So, about a month ago I was let go from Docker. In my mid-50’s, kids still in college, and though I was interviewing, I hadn’t actually secured anything yet. As of now, I have secured a new role that I can’t quite yet reveal (and let’s be clear, while I am excited, as you may know I’m not about ‘big reveals’ and such… I just haven’t signed the paperwork yet). I was – am still – bitter about it…
Read more: The only thing constant is change…
Docker wasn’t even a good fit for me – or me for them, quite frankly. I expressed my surprise to my boss on more than one occasion that they actually made an offer. As far as Docker is concerned, I’m a dinosaur! Thirty-plus years in the industry, and it’s the wrong industry – my earliest certifications were in Windows NT 3.51, I have worked with / for multiple VARs/resellers, Citrix, EMC, Datacore, and VMware. I don’t have a background or education in computer science, and while I have contributed a little to some OSS projects, that certainly doesn’t make me a developer.
But Docker is trying to change from an open source company to one that builds and sells enterprise software, and that was exciting for me. After the acquisition of VMware, I was ready to get far away from Broadcom – my last months there were a major turn-off, and I still mourn the loss of the culture that VMware created. I can still recall the first sales kickoff I attended in 2008, when Diane Greene pointed out how unique the opportunity was… The opportunity to inject a new layer into the datacenter stack – not just introduce a new computing architecture, but fundamentally alter how datacenters are operated. And she was right – just look at the impact virtualization has had on the design of servers, storage, and networking… so much in the industry has changed because of VMware… Yet in a masterclass of corporate pillaging, Broadcom hollowed out VMware—cashing in on loyal customers while burying its spirit of change and invention.
So I left, and joined Docker, ready for a change. Ready to join another iconic company that has fundamentally changed the industry. I was not prepared, however, for the massive culture shift, the vertical learning curve, and the completely different nature of their go-to-market (GTM) process. I was used to the extremely formal (perhaps even rigid) release process that VMware had built over time, and found myself completely flummoxed with how to operate in this new environment. I didn’t have any of the internal communication paths I expected to have, and none of the pre-release training I was used to – often new features left the building before I was even aware of the project. I had expectations of what it meant to be in the GTM organization, and I hadn’t ever worked for EITHER a company that builds developer tooling, nor an OSS company. My career has been 100% datacenter infrastructure software, and my customers have all been part of enterprise IT organizations. I found myself unmoored, floating in what felt like a sea of chaos, and I allowed my perception of my own inadequacies to turn into frustration.
On the other hand, Docker is trying to do something that is very challenging – find a way to make money to continue their long record of innovation, without losing their absolutely ENORMOUS base of users… Docker Hub boasts over 11,000,000,000 downloads every month – that’s right 11 Billion with a capital ‘B.’ By their own best guesstimates, Docker has over 27 million users of Docker Desktop globally – and that doesn’t take into account Moby / Docker CE (engine) running in pipelines. While it is true that today Docker does not offer a supported, production, runtime environment, the impact they have had on how software is build is incontrovertible.
So I get it – what they are trying to do is difficult – how to make money off their existing users, gain additional market share, and avoid alienating everyone? And to be clear, they are building some extremely cool tooling – right now. If you have the time, be sure to check out this video by Bret Fisher on his Cloud Native Devops channel – they do a great job of recapping everything Docker has done just this year. And while they are building some really excellent tooling, I think they are still struggling to figure out what it means to be an enterprise software company, struggling to find product market fit, etc.
Honestly, being laid off was probably best for both of us – I have a very hard time hiding my feelings, and I am sure my frustration was readily apparent. As Docker gets its ‘sea legs’ (had to drop at least one nautical reference in there) in selling to the enterprise, I believe they are setting themselves up to have another major impact on the industry.
So while it is hard for me to acknowledge, I have to give credit where due. They taught me an absolute, metric TON about what it means to be a developer (note that I am not claiming I am now a developer). They taught me:
- what an SBOM is, and how to secure your software supply chain
- how to containerize a previously non-containerized application
- what unit and integration tests are, and how to write unit and integration tests (at least in Python)
- how containers can fundamentally enhance productivity both in development and in continuous integration workflows
- what an agentic application is, and how containers can and will deliver all the same benefits to agentic applications as they did to cloud native applications.
- … and so much more
As for my next gig, I will be returning to software infrastructure, a domain that I know well and have excelled in previously. But I will be returning with a newfound appreciation for the beginning of the software development lifecycle, with new connections, friends, and perspective. Maybe I will even be able to continue to present at some of the local meetups.
My Occasional IT Blog... 