Take the 2-minute tour ×
Programmers Stack Exchange is a question and answer site for professional programmers interested in conceptual questions about software development. It's 100% free, no registration required.

I have a library for R (open source statistics package) mapped out on paper. I have started coding the different functions but I realize that I don't have the time needed to finish this in a reasonable amount of time. I know I can just throw the code up on a repo and call for others to help fill in the blanks. But I'd like to incentivize things a bit. I'm thinking about putting a bounty on each function of, say, $5-$20. There's no way that $20 would be a fair return on time for a developer to code each function. But my thought is that the cash (or Amazon gift certificates) would be an inventive for people to actually work on the project. And it would allow me to put higher bounties on the functions I am most interested in.

I have a few questions related to this:

  1. Good idea?
  2. I going to make the development work faster, or slower? I've read Predictably Irrational and I'm concerned that by offering menial pay for functions I might actually disincentive developers.
  3. Are there sites devoted to this type of activity? Can you recommend one based on personal experience?
  4. Would you recommend a totally different approach? I'm open to ideas!
share|improve this question
    
    
Turns out that a newer question was a duplicate of this one: programmers.stackexchange.com/questions/79561/… –  user16764 Feb 17 '13 at 23:48

4 Answers 4

Not a good idea, in my mind. None of the OSS programmers I know would respond to such a bounty.

So, what does incentivize people? According to Dan Pink people are motivated by:

  • Autonomy
  • Mastery
  • Purpose

Then to attract good programmers, find a way to provide some or all of those items.

A second approach that can be done concurrently with the first is to display a home page that tracks the progress of the project, showing the status of each of the functions along with the person who provided the function that first passed the unit tests (you do have tests, right?).

Finally, it's been my experience that a compelling project doesn't need much help in attracting contributors. Take a look at what you are doing and if you are having a hard time attracting and keeping programmers to work on it, think about what that is telling you about the usefulness of your project.

share|improve this answer
    
that seems like very good input. I've read Pink as well and his ideas are part of the nagging voice in the back of my head that keeps telling me "this might not be a good idea" –  JD Long Dec 21 '10 at 20:13
    
youtube.com/watch?v=u6XAPnuFjJc <- I know Dan Pink from this video. –  Joe Z. Jan 5 '13 at 3:45

I remember seeing some sites during the dot com days that were basically exactly what you describe. People would post small coding tasks they wanted done, a $ amount, and people could register to perform said task - there were some variations on that theme, but that was the basic idea. Being fresh out of school and looking for some extra moolah I would often poke around and look for a good one to do. The result? I never did a single one. Invariably I'd look at the tasks (that I could do) and do a price/performance in my head and realize it really wasn't worth my time to bother (exactly the point you make in #2). The other problem was that almost all of them were not compelling problems - there was a reason why they were being farmed out :)

I agree with KevDog that if you have a cool project & some decent PR (getting the word out) that people will come find you and do the work for free. While I've never gone the mercenary route I've certainly contributed code here & there to OSS projects that strike my fancy.

share|improve this answer
    
thanks for your opinion, Jeff. That makes perfect sense. –  JD Long Dec 21 '10 at 20:59

I don't think the idea is completely out of the realm of possibility, however the cost per task paradigm doesn't work as it isn't cost effective to the developer, nor proportionally scalable.

I think a better system might be $/Line Of Code where said loc resides in version control for x amount of time and is not committed over for reasons of incompetence (eg. bug).

share|improve this answer
2  
I can pad out lines of code if I have an incentive to. –  David Thornley Dec 21 '10 at 21:40
    
Indeed. The answer was 176 character representation of a basic idea, however. Any idea going into production would need many, many more rules and safe-guards. –  Craige Dec 21 '10 at 21:44
1  
But your first 3 lines are totally unnecessary, i.e. they are useless padding. If you get paid by the line, you can probably stretch it out into at least a couple of more lines... –  jmoreno Feb 18 '13 at 0:36

https://www.bountysource.com

From the about page:

BountySource was originally created in 2004 with the hope of increasing and improving development in open-source software communities. The first iteration of BountySource provided a variety of tools that allowed for easy management of open-source projects. Some of these tools included a Task Tracker, an SVN Code Repository, and a Content Management System.

BountySource was way ahead of its time...we'd like to think of it as a predecessor to GitHub.

After a lengthy hiatus, we're back with the same vision - overall improvement in open-source software development - but a completely different system.

We're shifting our focus from project hosting - repositories, issue tracking and all - to the crowdfunding aspect of BountySource's original idea.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.