Recap: DevOps Enterprise Summit San Francisco 2016
Recap: DevOps Enterprise Summit San Francisco 2016
This year, IT Revolution and Founding Partner Electric Cloud hosted over 1,300 attendees in San Francisco for three days of DevOps sharing, learning, and mutual support.
Join the DZone community and get the full member experience.Join For Free
Read why times series is the fastest growing database category.
What an amazing week we had at DevOps Enterprise Summit (DOES16)! I just want to say thank you to all the incredible speakers, sponsors, staff and, of course, our participants, for making this our best conference yet. This year, IT Revolution, along with our Founding Partner Electric Cloud, hosted more than 1,300 attendees in San Francisco for three days of DevOps sharing and learning and mutual support.
Many people commented to me, both during and after, about the infectious energy at the conference! I agree! My personal belief is that it’s because, like any great conference, fantastic things happen when you put like-minded thinkers with similar goals together, who have a passion for learning and sharing the outcomes of pioneering practices. My genuine hope is that the DevOps Enterprise Summit help this amazing community, both as lifelong learners and as leaders who are transforming how technology work is done in large, complex organizations — overcoming incredible obstacles.
Here are some of my favorite moments from the DevOps Enterprise Summit 2016, beyond just the amazing talks, many of which you can see already here. We will continue to post the videos as we receive approval from the speakers, so keep checking back often!
DevOps Workshops Lean Coffee Sessions
As in previous years, we had two types of talks: experience reports and subject matter expert talks. However, we added a third format that we called “DevOps Workshops” that were small-group discussions using the Lean Coffee format, with a small army of trained facilitators at each table. Read more about the structure of the Workshops in this blog post.
Holy cow. There was phenomenal turnout for each of the four sessions we had, with several overflowing. There was even a group that didn’t have a table, so they used the podium, but they still had a great session.
I sat in on the session that was facilitated by Scott Nasello from Columbia Sportswear on information security and compliance. It was awesome! I mentioned that I took some courses and workshops from various universities, and holy cow, I wish they had used this format because I think it really helps achieved the desired learning objectives.
I especially love that we each group captured any epiphanies or discussions to be continued — here are some pictures to give you an idea of the high-quality interactions that occurred.
The DevOps Handbook Book Signing Party
On the second night, we had a unique opportunity to have three of the four coauthors, Jez Humble, John Willis and I, assembled to sign copies of The DevOps Handbook. (We had a different combination of us assembled at DevOps Enterprise Summit in London earlier this year: Patrick Debois, John Willis, and I. The goal is to get all four of us assembled sometime!
All conference attendees received a copy of the book, thanks to XebiaLabs, who hosted our book signing in their booth. It was fun to see so many of the people who helped contribute to the book, as well as influenced and helped me over the years.
I was particularly delighted that Eunice from XebiaLabs was able to capture some video evidence that countered the claims of John and Jez that I am always the bottleneck.
Here’s a quick video that we played during opening remarks the following morning:
(Actually, as hilarious as I think the video is, it would be wildly inaccurate to say that “I was never the bottleneck.”)
To expand upon the point slightly: the book signing was interesting because the process times (i.e., cycle times) were short enough that you could see vividly the consequences of variability. I’m guessing that the takt time was around 15 seconds — in other words, the tempo of which signed books were being produced by the three work centers was one book every 15 seconds.
It doesn’t take much variability at one work center to disrupt flow, leading to books piling up in front of a work center. In the case in the video where I was having so much fun talking to Matt for over a minute, four books piled up—which makes sense: 60-second delay means four books got queued up.
When you’re behind, books get accidentally re-ordered, accidentally double signed, etc. Hilarity ensues.
Pauly Comtois, VP of DevOps, Hearst Media on Making Middle Management Great
I didn’t want to mention any specific talks in this post because so many talks deserve mention. But okay I’ll bring one talk that if you missed, you’ll want to go see because I’m pretty sure you haven’t seen a talk like this before.
Pauly Comtois is now VP of DevOps at Hearst Media. He told me a story over a year ago about some of the things he was doing to help win the hearts and minds of technology groups across his large organization, including showing at other people’s 2 a.m. outage calls, just to help. Talk about a way to make new friends!
How middle management is characterized by many in the DevOps community has bothered me quite a bit. Of course, there is some truth about how DevOps changes the role of middle managers, especially when moving from a heavily siloed, functional orientation.
But by no means does DevOps eliminate middle management — after all, you can’t have thousands of highly autonomous “two pizza teams” and then a CEO. There needs to be more structure to ensure that strategic goals are met, as opposed to just team goals.
Many would argue that “middle management” have some of the toughest jobs of all. They own both execution of strategy as well as the execution of daily work. They must ship projects and features, keep things running, and also potentially go in front of the board of directors to describe why an outage occurred in a peak traffic period because they’re close to where work is actually performed.
In fact, if we define “middle management” as 2nd and 3rd line managers (i.e., directors, senior directors, VPs), that’s actually very much the DevOps Enterprise community.
Pauly’s talk is a fable about a DevOps leader trying to help middle managers make the transition from the old world to the new. It conveys in high fidelity the hopes, fears and dreams of middle managers, and all sorts of wonderful aids in describing this transition.
And plus, it’s freaking hilarious. Watch it and enjoy!
Published at DZone with permission of Gene Kim , DZone MVB. See the original article here.
Opinions expressed by DZone contributors are their own.