The last week of July was OSCON. Now that I’ve had a week off to think about all the amazing conversations I overheard and took part in, I want to share some of the ideas that I was hearing.
I took Steve Yegge’s call to “work on important things” to heart in last week in Portland. I sought out hard conversations with the collection of open source users, developers and luminaries that attended OSCON. I started most of my hallway track sessions asking and thinking very broadly about our tag line, “from disruption to default.”
Growing the community
If open source is now the default, if the audience is now much broader, then are we talking enough about the primary audience for open source tools, projects and products? Now our constituency is not all developer peers. We must challenge the idea that if someone really wants to use a piece of software, he or she will be willing to slog through half-written documentation, the actual code base and an unkind user interface.
Gabe Zichermann made a fantastic point in his keynote about the introduction of new players to a game: In the first screens of a game, there is no way to do the wrong thing. In fact, there is only one thing to do, which is the first step in learning how to interface with that product. Likewise, lowering the barrier to entry is paramount to continuing to promote open source as a culture and a global good.
There was a time in open source when, if you used a product, you also were a contributor to the code. Now, we are experiencing a redefinition of contributor and community member. Users who make good bug reports are fantastic contributors, though they may not code. Designers who rework the user experience for broader appeal are also important contributors. An entire culture based on finding people to relate to as outcasts is now learning to deal with popularity. The label “geek” is no longer a pejorative, but instead is becoming a badge of honor.
These adolescent growing pains are causing tension within the greater open source community. Change is happening: ‘Open’ has gone from elitist anti-crowd to global buzzword, experiencing dramatic growth and shifts in business models. Hackers aren’t doing this simply for love but often looking for corporate sponsorship. There are more subtle changes, too, like the redefinition of ‘community’ to include not only developers but also end users, corporate sponsors, people of color and women, designers and UX engineers.
Boy, it’s hard to be cool. The early projects which had success in broadening the reach of a project or in securing corporate sponsorship, funding or a business model were decried as sellouts. Think RedHat, circa 2002. Who would ever pay for Linux support?
In more recent times, Jono Bacon has navigated that line well. He is not only the community face of Ubuntu and Canonical, but an actual rock star. He gave a keynote at OSCON last week about the future of communities. Then, in an incredibly authentic blog post, Jono reflected on his presentation and noted what he learned. Transparency and openness to alternate perspectives are what allow him to rally an eclectic community of developers, users, designers, and the curious who in turn promote and host events to add a three dimensional component for the community around Ubuntu.
Using data and open source better
In some quiet time at OSCON, Dave Eaves and Edd Dumbill had a fantastic conversation about making use of data to better understand the ebb and flow of an open source community’s engagement.
Dave has been working with Mozilla to seek out and understand what is quantitatively happening in the community to inform constant improvement. Ask “why are bugs in this section of the code taking twice as long to be reviewed?” Ask “who has contributed consistently over the last 18 months, but not in the last 30 days?” Take the information, both qualitative and quantitative, and then use it to continue to improve our communities.
In another example, when OpenStack was announced on the OSCON keynote stage last year, it gave rise to conversations with the thesis that there are “real open source” projects (and, presumably, “faux open source” projects). Under scrutiny were corporate sponsorships and business models which have funded a proliferation of newer open source projects like Ubuntu or Hadoop. Many argue that companies still don’t know how to take a product and “open source it.”
The fact that “open source” is a verb suggests we have made it, but the community’s limited engagement with OpenStack means there’s still a long way to go. At Nebula’s launch keynote, Chris Kemp asked who had worked with OpenStack; the response was anemic. There is hope. OpenStack has increased from a dozen contributors (primarily funded by Rackspace) to more than 250 in the current release and is seeking to engage more independent developers as well. (There are currently 1200 developers in the development tree.)
What does open mean?
Some arguments for open source are about transparency or safety. The more eyes see code, the more refined it becomes, and the more security exploits are found before impacting an end user. In her keynote at OSCON, Karen Sandler answered how open source will benefit consumers of medical devices. A corporate sponsor, however will be considering the talent pool of theoretically unlimited volunteer workers. The dogmatic among us want “free-libre” for the independence from corporate overlords.
I believe open is a mindset: accepting and respecting the views of others and encouraging more perspectives in our communities. Not every company will have goals or methods that align with the open source philosophy. There are many companies which will. The first step to hacking is to frame the problem and understand the biases. We have the opportunity here to share experience and find where we can work together. Instead of conversations about if a corporate sponsored project is really open, why not welcome the corporate sponsors to the table and work to find where our interests align. Nurture those ideas, then start discussing the harder edge cases.
We lack common motivations for a set of complicated goals loosely collected and anointed “open source.” When we address this underlying problem, the other symptoms we’ve been treating will ease. There is a culture behind the idea of open source. To move that vision forward, there will have to be more public successes.
Jim Zemlin suggested in his keynote that we need to brand the vision behind open source, much like “Made in the USA” or “Fair Trade.” This means working with and embracing our old rivals and finding where our interests align. In the least marketing-y keynote I’ve ever seen from Microsoft, Gianugo Rabellino spoke of a place for both open and closed source products and advocates embracing their interoperability. While I don’t see Microsoft being a corporate sponsor for Linux, they are kernel contributors. Alignment of interests give rise to opportunities where everyone benefits. We will see more in the future.
Have we succeeded in disrupting? Yes. Is open source the default? In some places. How do we do it again? We embrace the change we birthed and work to evolve and grow this adolescent into a robust, healthy and curious philosophy.