Bringing faux open source into the world isn't a justification for adopting an infeasible business model and then complaining that your business doesn't compete very well.
I would argue that was precisely the issue with Redis and its friends. As a rule they want to get credit for being an open source project and contributing to the global commons, but without actually contributing to the global commons.
I'm not going to knock people for charging money to write proprietary software. If that's how you want to approach business dynamics as a software author, then by all means.
But trying to make money by extracting rent through a proprietary hold on your "open source" property, even as you claim to be open source, is too cute by half. Which one is it? The OSI definition hasn't substantially changed since the 90s, it's not like people can act surprised by what counts as open source.
There are ways to try to make money from open source, but they often involve leaning into the commons aspect and then offering a proprietary license as a relief valve for organizations not ready to have to pitch in, but who would be willing to offer up money instead.
Absent that, if you're literally going to be outcompeted on a business perspective on software you wrote, I can scarcely imagine what to tell you.
Of course not, their individual contributors formed business models that make sense, and work at those companies instead. Companies that are, I must again reiterate, linked directly from the official PostgreSQL.org website, in case there's any confusion on pedigree.
"Our consultants are recognizable from their many contributions to PostgreSQL"
"Our co-founders have written several books ... [co-founder] is also a core developer and project steering committee member on PostGIS and pgRouting projects"
"With a strong focus on PostgreSQL, they recently launched Percona Distribution for PostgreSQL, which delivers a single source, enterprise-grade, open-source installation of PostgreSQL Core Distribution." (Custom sales on software products based on the open-source original? Sounds almost like what AWS did with Redis)
Obviously these companies are likely to give back monetarily to the things needed by the open-source team, if only because their viability of their own business relies on having the open source product to sell expertise around. But a lot of companies simply use PostgreSQL and pay nothing, and that's fine too. It's all part of what it means to be an open-source product.
There's ultimately no reason Redis Labs couldn't have been successful as the Percona or KDAB of their own product's ecosystem. I guess they figured there was more money in hosting than services, or maybe Redis was simply too foolproof to need consultants?
