Cloud Native Computing Foundation KubeCon + CloudNativeCon North America 2021, 29 Oct 2021

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Keynote: Sustaining a Contributor Community’s Next Generation - Christoph Blecker, & Paris Pittman

Description

Don’t miss out! Join us at our next event: KubeCon + CloudNativeCon Europe 2022 in Valencia, Spain from May 17-20. Learn more at https://kubecon.io The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects.

Keynote: Sustaining a Contributor Community’s Next Generation - Christoph Blecker, Principal Site Reliability Engineer, Red Hat & Paris Pittman, Program Manager, Apple

A

Hello, everyone, sorry, I'm dancing, um I'm paris.

B

And I'm kristoff.

A

We're so excited to be here closing out the week with you we're going to be talking about kubernetes sustainability through our lens. It's been a long fun intense week already insert your adjective here, but we've made it through together, there's still more to do and everybody's been super kind in the process.

A

I really wanted to kick it off with some personal journeys of ours, so first mine, when I started on the project four years ago, a little more than four years ago now, I already felt like what would be labeled as the project's next generation.

A

If you will, at the time kubernetes was a year and a half old about 8 000 contributors, dozens of contributing companies already questions like how do we sustain after growth, we're already being asked intentional sustainability efforts were already happening.

A

I grew under sarah novotny, our first community architect, who once told me the belonging, is high with this group.

A

Indeed, I found out it is I put my work in with upstream governance, operations and community at the start, and it was noticed a few months in I was in a contributor experience, special interest group meeting hanging with my usual suspects, giving my usual updates and aaron krikenberger told the group that he'd like to nominate me to lead it.

A

I'm not going to get sad, and I was so taken back that someone not only saw me but thought that I could be a leader in an open source community as monumental as kubernetes, and then garrett, elsie george, and this guy right here, sponsored me to leave this group and together we worked on initiatives that led kubernetes from six from 8 000 contributors to 60, 000 that we have today.

A

Sponsorship is a tenet of our project y'all. It is shown through trust, building, consensus decision making and it's shown through our membership processes through our contributor ladder and roles.

A

As a woman in this industry, I can't tell you how critical sponsorship is a plus one can go so far open doors that one can't imagine. I definitely didn't see this one coming home.

A

Our values and tenants like this will sustain us into the future so long as we have folks showing up to participate.

B

I had a similar sense of how tight-knit this group was when I started getting involved back in 2016.. I was working for disney at the time on a project to adopt kubernetes in our organization.

B

We were fairly early in the adoption cycle and, of course, ran into bugs or under-documented features, and when I did, I go and read the code and documentation and propose effects.

B

It started with things that were directly related to my day job, but I got hooked and found myself continuing to contribute my off hours.

B

I found myself not only fixing issues with the kubernetes code itself, but also the process of contributing that's how I got involved in the contributor experience sig and met folks like paris as I continued to work in the community. I too started being recognized for my efforts, paris and the other sig leaders at the time, sponsored me to become a tech lead years later. Both of us grew within the project and were elected to the kubernetes steering committee.

B

We kind of live and breathe the stuff, and that's why it's so important for us to talk about holding the door open for the folks who are coming in after us.

B

One of the other early areas I got involved in was our test infrastructure.

B

We as a project, had to make significant intentional investment in our automation, testing and tooling. In order to scale we needed to be able to delegate abilities like running tests, merging code and even adding folks to our github work prow. Our ci cicd system has been so instrumental to the success of kubernetes that many other organizations and projects in the cloud native space have adopted it.

B

But behind the software there is a small, dedicated team of folks who manage and maintain it. It's no small feat either. Did you know it cost between 200 to 300, 000 u.s dollars a month to run our test infrastructure?

B

Still, some jobs can't be automated.

B

One of the more recent additions to the project is our production readiness review team. This team of talented folks conduct an in-depth review of every single feature that gets promoted to general availability.

B

We do this to ensure that the features we ship are observable, scalable, supportable and ultimately safe to run in production environments back in those early days when I was in that end user role, I can remember being caught off guard when a feature or api changed.

B

If I didn't diligently decode the change log and update my application accordingly, it was extremely easy to suffer an outage as a consequence, production readiness reviews, along with other things, we've done like a clear api deprecation policy uh to stop things from sitting in that perma-beta state have truly made kubernetes a better, more stable piece of software.

A

The thing that I like about the production readiness review team is that their roles are scoped and documented roles like this help communities grow, but not only that sustain go on live past ourselves.

A

Roles like this, our core roles like reviewer and approver. You know the people reviewing 40 000 pull requests a year are also scoped and documented. This is required for contributor communities of our size. According to nadia ekbal, an open source researcher we're rare. Only three percent of open source communities are like this.

A

We also have api reviewers, who have some of the best eyes in the business contributor event leads moderators who shape online community presence of hundreds of thousands. You heard that right, ci signal keeping the lights on and I could go on.

A

Many have shadow programs too that onboard new folks while they learn more kubernetes, internals advanced kubernetes and what have you to take back to their employers. This is how our ecosystem flourishes.

A

Folks, like alana in production, readiness and node ci approver are known in newly minted sig infra veronica as release manager, 40 new sig chairs tech leads committee members and more in this year alone. They are our next generation.

A

Did you know that the release team has had over 300 people come through its shadow program in this last several years? That is the world's one of the world's largest open source release processes.

A

All of these folks are volunteers, too, especially the mentors and guess what many aren't paid to do this frankly, depending on the role that you have in the project, some of this stuff could feel like two, maybe even three jobs and speaking of release, it's clear by now how many companies contribute to this project that there's real business value here, especially through enhancements, it's so much easier to tie a feature to a business goal.

A

Are you adding a feature upstream because it's useful for others and you don't want to carry a messy patch, great goal? What about the engineer on your team, who wants to write automation to manage our slack community of 130 000, which has no direct ties to your revenue?

A

These roles allow the project to continue and are at the heart of what we say when we mean chop wood carry water, this isn't just an award to us. This is how we've intentionally created our community.

B

So, between dev, tooling and meta infrastructure, production readiness and our documented roles at scale is that all it takes to sustain a project this size, the rumor is: we've hit all our initial goals, so we must be done, is kubernetes done yet not even close.

B

We have 170 enhancements still in the alpha or beta stages of development.

B

Not all of these will reach that end goal of general availability, but it shows that the number of new use cases for kubernetes is not slowing down even a little bit, but for every new enhancement that adds or changes the functionality of kubernetes there's an immense foundation that must be supported normally, when we're talking to folks and encouraging them to get involved in open source.

B

Lots of it is focused on those individual contributors. Hobbyists enthusiasts students and other kinds of individual contributors are important parts of our community, but ultimately people need to be paid for their work.

B

B

What happens when these talented folks get hired into the kubernetes ecosystem and then stop having time for upstream?

B

So I want to talk to the companies who use kubernetes for a moment if you're, a company that makes a kubernetes distribution and likely already has engineers contributing to code and features, consider really supporting your employees to take on roles like leading a sick.

B

If you're a telco or an airline that runs kubernetes emission critical scenarios consider supporting your staff getting involved in efforts like production, readiness or instrumentation work to ensure the things that you're deploying are reliable and observable, if you're a bank or financial services company where security is really important to you, consider getting involved in our release and security efforts around software bill of materials or dependency management.

B

If your team has skills like running kubernetes in production or managing and automating community tools like zoom or slack, this is valuable to us. Send us your people, but on company time.

A

Digging in and helping with help wanted issues is still a very important piece to contributing here and building trust. We need deeper than that. At this stage, all we need companies to collaborate with us to incentivize and support reviewers, approvers plus chopwood carry water rolls find the balance between your new enhancement with the work that needs to be done to increase our sustainability as both a technology and the community.

A

We know promotions are tied to enhancements, but is it a norm yet for companies to promote because of a critical refactor to an open source project?

A

Misaligned incentives is how we get to burnout contributors, leaving the project employee turnover in order to get more support, promoted paid. Whatever many of you justify your coupon expenses for recruiting talent, if you want to be competitive start with upstream support and incentives, remember the belonging is high. In this group.

A

We have huge goals that require focus to help keep this the ship.

A

We need we need full-time folks to get us there. Are you a manager out there with a contributor maintainer under you, ask them what they're doing with their upstream adventures and if they feel supported, but most importantly, do they feel supported enough to grow people behind them?

A

Remember sponsorship is how kristoff and I got here- mentoring for the next tim hawkins, clayton coleman and the rest of the gang at the root of kubernetes. Kubernetes isn't going to happen overnight.

A

uh How do we pass down this knowledge in a world full of casual contributors and misaligned incentives? We've heard and implemented thousands of your unique use cases and built this community on radical collaboration that leads by our values.

A

If kubernetes is important to you, we need you to participate and to the hundreds of you that have shown up to date. Thank you so much. Thank you.