Marshal

Sounds as a good idea to wait. I really hope RedHat time will drop this idea and keeps the development for free. If nothing changes in that area, Debian (even Ubuntu) with their decently long LTS support periods seems as an equality good alternative. I come from a different background, but I wouldn’t really recommend FreeBSD nowadays due to not too much community support. Ubuntu made its success as being most documented server OS in years. Pretty much any issue that you could experience was already experience by somebody else and the solution documented. Try to combine CentOS with nginx and use two or more different php-fpm versions and you will end up with some undocumented issues, since there lack of solutions for free, unless you are using their paid support. Well, FreeBSD is even less documented than that.

From what I can see is that RedHat is trying to cut down on costs of beta testing while trying to innovate. Most big software companies cut down on their Q&A teams since few years ago, which backfired in quite a few situations (remember multiple Windows updates that failed?). RedHat is probably trying to stay competitive and replace proper software testing by community reports.

One thing that boggles my mind is what is there to be innovated on a server OS? Kernel and hardware support is written by Linus and community around him, web servers as well as databases are written by 3rd parties, no new system utility applications being invented in years, so to put things together - nowadays a Linux OS comes down to a “wrapper” integrating all those tools together and providing proper infrastructure for repositories (and paid support).

If they stay with this decision, they will convert part of their users to paid RHEL plan, while losing community support and the market share as being the most used server Linux OS. I could be very wrong and I honestly wish I am.