KubeCon Observations

Published: by

Two weeks ago, I attended KubeCon/CloudNativeCon EU 2018 in Copenhagen. The sheer size of the conference was astounding. Over 4,000 people attended.. In addition to the sheer size, the professionalism of both the conference itself - audiovisual, presentation, organization and administration - and the sponsor booths was very impressive. I have always enjoyed Linux Foundation events, very warm and friendly, but a little, as they say in Yiddish, “heimish”. They contrasted with the professionally run conferences you could attend put on by other organizations.

No more.

This cloud native conference was organized at the highest level, from food to layouts to registration to security to timing to audiovisual. Clearly part of the credit goes to the hard-working stuff of the Cloud Native Compute Foundation, but also partially to some key individuals who gave consistent feedback to the CNCF as to how to “elevate their game”.

What struck me most, beyond the size and professionalism - and hats off to the two leads, Liz Rice and Kelsey Hightower, both extremely smart and nice people who give freely of their time to help others with no attendant benefit to themselves - were the following two observations:

  • Real companies
  • Centre of gravity

Real Companies

Let’s start with real companies. Most of the time, tech conferences, especially those that are advancing the art and science of software and infrastructure, are attended by the practitioners of the art. The attendees and sponsors are mostly on the “sell-side”, people building software and services to sell to others, often large firms.

The growth of the Internet has changed that - large companies with serious open-source initiatives like Yahoo (RIP), Facebook and Twitter use these technologies in-house - but the overwhelming majority of the market for technology is real-world companies that sell things other than technology: financial services; cars; hotel rooms; airplane tickets; tables and chairs; corn on the cob; etc.

This is the first time I have seen a major leading-edge technology conference where a significant number of attendees are from real-world companies. These companies are - mostly - not looking to contribute to Kubernetes or create a new cloud-native project. Leave those to Synadia or Heptio or the dozens of others. No, these companies are looking to see how their practices can change using cloud-native technologies.

I do not know the breakdown of attendees’ organizations, although I would be interested to see if the CNCF would release some statistics, but the presence of these real-world mainstream companies was palpable.

When mainstream technology users (as opposed to creators) attend, your technology and, more important, your processes, are becoming mainstream.

Centre of Gravity

Up until fairly recently, the centre of gravity for containers was DockerCon. I attended DockerCon EU 6 months ago (ironically, in the exact same place as KubeCon was held this month). It wasn’t that long ago that there was no KubeCon at all, let alone a Cloud Native Con. The Linux Foundation’s container-focused conference was a mash of LinuxCon with ContainerCon, which become the impossible-to-pronounce LinuxCon/ContainerCon, or “LCCC”. Last year, LCCC become Open Source Summit (OSS - not to be confused with Open Source Software), and the higher-level utilization, orchestration and container “everything” moved over to KubeCon/CloudNativeCon (which is even more of a mouthful).

What became clear to me this month is that the centre of gravity for all things container has shifted to KubeCon/CloudNativeCon. This is where new ideas are announced; it is where open-source thrives; it is where venture capitalists go to learn of trends and meet new startups.

Interestingly, a number of companies that have been sponsors of DockerCon for years now expressed to me their hesitation about sponsoring DockerCon (coming up next month in San Francisco) at all. Some considered dropping from gold to silver sponsors; others said they either wouldn’t sponsor at all, or it would be the last time unless they saw a significant change in focus that convinced them otherwise.

Why has this happened? I only can speculate. Part of it is a focused push by the Linux Foundation, and eventually CNCF, to become relevant and important. Part of it is the maturity of Kubernetes and its growing ecosystem. Part of it is missteps by Docker Inc. in managing its own conferences and relationships with the market and open-source.

In the end, though, why matters less than the fact that it has.

Does this mean DockerCon is dead? Hardly. I suspect they will have a well-attended conference next month (I intend to be there as well, although I debated it for some time, until a client asked me to go on their behalf). However, unless something material changes, I expect to see more and more of innovation, focus and froth to happen at KubeCon as time goes on.