OpenELA Drops First RHEL, 'Enterprise Linux' Compatible Source Code (theregister.com) 39
Long-time Slashdot reader williamyf writes: In the ongoing battle between Red Hat and other "Enterprise Linux -- RHEL compatible" distros, today the OpenELA (Open Enterprise Linux Association), a body Consisting of CIQ (stewards of Rocky Linux), Oracle and Suse, released source code for a generic "Enterprise Linux Distro" (Sources available for RHEL 8 and RHEL 9). A Steering committee for the foundation was also formed.
War between Red Hat and what they call "clones" (mostly Oracle; CentOS, Rocky, Alma and others seem to be collateral damage) has been raging on for years. First, in 2011, Red Hat changed the way they distributed kernel patches. Then, in 2014, Red Hat absorbed CentOS. In 2019 Red Hat transformed CentOS to CentOS stream, and shortened support Timetables for CentOS 8, all out of the blue. Then, in 2023, RedHat severely restricted source code access to non-customers.
What will be RedHat's reaction to this development? My bet is that they will stop to release source code of distro modules under BSD, MIT, APACHE and MPL Licenses for RHEL and in certain Windows for CentOS Stream. What is your bet? Let us know in the comments.
War between Red Hat and what they call "clones" (mostly Oracle; CentOS, Rocky, Alma and others seem to be collateral damage) has been raging on for years. First, in 2011, Red Hat changed the way they distributed kernel patches. Then, in 2014, Red Hat absorbed CentOS. In 2019 Red Hat transformed CentOS to CentOS stream, and shortened support Timetables for CentOS 8, all out of the blue. Then, in 2023, RedHat severely restricted source code access to non-customers.
What will be RedHat's reaction to this development? My bet is that they will stop to release source code of distro modules under BSD, MIT, APACHE and MPL Licenses for RHEL and in certain Windows for CentOS Stream. What is your bet? Let us know in the comments.
Toxic relationships... (Score:1)
Red Hat aren't interested in you. Why do you punish yourself?
give Deb Mint-Buntu a call and move on with your life.
Open Source (Score:3)
Open Source is Open! OH NOES !!!
so about that source... (Score:2)
One of the key arguments for Open Source is transparency. Is that just not a thing anymore?
Re: (Score:2)
GPL says they can't do it.
IBM's army of lawyers says "try me".
Best bet is to move on.
Re: (Score:2)
(Full disclosure: I work for Microsoft and have a long enough memory to know I'm in no position to throw stones as long as they are buying my groceries.)
Prusa is doing some funny business with Open Source too because of clones. These are companies I expected to be bedrocks of open source. What's happening, and more importantly who is next?
Re: (Score:2)
Sadly, some experts say that, no matter how much they want to think otherwise, GPL says they can do it.
GPL allows you to discontinue access to new source code, so long as you also discontinue access to new binary. The most hopeful counter argument is that the threat of closing off support and update access somehow constitutes a "further restriction" and therefore that threat is not compliant, but many hold that the restrictions refer to the binaries and code in hand, and could not imply a forced ongoing bu
Nobody gives a shit (Score:1)
Re: (Score:2)
Red Hat has contributed a lot of important bits and pieces in addition to crap like systemd so it's a mixed bag, but it's a large one and there's a lot of stuff in there. I was just fiddling around with one of my KVM virtual machines and I noticed that bits of the driver stack had their name on it. Their slide into enshitification is a real loss for the community.
I don't know what will happen (Score:3)
Re: I don't know what will happen (Score:5, Insightful)
Re: I don't know what will happen (Score:4, Insightful)
It would set Linux back tremendously if RHEL and red hat died. Roughly 60% of the patches across all distros origininate in RHEL/Fedora/Centos. Canonical is to busy playing with snaps and SUSE doesn't contribute nearly enough and take alot of RHEL already. There aren't many more companies with vested interest in Linux with cash flow to fill the gap in a open manner.
That effort would flow elsewhere, indeed into many of the same packages maintained by the same people paid by someone else. Short term pain for long term gain.
I don't discount what Redhat adds to Linux, just that they don't want to share it makes them not a team player, which is both hypocritical with them still being an OS based on much open code they don't pay anything for, and also is completely anathema to the spirit of OSS.
Re: (Score:2, Funny)
The death of Red Hat would result in the death of Gnome.
Win-win.
Re: I don't know what will happen (Score:3)
And SystemD. Triple win.
Re: (Score:2)
Unfortunately, no.
These particular piles of "software," however awful they may obviously be to those familiar with software development, do make distributors' lives easier in the short term.
RHEL has already destroyed the good will it once had with the FOSS community, and is already well on its way toward becoming a niche distribution. But its demise will not prevent the need for us to invent something dramatically better than what either GNOME or systemd have become.
(Disclaimer: I use Gentoo, in which neit
Re: (Score:2)
There aren't many more companies with vested interest in Linux with cash flow to fill the gap in a open manner.
The money currently spent on redhate can go somewhere else, and then someone else will have the cash flow to do what redhate does now, except without the withholding sources. Fuck IBM.
Re: (Score:2)
Yes, go somewhere more trustworthy with a long reputation of magnanimous open source behavior, like Oracle, oh wait....
There's a good chance that the revenue ends up going somewhere even worse.
Also there's a good chance the revenue stays parked right where it is, because these companies *hate* change and generally don't care about the copyleft as long as they get theirs. Largely these are shops that see Linux as 'cheap, popular Unix'. They didn't move off of Unix systems because of ideology, they moved it
Re: (Score:2)
Oracle's involvement is going to scare a lot of folks, myself included, away from ever recommending this. Which sucks, because I have some level of respect for the other players. But Oracle. :(
They tried to copyright an API.
From now it continues to be Gentoo for myself personally, and Debian or Devuan if I have to recommend anything for others. Not Ubuntu because snaps among other things.
Re: (Score:2)
In this hypothetical, the revenue opportunity would go somewhere. Of course, whether that steward would be as good is a dubious proposition.
I think current Suse would be as good or better a steward than RH if they had the success. Their own distribution however hasn't had success and is generally not even mentioned in 'replace CentOS/RHEL' conversations, but maybe this OpenELA thing can reverse their fortunes.
Oracle, well, we've seen what happens when Oracle gets their way, this is probably a nightmare sc
Re: I don't know what will happen (Score:2)
SuSE is very ethnocentrically German in just about everything and their approach isnâ(TM)t much different than that of Red Hat. SuSE doesnâ(TM)t provide free access to their repos, including the source RPMs, and neither do a lot of other distros, including for example Proxmox and Ubuntu. All the code upstream is available, as it is with RedHat, but custom builds and custom patches are paid and donâ(TM)t allow redistribution.
The only reason SuSE exist is because EU agencies are required to
They Can't (Score:2)
My bet is that they will stop to release source code of distro modules under BSD, MIT, APACHE and MPL Licenses for RHEL and in certain Windows for CentOS Stream
Unless they want to rewrite these modules from scratch under a different license. I mean they can, but they open themselves to lawsuits.
Re: (Score:3)
>>"My bet is that they will stop to release source code of distro modules under BSD, MIT, APACHE and MPL Licenses"
>"Unless they want to rewrite these modules from scratch under a different license. I mean they can, but they open themselves to lawsuits."
They aren't GPL. Those licenses have no requirement to supply source code of the modified OR original code. They just have to keep the copyright/attribution intact. So RedHat/IBM can up the evil even more and totally withhold those if they want.
Al
Re: (Score:2)
Re: (Score:2)
That was the point of mentioning BSD, MIT, Apache, and MPL, those all allow RH to close off the source code just like that. The copyleft licenses (mainly the GPL family) they are stuck with, but they can readily close up all sorts of stuff. So the kernel, glibc, coreutils, and about half of their distribution is 'stuck' open source, but there's another half that they could close up and make impossible to create a viable clone. They have enough influence over a lot of the GPL projects, they might even be
Why make it RHEL compatible? (Score:2)
Re: (Score:3)
Someday that might be possible. For now RHEL is the enterprise standard for nearly all commercial and proprietary software packages, of which there are many out there that many enterprises rely on. Oracle Corporation produces hundreds. It is also the standard base for a great deal of internal software as well. The needs of both to have a long term stable base is why Redhat has a business. It will take a great deal of trust and experience for a new standard to supplant that, and the more incompatible thin
What is your bet? (Score:4, Interesting)
>"What is your bet? Let us know in the comments."
I already moved all machines off RHEL and CentOS and onto AlmaLinux. And I didn't wait for this latest round of evil, I did it over a year ago with the first round.
My bet is that a huge number of people/sites will do the same/similar. IBM/RedHat has destroyed their brand, good will, and standing in the community that provides and supports the code they "steal" (I say "steal" only because that is what THEY call what the RHEL clones were doing).
Re: What is your bet? (Score:1)
Why did you pick Alma and not another RHEL derivative such as Rocky? Curious because itâ(TM)s time to migrate our build machines from CentOS 7. We develop SDKs and have some customers on RHEL and derivatives, and for the rest, building on one of these seems to maximise our glibc and libstdc++ compatibility. Furthermore, the developer toolsets allow our developers to use modern compilers very easily. Some of the developers suggest moving to Ubuntu, but we donâ(TM)t really want to get in to back
Re: (Score:2)
Not him, but at least when this all started, Alma had their stuff together more quickly. They released first and even after Rocky released, Alma released first more often. Alma Linux had Secureboot down from the onset, Rocky had to figure it out after a couple of releases. So Rocky I'd best describe as being rusty in the beginning, and anyone paying very close attention would see Alma as the more together option.
Nowadays, Rocky has shaken it off, and has in fact climbed to be the most popular RHEL clone.
Re: (Score:2)
>"Why did you pick Alma and not another RHEL derivative such as Rocky?"
Because Rocky didn't exist at the time.
I would absolutely not move to Ubuntu. I would pick Debian over Ubuntu for servers and Mint over Ubuntu for desktops. But I do no regret using Alma for servers.
Re: (Score:2)
It all hits CentOS stream before RHEL (Score:3)
My guess is that in the next release, RHEL will stop offering official SRPMs even to their own customers of the non-GPL packages, instead referring them to CentOS Stream sources. So OpenELA will end up doing exactly what AlmaLinux does now, and pull from Stream after carefully evaluating the RHEL releases they wish to be compatible with.
I think AlmaLinux's method of using Stream is probably the only sustainable way forward and I wish OpenELA would have just joined them in this effort, rather than just trying to copy RHEL's packages verbatim.
Should RH go down the path of restricting non-GPL sources (outside of Stream), it reinforces in my mind the importance of the GPL to the entire Linux ecosystem.
Re: (Score:2)
Well, actually, it's more complicated.
Broadly speaking that's the story, but, for example, you'll frequently find that there is an RPM in RHEL that doesn't quite match the version of any rpm that had ever been in stream. One example, I once looked at a fully updated RHEL system and it's kernel was newer than the newest stream had to offer. Usually Stream is newer and the RHEL is some intermediate backport, but sometimes RHEL just flat out releases updates to RHEL without touching Stream first.
RHEL has mad
Oracle? Too bad... (Score:4, Insightful)
I'd feel a LOT better about this development if Oracle, of all companies, wasn't nominally one of the 'good guys'. I don't trust them as far as I could throw One (hundred) Rich Asshole(s) all at once.
Re: (Score:2)
That's what's good about FoSS. If they do it wrong then the very same work can be picked up and run with. That's what's so offensive about IBM's attempt to control distribution of the Redhate sources, they want to interfere with one of the very most important things about FoSS and GPL software in particular. What they're doing may be legal, but it goes directly against the spirit of the license that made Linux great in the first place. When even Oracle becomes the good guy by comparison, you know you have g
Re: (Score:2)
Nobody wants (Score:2)
Nobody wants to be locked in to a vendor. By removing alternatives that's exactly what red hat have done. Red hat will continue to shrink, Debian based alternatives will continue to grow.
Re: (Score:2)
Stop saying "drop" instead of "release" (Score:1)
Re: (Score:2)
I'd agree.
Ambiguity is rarely even acceptable in a language, any language, and it certainly is never desirable. I'd never encourage or recommend adding to it.
It does happen sometimes all by itself; for instance, the meanings of both "mean" and "let" have almost completely reversed since early modern English.
But please, oh please, don't purposely make it worse!!!