There were also a lot of assurances that "CentOS Linux isn't going away" when Red Hat acquired them. Stream was a tough sell, but many of us accepted and embraced and defended it. The move last week removes some of the basis for the good faith that those arguments depended upon. So, on one hand, it is about last week, but on the other is the incredible damage to the brand among people who have been Red Hat customers, contributors, and loyalists for so long.

It fundamentally feels like a redefinition of what Red Hat is about. Now, I know there have always been people in Red Hat who wanted to cut off downstreams and bring stuff in under more control. There has always been that tension. The times when Red Hat gave up that control to the community were often the highlights of good decisions, like merging Core and Extras in Fedora and giving the community governance of the entire project. That move also benefited #RHEL greatly.

It's also not lost on me that many prominent community facing leaders who championed Red Hat's cause in #opensource circles, such as Karsten Wade and Ben Cotton, were among those abruptly laid off recently in a choice made by #RedHat, and not IBM.  This is another telling of what Red Hat now values and what it believes are expendable.  That now "expendable" part was the very flagship of the brand that won so many of us to it.

Naturally, this is causing some "soul-searching" about my own alignment with the brand, personally and professionally, and to what extent that matters. I do not believe these changes and rhetoric reflect the community or values of Fedora and don't see that changing. I am sad to see some leave the project over it. I am not planning to do so. But I am frankly concerned over what's next for Ansible and Openshift and how far Red Hat will take this new mindset.

There has always been an area of trust in between the line of what Red Hat could and wouldn't do and those lines have moved a number of times in recent years so it's not unfair to ask if Red Hat will cut off their non-commercial versions of AWX, OKD, Quay, etc. It's completely within their rights to do so.

Oracle is proof that you can convince shops to give you money through fear of a shakedown or anti-competitive move. It is a way to "extract value from a product", but not one that will win the hearts of your customers and community.

@vwbusguy The viability of this strategy, long-term, depends on switching costs. Converting an Oracle shop to (say) Postgres is a major project for any company, even if they have good test coverage on most of their code and make relatively little use of Oracle-specific PL/SQL, etc. -- no DB's SQL is fully standard, and Oracle's less so than most. (Ask me how I know.) RedHat to SuSE or some Debian derivative may be less of a leap.