McGrath: Red Hat’s commitment to open source

Posted Jul 3, 2023 19:30 UTC (Mon) by pizza (subscriber, #46)
In reply to: McGrath: Red Hat’s commitment to open source by jccleaver
Parent article: McGrath: Red Hat’s commitment to open source

> Community bug reports are a Good Thing here, and it baffles me that RH is so myopic as to think that it's moot.

I don't think anyone, least of RH, would disagree with the importance of "community bug reports", and indeed, that's one of their stated reasons behind the CentOS Stream model.

But you need to stop looking at this in terms of absolutes. Does the [now-]former rebuilder model offer RH some benefits? Undoubtedly! But do the benefits outweigh the price? According to RH, not any more.

I can't be the only one who thinks that a decision of this magnitude doesn't have some hard data backing it up. It's not hard to imagine multiple now-former customers telling their now-former RH reps "We're not renewing our seven-digit RHEL licenses unless you match Oracle's pricing" or even "We're going to just use Rocky as it gives us everything we care about and we don't have to pay anything at all to get it."

And that's what it all comes down to in the end -- "We lost $$$$$$ worth of license renewals to rebuilders, and our sales folks say only $$$ came in from folks that decided to go from a rebuilder to RHEL. Why exactly should we keep paying our engineers and hosting providers $$$$ to enable these rebuilders to easily take business from us?"

McGrath: Red Hat’s commitment to open source

Posted Jul 3, 2023 19:37 UTC (Mon) by pizza (subscriber, #46) [Link]

> I can't be the only one who thinks that a decision of this magnitude doesn't have some hard data backing it up

Ugh, s/doesn't have/has/

McGrath: Red Hat’s commitment to open source

Posted Jul 3, 2023 22:31 UTC (Mon) by jccleaver (guest, #127418) [Link]

The problem here (for Red Hat) is that adopting an antagonistic posture toward rebuilds won't force a long-term change here. The underlying code is open source; the patches are open source. SRPMs can be re-assembled, and OSS collaborators can simply work harder to try to get it back to the same level that CentOS Linux had been in the EL6 and EL7 era. At best it delays things, and the cost was a huge amount of goodwill.

Furthermore, this recent action really can't be taken alone, separately from the decision two years ago to clobber EL8 rebuilds halfway through the RHEL8 lifecycle. McGrath at that point clearly demonstrated on the CentOS list where he seemed to dig even deeper with every possible opportunity: eg, https://www.spinics.net/lists/centos-devel/msg19597.html

Oracle and Amazon can throw whatever amount of money they want at reverse-engineering EL. If that's their concern, then they haven't figured out yet that it's better to co-opt your enemies than fight them. Demonstrate what RedHat's value is and cash in some of that goodwill. Instead, they've burned their goodwill and Oracle and Amazon will do whatever it is they want to do anyway, and CloudLinux, Rocky/Alma, and NASA or whichever lab decides to work on Scientific Linux 2: Electric Boogaloo will find a way to formalize support for things that they now actually have a larger claim to have placed real value into, based solely on the amount of (re-)engineering effort they're now having to put in.

McGrath: Red Hat’s commitment to open source

Posted Jul 4, 2023 8:36 UTC (Tue) by paulj (subscriber, #341) [Link]

"I can't be the only one who thinks that a decision of this magnitude [has] some hard data backing it up."

You may think so, maybe others, but anyone with experience of suits at big corporations also know decisions like this often get made without hard data. Indeed, anyone with experience will know such suits often have a very strong presumption /against/ things like goodwill and intangible value being of benefit to the corporation's bottom line.