Open Source Communities: How Design Choice with regards to Transparency and Accessibility affects External Participation
Joel West joined the conversation on community-led and sponsored open source projects. His newer paper on the role of participation architecture in growing sponsored open source communities explores governance issues at a deeper scale.
Transparency and Accessibility matter by Josh Sommers
Joel and Siobhán O’Mahony compared corporate and voluntary production models, or sponsored and community managed projects, as they called them. Their study had two research questions:
how did sponsors design open source software communities in the hopes of attracting external participation, and how did this differ from the design of autonomous based communities?
Interesting questions indeed, coming from real experts of open source community governance. Joel and Siobhán didn’t focus on boundary between work and hobby in an individual’s participation, or other individual motivations for open source contributions, as Joel explained. They focused on investigating how sponsored communities differ from their autonomous counterparts identified three key design choices: the organization of production, community governance and intellectual property. I tend to agree with Joel saying that the “meat” of the paper is in Table 3, reporting the specific trade-offs made across the various independent and sponsored projects, but I warmly suggest you to read the whole paper.
Talking about the organization of the production of code, they say (bold emphasis is mine):
Overall, the degree of modularity, associated dependencies, and the quality of code documentation affected the ability of outside members to understand the code well enough to contribute. [..] In addition to the technical architecture of the code, the organization of production includes control of the processes by which individuals participate in the community’s production process. These social measures are not necessarily correlated to a project’s technical design: for example, highly modular code can still be tightly controlled by a single firm. Thus, a project’s technical architecture is one subset of a community’s participation architecture. [..] We identified three design parameters that provided contributors with transparency and accessibility to production processes:
1. Live code access provides transparency by offering the community the chance to review the most recent “live†version of source code on the community website [..] . 2. Public commit process refers to the opportunity for community members to become directly involved in the production process by earning (through demonstrated technical proficiency) the right to directly commit software changes to the community repository.[..] 3. Subproject creation is a mechanism by which a community based on the sponsor’s original code can grow to assume new functionality or new directions.
The production of code is key to a sustainable open source business model, that’s why some open source firms are shifting towards an hybrid production model and also why retaining an existing one is vital.
Roberto Galoppini, and 
Giuseppe Maxia are discussing.
Joel West 10:07 pm on July 17, 2008 Permalink
I think over time this hybrid model will be recognized for what it is. Most OSS users and industry types know that MySQL uses OSS as a marketing vehicle (and a way to get adoption by students and small ISVs), and that’s not sharing participation or authority the way that an Apache or Eclipse does.
Don’t get me wrong: being open on 1 dimension is better than being open on 0 dimensions; let’s just not confuse it with being open on all 3 dimensions.
Giuseppe Maxia 11:54 am on July 18, 2008 Permalink
Joel,
what you say is true. MySQL, as a company, uses open source as a marketing vehicle. It means that the company seeks to enlarge the user base as much as it can be done, and then looks for potential customers within the extended user community.
This is a good model, and one of the reasons why Sun acquired MySQL, i.e. for its ability of cashing on open source.
MySQL, as a project, is sadly less open than others. While it can rightfully claim a user base far superior to any other open source database, it has a very limited number of external contributors. The company has structured its development practice around a closed source model, exacerbated by the RCS tool of choice (BitKeeper) which made openness even more difficult. We have been changing, though. Slowly, because development habits are hard to change, but the direction is towards openness. We now have many developers openly talking about the project in public IRC (instead of the internal one, the only one we used until one year ago), we have moved our code base to the more community friendly Bazaar, and this will make contribution easier. We have promoted 14 projects within Google Summer of Code, and the outcome of these projects can be added to our code (depending on quality review, of course).
In short, we are on the right path. We acknowledge that, while we have been a good example in matter of cashing on open source, we have a long way to go in matter of public participation. But we are learning.
Giuseppe Maxia
MySQL Community Team Lead
Sun Microsystems – Database Group
http://datacharmer.blogspot.com
Roberto Galoppini 5:42 pm on July 18, 2008 Permalink
Thank you Giuseppe to join the conversation, I think you are really honest about where MySQL stands today. As OpenOffice.org community member I do know how difficult is to cope with corporate production models. Hybridization is a pretty new thing, but I think that it is a process, and I welcome MySQL’s first steps.
As I observed in another comment to Joel’s ones, I am also looking forward to see what can do communities involving (also) consumers, like SAKAI, or “not organic” ones like those managed by the Collaborative Software Initiative.
Joel West 12:52 am on July 21, 2008 Permalink
Giuseppe,
Thank you for your frank comments. I am perhaps one of the few open source researchers that’s not a true believer, maybe because I spent too much time in the software industry. But frankly I think it’s cool if companies can find success using OSS as demoware, or as a way to get PR, or by providing full IP but not sharing control or development. Parceling out different rights to see what buyers will value is just another form of business model innovation, like Southwest or Wal-Mart or (frankly) MySQL. And such innovation deserves to be rewarded.
My only concern is false advertising, e.g. people who claim they’re promoting some noble cause but really are just out to make a buck. IT buyers are not stupid, and by now the industry has a fairly sophisticated understanding of OSS. So unlike in the consumer space, I think vendors will be punished for exaggerated claims and are better off just leveling with people. If nothing else, the dot-bomb era taught us that a free lunch doesn’t last forever and IT buyers certainly want their vendors to be around to provide support and upgrades down the road.
Joel West
http://www.joelwest.org