@espectalll if you use permissive licenses you don't have to worry about abuse

@feld actually that's exactly the issue they claim to have, that they're not getting contributions back because of loopholes on AGPL making it too permissive

@espectalll If your software is not trivial and if your software is being actively developed people don't successfully fork it. It's much easier to submit your changes upstream than to pay a team to constantly try to rebase their fork. This has been proven time and time again in the BSD ecosystem. Those who fork regret it and suffer the consequences of millions of dollars of technical debt.

@feld @espectalll I hear what you're saying here, but isn't Amazon Linux (RHEL fork, approximately) basically a counter example?

@telent @espectalll is Amazon Linux beating RHEL at their own game? Is it more than just a rebranding or slimmed down image for their infrastructure?

I've never even heard of "Amazon Linux", by the way...

@feld @espectalll amazon linux is the path of least resistance if you want a rpm-based distribution running on your aws cloud instances: it has preinstalled all the aws bits you need for easily managing those boxes at scale. By making it easier to choose that over installing and paying for rhel - they aren't beating redhat at redhat's game, they're making their own rules up for a completely different game.

@telent @espectalll I don't understand -- are you saying that a distro re-spin/customization is somehow an evil fork? Because that's not even a fork, IMO. Not in the slightest. They're just optimizing the OS for their specific platform. I'm sure they stripped out all RedHat trademarks because it's not an officially supported OS and they're serving packages from their own internal mirrors which are just alternative builds of the SRPMs.

Call me when they've actually forked *code*. Like massive changes to RPM/yum/dnf, for example.

If anything we should be mad at RedHat for forking the Linux kernel and having patches/changes that are not upstream in vanilla.

The FreeBSD AMI images are also different from the FreeBSD installation media, too. It's not a fork.

@feld @espectalll whether they've forked the code or not is largely immaterial, I think. What matters is if they've established themselves as supplier of software (as a service) to people who would otherwise have spent money with "upstream" and thereby supported upstream development.

@feld @espectalll and for the record I didn't say anything about "evil". You could argue that they're eating red hat in the same way red hat took chunks out of Solaris/proprietary Unix - not saying it's good or bad, just that there are winners and losers

@telent @espectalll Is it a fork or is it the inclusion of some additional software and kernel modules enabled by default to support the platform?

Sounds to me like RedHat should include the support in an image and ship it. It doesn't seem complicated at all.
Sign in to participate in the conversation

Server run by the main developers of the project 🐘 It is not focused on any particular niche interest - everyone is welcome as long as you follow our code of conduct!