I'm a huge advocate of free software, but mostly for personal use and individual freedom. So I don't feel like the loss of Scylla is a great loss in that respect. I can't imagine a lot of individuals are harmed by this move.
But still, it makes me distrust corporate "open source" more and more.
Buuuttt... projects like these would probably never haven gotten the traction they have now if they wouldn't have started out as opensource but being closed source. For example Hashicorp and Redis definitely wouldn't be where they are today if they had start out as opensource. So in that sense these license changes are a bait and switch.
What makes you think there are genuine open source projects that don't get traction? Ansible is still GPLv3 even while owned by RedHat^W IBM and works fine. Any one of the bazillions of front-end toolkits, build tools, bundlers, whatever, many initiated by some company and practically all under open source licenses.
My experience has been that if something is useful, and its open source license means one can fix bugs they encounter or at least have a small chance it'll remain around indefinitely (not go out of business) in order to bake it into your workflow, then it'll be adopted and blogged about and show up in HN and Reddit
There are plenty of corporate projects that don't require rights assignment to submit changes, just a DCO or even just $(git commit -s)
In the 90s, we had shareware/freeware/beerware. No one expected source availability. At universities, the Unix world had open source, in the real sense. Then something happened that caused these two tracks to merge.
Maybe the university students grew up and carried on their belief in open source, but also had to create the business. After 20 years, they realized (1) it doesn't have to be open source to _feel right_ and (2) maintaining in the open is more expensive. They're now in management positions, not coding. Selling B2B doesn't require code to be open to the public, since you can have source available licenses. And as you said, perhaps ScyllaDB isn't really targeting the hobbyist. We'll continue this trend.
Or perhaps open source has stopped being a buzzword. We're now much more of a SaaS world, where being open source isn't as important as costing $10 per month or having 200 "data partners" that need to track you.
That said, I think there's a really good reason for core libraries and security-sensitive libraries to be open. I want to be able to inspect them, before using them. And I'm in the HN crowd of actually using open source code because I can fix bugs as I encounter them, but I realize this is a small crowd.
I'd love for open source to be a useful word again, and not something that goes on the Silicon Valley PR budget.
Nobody will touch your product if it's not OSS doesn't mean you should call it OSS and then rug pull. Build it proprietary from the beginning you cowards. The people who do this are the same as the corporate shills who infiltrate subcultures to monetize them and ultimately destroy the community in the process. If you don't actually share the values of the community you're trying to join, then don't join it.
This little SV "growth hack" where the success of your OSS project is the proving ground to get funding has to die. It's turning a high-trust community into a low-trust one.
It doesn’t make sense to people who contribute small hobby projects or work as researchers / are government funded, which means ultimately paid for by businesses and consumers thereof.
This is the primary appeal of open source software to most businesses - free software and free labor.
An open source project that doesn’t receive in kind development cost reduction benefit from free contributors should not stay open source.
Where it exists in a small community of other developers, most of whom have a similar mindset, it works fine.
And for very popular projects, there’s enough users for whom it makes economic sense to have some devs on staff to support their usage, tailor stuff, and keep it moving forward to maintain the project.
But in the middle, where most things are, it kinda doesn’t work.
I understand it will have limitations on the data that you can use (up to a certain size) and the CPU power (up to these many cores). Will it also limit the activity type (commercial, SaaS offering, ...)?
Although a FAQ is not a real substitute for the actual legal text of the license.
The other interesting example is Copyleft software, where "Corporation" does not have complete copyright holder. For example Percona or MariaDB can't "close source" their MySQL forks of the core software, though it does not prevent them from doing it with other parts of the complete platform, think MariaDB MaxScale.
If they did not get back from community it may mean that their code was bad for contributions.
Sorry!
They can keep the unmaintainable mess to themselves because then, the code is available but not really open.
License is one thing for openness, code quality is the other.
I doubt anyone criticizing is doing so because of "How dare they want money?"
Which just begs the question, why did you think that? That take just exposes a complete missing of the point of open source software and why (honest) people invest their time in it.
There are others who just want the caché and benefits that come with producing open source, but without the annoying actually being open source parts.
If you want to get paid for renting out copies of software, just do that honestly. If you want to make the world a better place and pay it forward for all the free stuff you were given, then do that honestly.
But wanting the benefits of sainthood without having to volunteer at the soup kitchen, no, audacity is not the word for that.
And merely wanting a comfortable living is not audacious, and no one who's not an idiot thinks it is, and is actuall othogonal to whether you have any integrity wrt open source licenses.
You can want, and have, that comfortable living either way.
