Re: Bounty for Bugs in Open Source Projects?

Frank Hecker <[email protected]> Mon, 18 Apr 2005 11:42:29 -0400
Newsgroups gmane.comp.misc.free-software-business
Message-ID <[email protected]>
Stephen J. Turnbull wrote:
>>>>>>"Nik" == Nik Cubrilovic <[email protected]> writes:
>     Nik> I am interested to know if there is currently an initiative
>     Nik> for commercial organisations to offer a bounty (cash
>     Nik> incentive) for bugs or features to be implemented in Open
>     Nik> Source projects.
> 
> Well, there's sourceXchange (dead) and cosource.com (dead/morphed into
> collab.net) and BeOpen.com (dead).  I don't think it is an accident
> that Red Hat and SuSE, on the other hand, are thriving by hiring core
> developers from various projects, as well as "generic" developers.

Minor correction: SourceXchange was a service of CollabNet (back in the 
days when it was known as Collab.Net); after the death of SourceXchange 
CollabNet then switched to focus on SourceCast (now CollabNet Enterprise 
Edition). I don't know what if anything survived from Cosource.com.

> The only case where having a middleman might make sense is if you need
> somebody to represent a bunch of buyers or a bunch of sellers, or
> both.  But that's hard work; unless the middleman has very good people
> skills, you're going to end up with both the buyers and the sellers
> mad at him!  As I understand it, that's what happened with cosource/
> collab: they realized that they were in the relationship business, and
> started focusing on that, rather than trying to build a market (or
> even just a bazaar).

Again, I don't know about Cosource, but IMO the problem with 
SourceXchange was basically a combination of insufficient demand and 
excessive transaction costs: not necessarily monetary costs per se but 
the whole hassle of coming up with Statements of Work, getting 
customer-developer agreements in place, dealing with international 
issues, and so on. Most companies didn't see any real need to fund 
development of open source software, and those who did see the need 
found it easier to accomplish this in other ways, e.g., by finding 
existing employees with the necessary skills or hiring new employees.

What CollabNet then did was to refocus on tools to support collaborative 
development and consulting services to advise companies on how to do 
collaborative development successfully, both based on experience gained 
in the open source/free software world.

In any case, I agree that the idea of a bounty-based development market 
has been tried before and based on those experiences is not likely to be 
successful on a large scale (or even necessarily on a small scale).
IMO the primary people who are interested in it are developers who 
imagine this might be a way to make a living (or at least supplement 
their income), and individual users who would like to see features 
developed but don't have much money to fund them. These do not a viable 
market make.

Frank
-- 
Frank Hecker
[email protected]