Making free software development profitable

oxplot · 2012-03-15 07:27:45

OK, this maybe rather off the topic of GNU/Linux but bear with me

Free software has come a long way and all thanks to generous contributions of numerous souls who have put in their most valuable asset, time, to develop a huge body of software we can all enjoy.
Now a good number of free software developers work for commercial companies and get paid to do what they do. That is cool too. In fact that is great! If one could live off money earned by developing free software ... I mean that is fantastic. Financially benefiting in addition to helping the community is certainly better than one or the other alone. But this kind of job is rare.

What if more (or every) free software developer could be financially compensated for their contributions? I'm sure there is a good amount of pessimism about the idea of expecting money for something that can otherwise be obtained for free. And that's justified, except there is always the assumption that the developer makes something and then asks for money. That is the trend (and perhaps has been). I'm referring to donations. There is always the software before the donation button. There are two issues with donations:

1. you have to have something first before getting paid for it (which provides little to no motivation for those able to contribute but whom don't have time due to their commitments to their paying jobs)
2. donations are not directed at anything specific, you don't know if the $5 you sent to that great girl/guy is going to his project A or B (which provides little motivation for users to pay because of uncertainty)

Now a model like that of kickstarter.com would solve less than half of the issues here. I personally think that kickstarter like services are geared towards creative projects, projects for which there is not necessarily a need at first. In free software communities however, it's very prevalent for users to be looking for something they need but not quite finding it. It could be a feature they need, a bug that needs fixing. Do you see where I'm going with this? Yep. Issue trackers with a twist!

A service which allows users to submit bugs/requests/ideas and offer money to whoever can get it done. But (a big but), the end result of the software has to be free software (otherwise you got yourself a rentacoder kind of service). A simple (or complicated if necessary) reputation based system would create trust in developers not to steal money and in users to pay what they pledged. This sounds like fundraising and that's exactly what it is. The difference is: it's focused and streamlined. You know who/what/why you're paying.

What's the incentive for the user to pay? Well they *need* something that doesn't exist. They can file a bug report and hope someone will take notice. Or they can get attention by offering to pay for what they need. From view point of some, this kind of transactional environment goes against the values of free software. It may seem unfair for those who can't afford (or want) to pay as much to get equal attention. This and other kind of issues are what I like to discuss with you.

I believe a well thought implementation of such system can change face of free software and accelerate its development.

Thoughts?

t1nk3r3r · 2012-03-15 07:53:31

Check out Wine/Codeweavers:
http://www.winehq.org/about/
and
http://www.codeweavers.com/compatibility/toplists/

Trilby · 2012-03-15 12:16:02

That is basically the pay-for-service model rather than the pay-for-product model which I thought was already the norm in the open source world (for those who make software a profession).

If I start a consulting business I can help people set up computer systems and get paid for that help. I'm paid for my time and expertise, I'm not paid for any of the software I put on their computer.

If they wanted some of that software customized to their own needs they'd pay for more of my time and effort to either go out and find the branch/patch that does what they want, or write it myself. They are still paying for the service, not the product.

I plan on starting a (very small) side-business doing just this sort of thing once this pesky dissertation is out of the way.

jgreen1tc · 2012-03-15 13:11:43

In nonprofits, donations go towards the nonprofit entity, not directly to those doing the work/service. How the nonprofit uses those funds is up to them. It is usually better if 15-20% (or less) of donations goes towards overhead cost (such as payroll). The rest should probably go back into the nonprofit programming. In this case, I'm guessing a lot of donations go towards hosting/servers. Should more of the donations go towards all the people who contribute to Arch, we might see a price tag on it unless increased and more consistent funding is found (speculation, but likely).

I'm a nonprofit major btw

oxplot · 2012-03-15 18:31:44

Trilby wrote:

That is basically the pay-for-service model rather than the pay-for-product model which I thought was already the norm in the open source world (for those who make software a profession).
If I start a consulting business I can help people set up computer systems and get paid for that help. I'm paid for my time and expertise, I'm not paid for any of the software I put on their computer.
If they wanted some of that software customized to their own needs they'd pay for more of my time and effort to either go out and find the branch/patch that does what they want, or write it myself. They are still paying for the service, not the product.
I plan on starting a (very small) side-business doing just this sort of thing once this pesky dissertation is out of the way.

I think pay-for-service does encompass the majority of what happens in (free) software world. This is not to be interpreted as implying lack of innovation. It is just the nature of software that requires constant maintenance throughout its lifetime (in contrast to a music record which is more or less complete after its release).

Now using such a service, you could find paying clients (aka software users) a lot easier and in larger numbers. There would also be a lot more variety (eg the entire spectrum of free software, not just one app or the other).

I admit there is a caveat, which might still keep some developers away from this and that is job security. But like you said, it doesn't have to be one's full time job.

My main question is perhaps whether you as a developer/consultant would be interested in such a service and take it as a serious source of income? Can you see any significant issues or downsides?

oxplot · 2012-03-15 18:49:32

jgreen1tc wrote:

In nonprofits, donations go towards the nonprofit entity, not directly to those doing the work/service. How the nonprofit uses those funds is up to them. It is usually better if 15-20% (or less) of donations goes towards overhead cost (such as payroll). The rest should probably go back into the nonprofit programming. In this case, I'm guessing a lot of donations go towards hosting/servers. Should more of the donations go towards all the people who contribute to Arch, we might see a price tag on it unless increased and more consistent funding is found (speculation, but likely).
I'm a nonprofit major btw

This is a very good point. You are essentially saying (if I understood correctly) that shifting the financial support from infrastructure and hosting to developer's time would put pressure on the former which may lead to charging people for downloading Arch.

Now, considering the enormity of any GNU/Linux distribution, I think it is safe to assume that users recognize that the money they spend on getting a new feature in one package amongst thousands, will not also compensate everyone else involved. We are talking about $50, $100 features which are crowdfunded from dozens of users collectively. Therefore I think the impact on donations would not be at all significant.

However, for smaller projects, run by perhaps one or couple of developers, donations may indeed suffer as the result of this process. This can be a good and a bad thing:

1. It is bad in case the reduction in donations is not matched by the new source of income for developers
2. It is good because developers have now more control on how much they could earn by getting more bug fixes and feature requests through.

jgreen1tc · 2012-03-15 19:13:22

You have to remember though, that developers of Arch are not necessarily responsible for the development of packages. New features or bugs in packages aren't patched by Arch devs, but upstream. Now if the Arch system is the cause of the problem, that can be fixed. I don't know whether you're talking about the former or the latter in your examples, but it matters for the distribution of a donation. If you mean you want features added to a specific package, or bugs to be patched faster, it might be better to donate to the people who manage that package.

I don't know how Arch Linux manages their funds, but as I said before, I suspect that increased and consistent funding would be needed in order to pay Arch developers a salary for what they do. If this is what you mean, then a solid fundraising/marketing plan needs to be put in place in order to get it.

oxplot · 2012-03-15 19:26:43

jgreen1tc wrote:

You have to remember though, that developers of Arch are not necessarily responsible for the development of packages. New features or bugs in packages aren't patched by Arch devs, but upstream. Now if the Arch system is the cause of the problem, that can be fixed. I don't know whether you're talking about the former or the latter in your examples, but it matters for the distribution of a donation. If you mean you want features added to a specific package, or bugs to be patched faster, it might be better to donate to the people who manage that package.
I don't know how Arch Linux manages their funds, but as I said before, I suspect that increased and consistent funding would be needed in order to pay Arch developers a salary for what they do. If this is what you mean, then a solid fundraising/marketing plan needs to be put in place in order to get it.

I wasn't specifically talking about Arch. Given my initial post, I'm thinking of this service that's like a huge issue tracking system, including every free software out there. So if a user submits a feature request against a software package, and 10 others back him/her up, then any developer, be it the upstream developer or some college kid on holidays, can agree to implement it. However, for features that require an in-depth knowledge of current software, you will more likely need an upstream person to look into it (as you mentioned).

Regarding a salary, that is what I envision this service to accomplish in the long term. In the short term, I don't believe it will provide enough funding to support developers in that capacity. However, when and if it does, a developer's income would be very tightly correlated with how much improvement is done to their software. I think that can have tremendous effects on innovation and development.

karol · 2012-03-15 19:48:34

You can pay for the openBSD discs http://www.openbsd.org/ftp.html, or Ardour download http://ardour.org/download
Some people are trying to make a living this way.

Edit: You can also sponsor specific issues for Ardour http://ardour.org/node

Popular Issues
Ardour users want these issues fixed. Click to sponsor these or other issues
Description Users Total
MIDI editing ... 27 $1365
AAF / OMF su ... 11 $505
performance ... 2 $140
Connection m ... 1 $100
...

(sorry for crappy pasting)

Arch had a Bounty once but it never took off http://archlinux.me/dusty/2010/01/22/de … ch-bounty/

Last edited by karol (2012-03-15 19:52:16)

oxplot · 2012-03-15 19:53:10

karol wrote:

You can pay for the openBSD discs http://www.openbsd.org/ftp.html, or Ardour download http://ardour.org/download
Some people are trying to make a living this way.

True. In fact charging for distribution of free software is specifically mentioned and encouraged by Free Software Foundation.

However, I am more interested in money that is directed at specific features/requests rather than an entire distribution.

karol · 2012-03-15 19:56:34

oxplot wrote:

I am more interested in money that is directed at specific features/requests rather than an entire distribution.

I edited my post; see the Ardour user-sponsored feature requests.

alphaniner · 2012-03-15 20:01:13

Haikuware also has moderate success with a bounty program.

Haikuware Bounties

oxplot · 2012-03-15 20:11:18

karol wrote:

Edit: You can also sponsor specific issues for Ardour http://ardour.org/node
Popular Issues
Ardour users want these issues fixed. Click to sponsor these or other issues
Description Users Total
MIDI editing ... 27 $1365
AAF / OMF su ... 11 $505
performance ... 2 $140
Connection m ... 1 $100
...
(sorry for crappy pasting)

Arch had a Bounty once but it never took off http://archlinux.me/dusty/2010/01/22/de … ch-bounty/

Thanks for that! Precisely what I'm talking about, except in larger scale, centralized and with more reach to all developers, not only the main ones.
If you were an independent developer looking to get some work done for money, it would be more convenient to look for it in one place as opposed to countless websites with countless types of systems.

alphaniner wrote:

Haikuware also has moderate success with a bounty program.
Haikuware Bounties

This is excellent! Shares the same view as I have. Again, like I mentioned, I'm researching (and hopefully will be building) a system which brings all these scattered efforts into one place.

karol · 2012-03-15 20:15:12

http://www.fossfactory.org/

Trilby · 2012-03-15 21:54:50

oxplot wrote:

My main question is perhaps whether you as a developer/consultant would be interested in such a service and take it as a serious source of income? Can you see any significant issues or downsides?

I certainly wouldn't call myself a developer. A proficient hobbiest at best - but in a world of blind <insert other OS name here> users the one eyed man ...

I can imagine such a consulting business would be quite difficult, but I'd actually market myself not as a software or computer consultant per se. In my case statistics and data analysis are my strength, and I have enough knowledge of programming to help automate much "menial labor" often done in scientific research.

I've become the 'go to guy' for anything related to data analysis/conversion/etc for several departments at a major university. I figure I could just keep doing what I'm doing, but make a few extra bucks at it.

Grinch · 2012-03-16 07:32:03

In a larger perspective I hope 'crowdfunding' will continue to help fuel open source software, and yes I count bounties as part of 'crowdfunding'. It seems like the most effective way for end users to target development in the direction they want apart from actually doing the coding.

As for making money of developing free software in general, one viable option is that which is practiced by the x264 devs. They release x264 under GPL and as such it's free to use for open source projects, however they also offer a proprietary licence for which they charge money. Apparently they are doing good business this way, and I do think it's a good solution if you are developing code which is of interest to proprietary projects.

oxplot · 2012-03-16 21:42:50

karol wrote:

http://www.fossfactory.org/

This is interesting. I hadn't come across these guys before, either because they have poor SEO or I am incompetent in searching

Having looked at some of the projects, it seems to be more or less active (not sure if they're thriving). I also had a read of their process and FAQ. What I have in mind seems to differ quite significantly with what those guys are doing in regards to the funding and the general process. They're doing it in a collaborative manner using an organic approach of allowing change to requirements and funding along the way. I'm thinking of the same type of collaboration in planning and design stages, but with a more stable state past the funding stage. In particular, I think kickstarter.com method of goal based time limited fundraising to be more successful in attracting developers to work on a project. I would describe the process I have in mind as follows:

- a user/developer submits a project idea (could be as big as an entire app or as small as a bug fix)
- other users can discuss the project, or create sub-projects and discuss those
- users also indicate on each project how much they're willing to contribute financially (without commitment)
- developers can submit proposals on any (sub) project
- proposals must have clearly indicated specs and deliverables as well as a funding goal and deadline
- proposals can be discussed just like project ideas and developers amend the specs as necessary to reflect the discussions
- developer decides it is time to finalize and start fund raising
- at this point the specs are frozen and a time limited fund raising window is open for users to "pledge" (kickstarter's term)
- if the funding goal set by developer is met, all the funds are transferred from users' accounts to the developer's account, otherwise no one's charged
- developer starts to work on the project, checking in regularly to post status updates
- a reputation system awards or takes away reputation points from developers based on their ability to meet deadlines, deliver what they promised, etc.
- developers can be permanently banned (based on their billing details) from the site if they simply don't get any work done or otherwise violate a term
- developers are limited on how much funding goal they can set based on their reputation (this helps building trust)

Arch Linux

#1 2012-03-15 07:27:45

Making free software development profitable

#2 2012-03-15 07:53:31

Re: Making free software development profitable

#3 2012-03-15 12:16:02

Re: Making free software development profitable

#4 2012-03-15 13:11:43

Re: Making free software development profitable

#5 2012-03-15 18:31:44

Re: Making free software development profitable

#6 2012-03-15 18:49:32

Re: Making free software development profitable

#7 2012-03-15 19:13:22

Re: Making free software development profitable

#8 2012-03-15 19:26:43

Re: Making free software development profitable

#9 2012-03-15 19:48:34

Re: Making free software development profitable

#10 2012-03-15 19:53:10

Re: Making free software development profitable

#11 2012-03-15 19:56:34

Re: Making free software development profitable

#12 2012-03-15 20:01:13

Re: Making free software development profitable

#13 2012-03-15 20:11:18

Re: Making free software development profitable

#14 2012-03-15 20:15:12

Re: Making free software development profitable

#15 2012-03-15 21:54:50

Re: Making free software development profitable

#16 2012-03-16 07:32:03

Re: Making free software development profitable

#17 2012-03-16 21:42:50

Re: Making free software development profitable

Board footer