Bridget Kromhout

Ops Sundries

The Blizzard of the World

Minnesota’s north woods in winter, though a study in snow, aren’t exactly monochromatic. Peeling birch bark, glistening icicles, fir boughs burdened with a textured dusting, flat matte of a frozen lake, idle flakes blowing upwards, ever-so-slightly periwinkle-tinged sky: a boreal forest in winter is replete with variations on white.

I’m ending this year as I usually do, in a cabin beyond the reach of cell service (but with internet!), enjoying friends, food, snowshoeing, and pesky squirrels balanced on the bird feeder outside our window when enormous blue jays don’t chase them away. Sitting cross-legged in front of a roaring fire, before we play Settlers of Catan, I’m contemplating my 2016 of airports and stage lights, of jet lag and seeing the world I’m working to improve. Despite everything we can all say about 2016 (and there’s plenty), I’ve found delight.


By the numbers, this year on the conference circuit: I have 29 speaking gigs listed on my website, with variations on a theme: 19 cities on 5 continents. 16 talks, 6 keynotes, 4 panels, 2 ignites, 1 tutorial – mostly conferences, with a handful of meetups or internal events – and I’m not even counting the non-public customer stuff! (If you liked the Agile 2016 presentation I gave as a conversation with Andrew, just imagine the customer whiteboarding session we did in yet another city the next day!)

Travel like this is ridiculously improbable in most people’s lives. I’m fortunate that my spouse Joe has chosen to work part-time so he can accompany me on most of the international travel. Sure, it’s work, but it’s also fun when we go from Cape Town to Paris to Bangalore to Delhi, from London to Budapest, from Copenhagen to Stockholm. Add a couple of Toronto trips and one to Vancouver plus a visit to Havana because why not, and round the year out with a jaunt to Sydney. I’m getting tired just thinking about all the jet lag, but between criss-crossing the US and venturing to the rest of the world, I get a chance to meet with many amazing people. If we saw one another, even in passing, then hi!

But even with that much mic time, I don’t scale; an even greater impact I had was in leading the growing devopsdays organization. 42 conferences on 6 continents in 2016, with a typical attendance around 250, means that upwards of 10,000 people had a chance to learn and share and grow. I’m humbled and delighted by that!

In the Pioneers-Settlers-Town Planners model, I’m pretty firmly in the Settlers camp. I don’t need to be first on the ground (Patrick did a great job of that here.) I like looking at something great and asking, how does this scale? How can we make it repeatable? How can it operate without a single point of failure? (And you’re saying yes, Bridget. #opslife. We know. True story: team ops for life.)


Nearly doubling the number of devopsdays events (from 2015’s 22 cities) wasn’t a specific goal. I knew I wanted to lower the barrier for entry that might dissuade new organizers, because I wanted these events to be ubiquitous. Travel for conferences is expensive (even if someone else is paying) and time-consuming (sadly, teleportation is not available and the Concorde stopped flying before I could have considered taking it). Limiting conference access based on “let’s guess who has the ability to leave home for days at a time” further homogenizes tech (even if our conference price isn’t as high as with some larger events). Learning and sharing in your local area has follow-on effects of building community and connecting people with employers (and I say this even though I work remote and travel all the time! because I still love my home and want it to be great.)

Growing an organization means learning to let go, just like scaling your infrastructure means it will get too large to fit in your head. Even though I spend enough time on flights to have a favorite snack (Biscoff cookies) and least favorite seat (bulkhead), I knew we wouldn’t be growing devopsdays by having me run (or even go to) them all.

Finding new people excited about running an event involved some direct outreach. I mention it in every conference talk, which notably led to the fast turn-around of Cape Town, where the organizers met me in March and ran their event in November. And I convinced my former DramaFever co-workers Pete and Yash to run one, because a year remoting to Philadelphia means it has a special place in my heart. (Watch Tim’s keynote from devopsdays Philly! So good!) But even for recruiting others, I don’t scale (oh, do I ever not scale), so I also edited, clarified, and expanded the guidelines for organizing a new event. This has made it possible for people to attend or hear about a devopsdays, visit the site, and imagine themselves organizing one.


Coming into a devopsdays core team already in progress means appreciating the wisdom of long-time members who still make time to weigh in, while listening to new ideas from other perspectives. When I joined the core team in January 2015, I asked every core organizer if they still wanted to be active (as shrug-emoji isn’t useful for listing participants on a website). In 2016, I also recruited a few more core team members. If growing your community’s leadership, I suggest looking to the people who are already reaching out and helping others; works best if they’re intrinsically motivated.

Slack has been a great way for organizers from different cities to share tips of what’s worked for them, get speakers from other cities’ organizing teams beyond the core organizers, and (hopefully) avoid making missteps that others have already encountered. The core team can weigh in on email discussions, but it’s even better having access to hundreds of people worldwide who’ve experienced similar things. There are upwards of 350 local organizers, and they help one another on Slack – and a few had enough focus on helping other events that it was useful to ask them to be core team members (which of course comes with expectations!)

Seeing organizers from so many different employers (including competitors) working together on these community conferences is gratifying. Often our companies sponsor, but the goal is always building the local community and encouraging sharing between participants, which gives devopsdays a different flavor from a more commercially-focused event, whether single-vendor or corporate-run.

Oh, and in 2016 I overcame my reluctance and updated the website to list myself as leading the global devopsdays organization. Naming the work I was doing felt strange, possibly because women are so often socialized to deflect attention and credit to others, but it’s true and accurate. And it’s also important for other women to see that yes, a woman is running this tech organization.


One surprise I encountered when I started working in Cloud Foundry tech advocacy at Pivotal is how many people want opinionated guidance. After years in ISPs, academia, and startups, I had developed a habit of “figure it out yourself and get something working”. Now that I’m at a vendor, I also respect the fact that often, people want both enough guidance to get started and enough help so they can focus their limited time and attention on what matters to them.

When you make the decision to run a conference, you have a lot more to do than think about the mechanics of the website (and ours had become complex and brittle with time). I knew I wanted it to change, but such a change requires dedicated time, skill, and attention. Enter Matt, who’s the founder of the Arrested DevOps podcast I co-host with him and Trevor. (Side note about ADO: so much fun. So many great guests this year – too many to name – but do check it out.) Matt’s also has been running devopsdays Chicago for a few years, and in 2016 I asked him to join the core devopsdays team because of the work he is leading on the devopsdays website. Sure, it looks more or less the same (for now – that will change in early 2017, though!) but in 2016 he moved the site from webby to Hugo, and that’s made it far easier for new cities to get started or existing cities to update their pages. (Also, I don’t miss needing to use Ruby 1.8.7.)

Like I’ve been saying in conference talks all year, tools won’t solve all your organization’s problems, but good tools do help. The best tools provide enough guidance and usable defaults to get started quickly, while they’re also flexible and extensible enough to let you grow as your needs change. (You still need to be willing to change, which is so often that cultural shift that orgs need even more than new tools.)

And opinions don’t only come in the form of your opinionated tools. The template I put together for a new devopsdays conference comes with a code of conduct. Events are welcome to rephrase it, but we’re not going to merge their initial pull request without one. My opinion is that making sure everyone in your community has an opportunity to participate in shaping the conversations and to benefit from the interactions they’ll have at the event means you’ll have a better conference; we find this to be the case in Minneapolis.

When I do event kickoff hangouts, I remind organizers that they may have to reach out to potential speakers beyond those who submit to their CFP unprompted. Local entrepreneurs, bootcamp instructors, or meetup organizers may have completely different networks and experiences to share, especially if they are part of under-represented minority communities. If we want to build tech that serves everyone’s needs, including everyone in the conversation is how we make that happen.

The Future (not just a Leonard Cohen song)

Sorry not sorry: there is no perfect future-proof solution to everything you need starting from this moment until the heat death of the universe. Iterating within reasonable constraints works better than waiting until something is “perfect” (lolwat). Running a conference makes that obvious; running a conference-generating organization, doubly so.

Pivotal marketing asked me to contribute to a predictions post for 2017, and I talked a bit about observability (because I listen to Charity, as we all should). What I know is that looking at the results of our work for 2016 will help us make a better 2017, in every area where we can effect positive change. That means volunteering, donations, political activism, and simply considering how your choices can make the world that much better.

“Save the world” is a grandiose scheme that’s pretty tough to implement (even if it’s my not-so-secret plan, because under all that ops cynicism I’m an idealist). “Make it possible for your co-workers to have a better day” is a more reasonable scope, and that’s where devopsdays comes in. I’m hoping to see you at one in 2017, or, better yet, hoping to hear about how you want to run one in your own town!

As for my town, snow (or lack thereof) is a pretty big point of discussion for months around these parts. How much is going to hit during rush hour, what plans must we adjust because of it, why isn’t there as much as when we used to build multi-level snow forts as kids, which sidewalk is it making impassible for the less sure-footed among us… but winter leads to spring, snow pack leads to melt which we need to grow food, and just because the future is going to be different doesn’t change the now.

Grab a shovel, fill the bird feeder, build a fire. Winter is coming, but winter’s not forever.

Thanks to Tim Gross for beta-reading. All errors and #questionablelifechoices are my own.