Many reading this blog are familiar with Dr, Elizabeth Kübler-Ross’s “Five stages of Grief” and probably mid-career I started thinking about applying it to open source software adoption in companies large and small. Here’s how it goes…
Stage 1: Denial
About a decade ago, I was invited to speak at the engineering summit of a manufacturer of, er, flying machines1. My contact lead with:
“Just so you know, we don’t use or allow the use of any open source software anywhere in our company. So you’ll be talking with an audience of people new to the topic.”
I reviewed their product line and after a short review of their ‘shipping2’ products, I was able to lead the talk with praise on their use of open source compilers and libraries clearly in use throughout their website, production hardware and more. I thought it showed how innovative and smart their engineers were to use the best tools available to ship the most reliable … things that fly… that they did.
I focused my talk on how they could continue to count on the tools they were using and where I felt their patches and improvements to those tools would be welcome and avoid endangering their position in the industry or violate ITAR. It was well received, I think, and I’m egotistical or foolish enough to believe my contact when he praised my performance.
Stage 2: Anger
Early in open source’s history , and indeed in many of the projects I would work on to free up data or code, I would often be presented with accusations of being a spoiler, killing programmer jobs, to destroying industries. There’s the famous ‘cancer’ accusations from a long past Microsoft, but I always tried to take these accusations in stride and see them for what they were: People having problems showing the value of their work. They would lash out at free and open source alternatives instead of looking into their actual selling product and seeing what’s working and what isn’t.
Getting past this stage isn’t easy, and many, well, founder and fade away in the anger gap….
Stage 3: Bargaining
One thing that is frighteningly common stretching back 30 years now is people asserting things about open source licenses , or adding stanzas to licenses, that simply aren’t true, useful or relevant.
23 years ago I was on a panel with a prominent MySQL sales executive (this was before the sun acquisition) and he asserted that “If you’re running MySQL in production, you should be licensing every cpu core you’re running on from us, that’s what the GPL requires”
Now this was a incorrect interpretation of the GPL3 and I gently corrected him that wasn’t how the GPL was written at all. At the time, they were having real problem getting past ‘support contract’ revenue to show ‘software license’ revenue to their investors.
Let me let you in on a secret: If you’re giving software away it can be awfully difficult to capture ongoing per-copy revenue. They would go on to change the license of the client code from BSD to the GPL and back again depending on how the user community reacted to it, and it didn't’ change the underlying problem one bit.
At the time, people were experimenting with ‘addons’ to licenses, that made the code no longer open source (again, a longer article) and some would even berate open source and free software developers when they would attempt or succeed at competing with their proprietary offerings.
I always found this odd. In my experience Open Source software would not win against superior technology just because it was free, but because it did something for the users and developers they could not get out of the proprietary offering. Even when co-existing with powerful open source alternatives, proprietary software, properly sold and engineered, seems to be doing quite well.
Take VMWare: Xen came out in 2003 and the Linux Kernel developers released KVM in 2007, and yet, VMWare is still going strong despite the ‘VMWare tax’ jokes amongst CIOs. If anything, That Xen exists and was adopted heavily by the hyperscalers early makes VMWare’s omnipresence in IT even that more interesting.
Stage 4: Depression
“How can we compete with free!” is a lament that I’ve heard so often. It’s unkind, but it does seem to me that if you can’t compete with free, maybe you aren’t meant to. I know that sounds harsh, but read on…
For decades, the miracle of software licensing was that you could charge many many times for things that cost nearly nothing to duplicate. But at some point, users of software and software developers in particular, were done paying for software long settled. It’s easy to, well, succumb to depression. It’s much harder to see what people value about your product, software, or company and lean into that.
Look at Wordpress & the commercial WordPress service out of Automattic: Wordpress isn’t exactly *hard* to host, but hosting it well and securely and keeping up with the innovations and security patches in that community isn’t easy. So when Automattic said (for what I consider to be a remarkably reasonable price) they can take those worries off your shoulder? Great. Make no mistake: It’s a remarkable, and remarkably huge business.
Again, larger article, but getting past the failings of your commercial offerings and creating real value for your users is the trick to getting out of this stage.
Stage 5: Acceptance
Understanding open source as a resource for your proprietary service, software, hardware, or offering is the first step to being an effective development or IT organization. Further deeper understanding of how you can take part in an open source developer community as a user, patcher, addon writer or even as a maintainer is often within companies grasp, should they chose to go that way. But, seriously, be a great user of open source software, libraries, hardware and then decide how you can help maintain the resource you’ve gotten so much value from.
Again, that’s a whole another article, but the more you help and take part in an open source developer community, the more insight and influence you have over the direction a project might go.
That’s pretty much it for today, and while it’s not as entertaining as my application of Dr. Kübler-Ross’ stages to the acceptance of Richard Stallman, it is just a little substack post before the Harris-Trump debate throws us all into stage 4….
Not Boeing! Different company altogether.
I mean, they shipped flying stuff, so not like people were downloading them off the internet…
See how nice I’m being?