who choose to tip to do so in amounts that are commensurate with the education, training, and thought required to build useful open source software.
The rejoinder would be that, unlike my bartender, I could conceivably have 100,000 people tip me $0.25 through Gittip. When you compare $0.25 to a Facebook "Like", it's a lot more valuable.
That said, I appreciate people who want to make even more on Gittip than I do. Go for it! :-)
>> unlike my bartender, I could conceivably have 100,000 people tip me $0.25 through Gittip.
And if the minimum were $1 or $5, I don't see how that's a problem. Your bartender doesn't have global reach and his/her services don't have infinite scalability. The bartender/dry cleaner/barista examples were to illustrate lower bounds and not the maximum.
There's no reason to place artificial ceilings on how much developers can make with something like Gittip.
You're artificially limiting the maximum here. The range of contributions for e.g. a game will be different from B2B software, because the values derived are completely different. There doesn't seem to be a good reason to enforce a global limit.
If someone is willing to pay a large amount of money to influence a programmer, they'll be willing to manually negotiate / transfer it (rather than doing it through gittip).
I think all that is needed is a link that says... want to tip more? then donate directly to programmer by mailing him. This will remind the donor that he probably could give more... see how many people opt for this and figure out the nature of the donations.
That way, the service remains simple and manageable and won't end up being a money laundering operation for some terrorist.
Hey Chad, I wonder if Apple would allow a Gittip app using in-app purchases? They just released subscriptions too. That way every downloading user would already have his CC on file, and be a click away to donate.
The idea is that the volume makes up for the tax - 100% CCs on file vs 4% is huge. Not the first thing you'd think of for supporting open-source, but something to consider.
Consider using Dwolla.com instead of credit cards. No transaction fees at all for < $10. And it's run by a brilliant entrepreneur determined to disrupt the credit card industry. Cutting edge stuff.
On the other hand, it adds up to $13/year, which is more than many commercial services charge for a subscription or an application download. It's a matter of perspective. But I'm with you on disliking the $0.25 amount.
I think it would even be better as $1.00/months as opposed to $0.25/week, even if there is no actual difference, due to it not being a sub-dollar amount.
I think you mean $13/year not $13/mo; I think you're comparing monthly fees with annual fees. In any case, tipping is optional and there is already a floor > $0.01.
We can't have it both ways. Either programming is worth something (we hate the idea of the quarter) or it's not ($13/yr is sooo expensive!).
Yes, that's what I wrote. Software sales spreads across the whole spectrum (example at hand: Sparrow, $9/eternity). I just wanted to provide a different view.
$0.25/week is not something I'd really consider a tip
Should weekly tips only be given to people whose work I use weekly? If so, then yes, I agree that $0.25/week is not a good tip.
But what about the person whose Emacs config files I read at GitHub and found somewhat informative? And what about the person whose personal wiki (e.g., Org Mode files) I read at GitHub and found somewhat informative? Should I give these people good one-time tips instead of small weekly tips? Maybe I should, but I think I'd like to have the option to give them small weekly tips. Or is there some reason I shouldn't have this option?
For those one-offs where I have interacted with the person directly, I would equate this very closely to the barista model, and expect to be able to give them a single-instance tip that doesn't recur.
I mentioned this before (I think, or somebody did) and Whit's response was that he wanted to be able to shore up the 'regularity' of the revenue stream, but I think that does the whole service an injustice.
I am likely to tip small amounts that recur, and larger amounts for, pardon the analogy, 'good service' or for someone who went out of their way to help me, or did something awesome. I might tip a $1 a week to the Apache Foundation, for example, and/or $10 to the guy who helped me figure out why Apache kept crashing.
As for the 'regularity' of the income, I see no reason why you couldn't just represent the values separately on admin panel... "You are 'earning' $200 a week on GitTip, and also got $50 extra in your tip jar this week."
Add to that the ability to integrate a 'tip jar' into websites, Trac, FogBugz, Github issues or anywhere people interact with other people who help them, and you've got a more rounded system (IMHO).
> I see no reason why you couldn't just represent the values separately on admin panel... "You are 'earning' $200 a week on GitTip, and also got $50 extra in your tip jar this week."
Right, this is how we would do it. I've reopened this ticket:
This is more of a philosophical question for Gittip. I'd agree that the option for one-time tips makes sense. Maybe I periodically contribute to someone who periodically updates a personal wiki, or after a new release of a project I care about. I'm not sure why Gittip decided for subscription-only, but I'd agree that one-time tips make sense as well. If I had to choose only one model, I'd opt for one-time tips.
But (crucially) I still think a quarter is way too small a floor. If I don't value the project enough to make a one-time contribution of even $1 (ideally more), wouldn't everyone be better off with me just sending a nice tweet and/or email? The quarter doesn't really help the developer much more than that. If I were mailing a shareware payment in the olden days, a quarter would barely cover postage.
Taking a cut of a transaction is a hard business to be in. Taking a cut of a microtransaction is an astronomically harder business to be in. Taking a cut of a microtransaction that has no reason to exist is even harder than that.
I sympathize that devs want ways to get paid for developing software, but they exist, and they start with charging money for developing software and not futzing about with sub-latte level payments. Since you cannot meaningfully improve a dev's life with sub-latte payments, you will not successfully create a business on top of them.
I cannot disagree with you, and I'm not about to say "but this time it's different, it's all about the long tail!" I just want to point out, I don't think whit537 is in the business of taking cuts of transactions[0]. He's very much dogfooding here[1], and I would hazard a guess that whether gittip survives or not will depend on whether people like it enough to make it viable for whit537 to pay himself.
Talk to Gregory Brown, who is a Rubyist who has a monthly donation model, about the economics of doing this as a solo dev. (I know that factoid because he wrote a component which makes my business possible and I give him a sale a month as a donation -- which is the exact opposite of generous and I know that and yet I somehow can't justify ever doing more.)
Making a living on Gittip is an implementation detail, not the goal itself, unless I'm greatly mistaken as to your internal calculus. Your goal is to make a living writing code with some X-factor about freedom or project selection, right? There are far better ways to do this. For example, you're currently looking for 50 hours a month of contracting or $2k of donations. This (and gittip) strongly, strongly suggest that you're undercharging for contracting, probably by a factor of 2x to 5x or higher. You can make a huge impact on your standard of living and job satisfaction just by raising rates, and the execution risk of that strategy is absurdly easier than hoping you hit the multiple independent dominoes you need to do to make the gittip plan work.
There are other product offerings you could make as an OSS dev which would complement your desired lifestyle and be orders of magnitude more efficient that gittip. One such product offering is online training material, such as e-books, screencasts, etc. $2k a month is 40 copies of a $50 ebook, which is very achievable. Or it is a one-day make-your-employer-even-more-money-with-Python training session that you run every six months. Or it is... there are a lot of options here and they're all better than competing with baristas for tip money and losing.
A) My current funding goal is $2,000.00 per week, not per month.
B) I want to make the X-factor of freedom available to as many as possible. I think Zuck won the startup game. How are you going to top him? With an even more bloated IPO? Time to rewrite the rules, IMO.
Yeah, what's up with that? Are US wages so different from European ones? Around here (Belgium, a rather wealthy country), $3000 a month is a good wage. 8k/month seems a little excessive for creating and maintaining a relatively simple service? Am I underestimating the effort involved or the tax pressure in the US?
But owning a company, you have far more costs than being employed. E.g. in The Netherlands (it must be similar in Belgium), you have to pay 19% VAT and add lots of insurances that your employer normally pays a substantial part of (disability insurance, liability insurance). Then you probably need a lawyer, and/or insurance for that. And the list goes on... Things become even more expensive once you start hiring employees, etc.
I know a guy working on a related idea. Instead of funding people, his system works by funding issues. The link is http://www.freedomsponsors.org
The workflow, as I understand it, goes something like that:
1) There's an issue on a project you use that's getting no love.
2) You go to freedomsponsors register that you'd like some issue solved, so much that you'd be willing to put your money where your mouth is. Also, you'll probably want to put a link to freedomsponsors on the project's issue tracker.
3) When one or more people solve the issue, the sponsor pays them (if more than one person has worked on the issue, the sponsor can distribute the money in anyways).
Kind of a neat twist on those third-party support sites like Fixya and that GreatSomething one, but that domain name isn't very good. I'd look more towards a "bug bounty" type name.
Whoa, I just noticed a huge spike in access count from Google Analytics coming from here. So this is what happenned...
Hi everyone. I'm Tony, creator of FreedomSponsors.
You're absolutely right @ebswhat. Sorry about that. The layout and webdesign still have major room for improvement - we're working on that. Functionality is pretty stable though.
To sponsor an issue, just do like @reginaldo said.
Gimp is not registered yet so you'll need to fill in more details about it: "Where does the bugtracker live?" and "What's gimp Home page?"
Would you please let me know about you usability experience in the Feedback section?
I'll tell him to put a giant Sponsor button on every screen.
First you'll have to click the tiny "Login with google" button in the upper right corner, then the "Sponsor new issue" button. Then you paste the link to the original issue. After that, the site will guide you a little bit better.
We will add more ways to signup: facebook, twitter, yahoo and github, at least.
I'm not a big fan of owning users' passwords, so unless people really really ask for it, we probably won't add a "create login/pass" functionality.
Also, there is the concept of reputation - Sponsors that pay their offers as promised will have good reputation.
For reputation to mean something, we need to know something about the user, and I believe linking with a social account is our best option for that.
Anyway, talking about FreedomSponsors on GitTip's thread is probably not good for MY reputation here on HackerNews (right? I don't really know, I just signed up). So I'm gonna create a new topic and answer questions there...
I didn't know about that untill now. I'll look more into it.
But the main idea behind the way accounts work on FS is that the user should have some sort of online presence before he can sign up.
So if Mozilla's Persona doesn't offer a way for the user to publish some sort of online profile, then we probably won't add it as a primary form of signing up on FS. Linking an existing account to authenticate later is ok. But we do want to make it more difficult to create completely "anonymous" profiles. At least that's our thinking today anyway.
This is going to sound pretty lame, but I've been suffering from a severe case of Signup Fatigue recently. My business has been doing pretty well so I'd be OK with committing to a weekly debit to my business card, but I don't have a GitHub account and Signup Fatigue is preventing me from setting one up just to give money to people.
Am I being excessively whiny? Are there any plans to not require a GitHub account to give people a little money?
Really neat effort, I hope it continues to take off.
The whole site is about giving money to people on Github. I'd usually agree that enforcing a certain other secondary login to sign up can be a bad thing, but for a product so tightly coupled with another service like this, I think it's okay. At least for the start.
> The whole site is about giving money to people on Github
Yes. But I think the parent comment is saying that also requires that the money is given BY people on GitHub.
Scenario: I don't have a GitHub account because I work for someone that is not doing open source work. However, my own work greatly benefits from a project I found on GitHub (a PhoneGap Library, a Python script, etc). I would like to tip the people responsible for those projects.
Quite the opposite: GitHub would love if you signed up.
You probably won't log out, and chances are you'll end up on GitHub again sometime soon. You'll discover the wonders of social coding now that you have an account, and soon enough, you're a paying customer.
I'm already on there from time to time (as a "consumer"). Likewise I am aware of the wonders of social coding... which existed before GitHub did. And all of this without a GitHub account. But, AFAIK, currently there is nothing I do on the GitHub site that would be enhanced by being logged in.
But the point is not how easy is it to sign up for a free GitHub account. The point is that it seems like an unnecessary requirement to tip someone on Gittip.
I misunderstood that ticket. It reads like you are trying to open it up so you don't need a GitHub account to get tipped. That would be great as I assume that opens up the same options for giving a tip.
Those 319 people (74%) are contributing only $79/month. If minimum contribution was raised to US$1 it would generate more donations even if >60% leave.
Yeah, the current pricing jumps from $1 to $12 a month, $1 to $4 is $4-$24, quite a gap. Honestly, I'd also prefer monthly contributions, thinking in week increments is an american thing. Credit card statements and salaries come in monthly for most of the world, I think.
All the people telling you to increase the price are right, and then some. So some advice:
- dont focus so much on the individual developers, focus more on getting money in the system. You want to know how best to help an individual developer? It's by having lots of users, and lots of money.
- first way you can help do this is by having a minimum monthly charge. I pay $150 a month on a database. I pay 4 digits worth of EC2 bills. I can afford a $10 minimum.
- 2nd way is to strongly focus on enterprise plans. For $1000 a month, you get a big giant banner add on gittip, that everyone else sees every time they add a tip or receive one, or play with their settings.
- dont make people spend the whole thing - distribute the excess to the developers that others have chosen, in proportion to how popular they are. Or use it to highlight projects you like, or projects for people who contribute money, or something like that.
Stop focussing on the low end - bump up that price, and who cares if it's too expensive for some people. Giving $0.25 isn't charity - its actually costing you money.
The stats page should have permalinks for each week so that if someone clicks on this HN submission in 1 month they can see the intended stats, not the latest.
As you can see, we slowed this past week, to be honest. Looking for advice on getting back to ~doubling each week. :-)
I'm afraid you're probably going to be looking at a plateau until you can get the payments thing straightened out permanently so that recipients can depend on getting an ACH every Friday (or Monday, whatever). Closing that loop will instantly motivate every open source hacker to put a "Help fund my work @ Gittip" button everywhere. Bummer the BalancedPayments option didn't pan out.
Bummer the BalancedPayments option didn't pan out.
It's panning out; we're in the process of helping him with the integration.
Also, the "once a week" payout period is unique to Gittip's model (maybe because Stripe requires a 7-day wait?). With Balanced, a marketplace can pay out as frequently as it's earning: even daily. So that's an option Gittip can consider in the future.
Ah, sorry; last I saw, he was heading towards Braintree with a a 6-month horizon for the various merchant account approvals, etc. I hope the current path works out. :-)
To be honest their product isn't as tight as Stripe (had to work around a few library bugs and deployment hiccups), but they have been super open and supportive and respectful of Gittip as a community.
You're comment should be required reading for HNers. SUPPORT is critical even over "coolness and hype"
The fact that these other solutions MAY be better - your experience is with Balanced is going to be what sets them apart and gives them your business.
I'd be willing to bet that the Balanced folks have experience working as contractors. Contractor work is founded on features and capabilities, but their work is kept alive and well through the relationship and support they give their customers.
This is the failing of many "rock star" developers - you may have an incredible resume, but if you're an asshole who would want to work with you.
For the record, I did receive an email from WePay in response to this comment: "I realize that there's nothing to be done now, but wanted you to know that we do take customer service seriously and apologize for the lack of follow up."
This is really cool. I look forwarding to support people in the Django community who make invaluable apps through this.
I would prefer if I could reward people for specific repos, though. Otherwise people, myself included, will probably be clueless why I am receiving money, and as a result it won't create an incentive to improve a particular product.
The text could read.
Hey, I just met you,
And this is crazy,
But here's my money,
So keep working on django-registration maybe.
I like the fact that the donations are per individual, rather than per project. It gives the recipients the freedom to explore other opportunities, rather than tying them down to their one popular project.
Gittip charges fees of $0.30 + 3.9%. They say that they use Balanced Payments, which charges $0.30 + 2.9%, so Gittip is taking a 1% commission on all money that enters the system.
For the sub-dollar charges, keep in mind that the fees are levied for the sum of your tips (assuming no one tips you). So if you only tip one person $0.25 then it's highly inefficient, mostly a 'tip' for the credit card companies.
At 1x$0.25 tips, this structure has an effective overhead of 124.9%; at 10 tips its overhead is 15.9%; at 100 tips, it's 5.1%; and at 1000 tips it's 4.0%.
One solution would be for Gittip to implement something like an account balance. Say I charge my Gittip Wallet with $25 seed money. My CC gets charged $25 + $.30 + 3.9% ($26.07). I set up $1.75/week of tips to various people. My Gittip Wallet gets dinged $1.75 per week for 14 weeks ($24.50). On week 15 when it comes time to give my $1.75, my Gittip Wallet is too low. So it gets another $25 bump (my CC gets hit with another $26.07) and we're good to go for a bit. Wash. Rinse. Repeat. Make the auto-bump amount a user selection in case someone is gifting $30 a week, they could auto-bump at say $100 a pop.
Edit: Also, if I understand correctly, if I am getting tips, that would also go into the same wallet. So in theory the tips I receive would extend the amount of time between auto-bumps. Or not at all if I'm getting more than I am giving.
yes. In SF Bay Area we have FasTrak for bridge/road tolls. Clipper card for public transit. For each, you keep some balance in your account that gets depleted as used and is recharged as needed.
If they took PayPal (which works for both sides of the system using Mass Payments to pay all the recipients), they'd be able to opt for 5% + $0.05 with the micropayments service. On a $1 tip, it's the difference between $0.33 and $0.10 in fees.
I suspect I'm not alone in saying this: If someone paid me, I'd be able to contribute a lot more to open source projects. As it stands, I am inclined to worry that sharing my source code will undermine my ability to make money while writing it, and this might not be an ideal situation.
Excuse me, but I don't understand why anyone would use this instead of flattr? Sorry for sounding blatant, but havent they got this micro-donation thing going quite well?
But I strongly object to the tip amounts. Without going into too much detail, $0.25/week is not something I'd really consider a tip for a barista making my coffee, my dry cleaner, or any of the other myriad service folks I interact with on a regular basis. $0.25 will get you about 10 minutes of on-street parking in many cities; it might take longer to get your coffee. I'd guess that my attorney, physician, or my lawn guy would not consider $0.25/week a genuine "thank you" gesture as much as "hey, I have this quarter in my pocket." Personally, I think programmers are worth a lot more than most programmers think they themselves are worth.
Every other day, there's a HN post about some clueless business guy looking for technical co-founders. The HN community rages about how business people always undervalue their technical contributions. And then here's a project where programmers will tell you that a good project is worth $0.25/week. It that's true, maybe the "clueless" business guys are right to devalue technical work.
So while I really like your idea, I really do wish you'd reconsider the worth of programmers who contribute to these projects. I'm trying to be as constructive as possible here, because I think good things can come from what you're doing if you don't advance the notion that good programming in any meaningful quantity is worth $0.25.