While planning the first Platform conference, I asked Andrew Shafer for advice and to facilitate an unconference. I first met him at OSCON 2010 and have enjoyed Andrew’s thought provoking perspective ever since. I sat down with Andrew to learn more about his experience with conferences, communities, open source, business and his plans for the Platform unconference.
Andrew, most people who are following the cloud computing and devops storyline for the past few years probably know who you are, but why don’t we start with an introduction.
Right, I’m Andrew Clay Shafer (@littleidea). A lot of people know me as a co-founder of Puppet Labs, and for evangelizing devops, but I worked for a couple of funded startups before that. I worked on an engineering team with David Flynn on devices using flash technology before he started Fusion-io. Since the Puppet days, I led teams that implemented some of the largest CloudStack and OpenStack deployments as VP of Engineering for Cloudscaling. I also like to organize events. I helped put together the first Puppet Camp, the devopsdays in Silicon Valley, Agile Roots and a bunch of Ignite events. That’s probably enough to get the conversation started.
I asked you to help with the planning for our Cloud Foundry community conference and to lead the unconference. Please share a bit about your experiences with similar events.
As I was saying, I’ve been involved in planning a few conferences. I’ve also been to dozens of conferences in the last few years. Some things you have to learn by doing it wrong and some things you learn from seeing other people do it right. Conferences can literally change the trajectory of a career. I learned a lot about that from watching and working with my co-founder Luke Kanies. I’d been working as a developer for a few years before I really understood that. I remember leaving Mountain West Ruby Conf in 2008 and I was just so pumped. You see the technology, but it’s really about the people, seeing what they are capable of, seeing what is possible, that’s what is inspiring. Seeing other people do something makes it possible for you.
Unconferences are really my favorite type of conference. Rarely do you have more expertise on the stage than you do in the audience at an industry conference, but most conferences are all about one person broadcasting, so you have the real dialog in the hallway. A good unconference inverts that. The group co-creates the experience, first deciding what is important enough to discuss and then discussing that as peers. I’m not going to direct people to do anything so much as articulate a framework for that co-created experience. Sadly, I think it is harder to get companies to support sending people to a pure unconference. The first conference I know of to do half presentations, half unconference was the first Puppet Camp, then devopsdays. That was a somewhat strategic decision because we knew we wanted to get a bunch of smart people together to talk about important things, but we thought it would be easier if we published some names who work for companies with well known names and logos to get managers to sign off on travel. And here we are… nice logos by the way.
You’ve also been involved in a few projects helping to build businesses and communities around open source. I’d like to hear what you think of how we are doing with Cloud Foundry so far.
To be honest, I don’t believe open source business or community is a formulaically solved problem. I know quite a bit about a lot of projects and a lot of communities. I’d say there are qualitative and quantitative differences in all of them so it makes it hard to compare apples to apples. Projects get started with different goals. Projects get led by people with different perspectives and personalities. That has an impact on everything to do with the community and business that can grow around the project.
Communities are defined by how they handle critique and dissent. I obviously think community is important, but I think people often mistake politeness and a lack of open conflict for cohesiveness. This is often the case in organizations as well, so there is no surprise to see that brought into a larger collective. The end is a lot of unresolved sublimated conflict and compromise with sub-optimal results. That doesn’t disregard the need for courtesy and respect, just that those are not ends.
Business is about creating and exchanging value. Open source can change some of the dynamics, but ultimately it comes down to convincing someone that what you provide is valuable. If you can do that, you can build a business, if you can’t, you have a hobby. Sometimes what you have isn’t valuable and sometimes you need to improve the narrative you are selling.
All that being said, I think Cloud Foundry has had some stutter steps along the way but appears to be trending in a good direction with regards to both community and business. In the end, my opinion doesn’t really matter, what does matter is who runs the code, who contributes to the code, who gets value from the code and how that allows for relationships that create more value for everyone.
Also, you could use better documentation. How’s that?
Is there anything else you would like to add?
Just that it is an amazing time to be building applications and operating infrastructure. So many interesting threads and possibilities are coming together with this. When I was first drawn to working with Luke on Puppet, the mission was literally to change the way that people manage computers. The first discussions we had about that were over a decade ago and look at where the devops discussions are now. A big influence on how I see software and what I see as my contribution to devops come from lessons I learned from Agile, especially the community around Alistair Cockburn and the Salt Lake Agile Roundtable. In the Agile community, I was always impressed by Pivots and Pivotal Labs. I can still picture their logo as a sponsor of my first Mountain West Ruby Conference and some of the great conversations I had with Pivots in the hallways.
There is massive narrative building around continuous delivery, cloud, devops and Agile. Software is eating the world and these things I’ve worked on for the last decade are all blurring into one story. I expect a lot of competition. We both know there is competition out there and on the way, but Cloud Foundry is positioning itself towards the center of that hurricane. You know I don’t think Cloud Foundry is perfect, no tool is, but people are getting real value now and I think it raises the bar for what people expect is possible deploying applications with open tools.
This is an amazing journey. I’m grateful to be a part of all of it, and grateful to get your invitation to help out. I hope people will come to learn and share their journey.
Thanks Andrew, we’ll look forward to seeing you at Platform.
I look forward to being there.