Red Hat's Decision Prompts Outrage and Sympathy, Called 'Necessary' and 'Embarrassing' (siliconangle.com) 118
SiliconANGLE reports that Red Hat's decision to limit access to RHEL sources "has sparked outrage in some circles," but observers contacted by the publication "were mostly sympathetic" to Red Hat's position:
Most acknowledged that the company's explanation that it couldn't keep funding the development of software that competitors then gave away for free was reasonable. But not Bill Ottman, founder and chief executive officer of Minds Inc., a social network built on open-source code." They are completely embarrassing themselves by betraying the community and their own model," he said. "Their best bet is to immediately reverse course and apologize."
Others were more inclined to agree with Josh Amishav, founder and CEO of data breach monitoring firm Breachsense. "If we want commercial entities to support our underlying operating system, they need to find ways to be profitable," he said. "If you disagree with Red Hat's policy change, then there are plenty of excellent Linux alternatives to choose from."
Some saw the move as a consequence of pressure inside IBM to justify the $34 billion it paid to buy Red Hat nearly five years ago. "Red Hat has to change to protect its business," said Joe Brockmeier, head of community at open-source developer Percona LLC and a former Red Hat employee. "They seem to have tried to find the least harmful way to do that. It's a necessary decision, although one that could have been communicated a little better." Brockmeier agreed with Red Hat's argument that it can't continue to fund innovations and give them away for free. "Copying a company's product isn't what open source is about," he said. "The code is what allows every company and individual to run, study, modify and distribute work based on a project. The members of the community can do those things; what they are finding harder to do is to 'clone' RHEL."
Not everyone buys the argument that IBM needed to wring more revenue out of its subsidiary. "Considering IBM's gross profit for [fiscal 2022] was $32.863 billion, this certainly wasn't a make-or-break decision for IBM's profitability," said Kadan Stadelmann, chief technology officer at Komodo, developer of a cryptocurrency and blockchain platform. And there's some risk to Red Hat in closing down source code access. "By totally removing free and open-source software, Red Hat may not necessarily increase revenues that much while alienating its large community of open-source developers," Stadelmann said.
There's evidence that's already happening, at least for now. Red Hat's action has both energized and elevated the profiles of some open-source alternatives.
Others were more inclined to agree with Josh Amishav, founder and CEO of data breach monitoring firm Breachsense. "If we want commercial entities to support our underlying operating system, they need to find ways to be profitable," he said. "If you disagree with Red Hat's policy change, then there are plenty of excellent Linux alternatives to choose from."
Some saw the move as a consequence of pressure inside IBM to justify the $34 billion it paid to buy Red Hat nearly five years ago. "Red Hat has to change to protect its business," said Joe Brockmeier, head of community at open-source developer Percona LLC and a former Red Hat employee. "They seem to have tried to find the least harmful way to do that. It's a necessary decision, although one that could have been communicated a little better." Brockmeier agreed with Red Hat's argument that it can't continue to fund innovations and give them away for free. "Copying a company's product isn't what open source is about," he said. "The code is what allows every company and individual to run, study, modify and distribute work based on a project. The members of the community can do those things; what they are finding harder to do is to 'clone' RHEL."
Not everyone buys the argument that IBM needed to wring more revenue out of its subsidiary. "Considering IBM's gross profit for [fiscal 2022] was $32.863 billion, this certainly wasn't a make-or-break decision for IBM's profitability," said Kadan Stadelmann, chief technology officer at Komodo, developer of a cryptocurrency and blockchain platform. And there's some risk to Red Hat in closing down source code access. "By totally removing free and open-source software, Red Hat may not necessarily increase revenues that much while alienating its large community of open-source developers," Stadelmann said.
There's evidence that's already happening, at least for now. Red Hat's action has both energized and elevated the profiles of some open-source alternatives.
Bussiness model (Score:4, Insightful)
a consequence of pressure inside IBM to justify the $34 billion it paid to buy Red Hat
I've seen it once I've seen it a hundred times. You buy a company, successful for doing things in a particular way, but you cannot stop yourself from changing that way.
After all, it's not like IBM has a scarcity of closed source products to offer, or anything.
Re:Bussiness model (Score:5, Informative)
During my tenure at IBM, I got to see it happen over and over.
IBM decides they want to play in a market
They figure out they can't organically participate, so they acquire.
After acquisition "we are going to leave it alone, we paid for the value of the company the way it has been running"
Shortly thereafter "We think we can now make that business work better, so we are going to take a stronger hand in directing it"
Thus far the next step has been "for some reason, the business unit surprisingly tanked, so we are shutting it down or selling it to get rid of that worthless business unit.
Of course, for the last 20 years, RH has consistently shown this mindset already floating around, so IBM may have given it a little shove, but it was already in the ballpark of this thinking.
Re:Bussiness model (Score:4, Insightful)
This doesn't look like an IBM move, though. Getting screwed by IBM is usually death by a thousand cuts. Whatever their sins, abruptness isn't typically one of them.
OTOH, RH has been routinely springing unpleasant surprises on their customers since at least the turn of the century. This move is so typically Red Hat I'd bet they'd have still pulled it even if IBM wasn't in the picture.
Re: (Score:3)
This doesn't look like an IBM move, though. Getting screwed by IBM is usually death by a thousand cuts. Whatever their sins, abruptness isn't typically one of them.
OTOH, RH has been routinely springing unpleasant surprises on their customers since at least the turn of the century. This move is so typically Red Hat I'd bet they'd have still pulled it even if IBM wasn't in the picture.
They've been really good to their customers, but they're always had a loath / hate relationship with the clones. Though it typically took the form of alternating between making stripping out trademarks and rebuilding the distro difficult and trying to divert potential customers away from free RHEL clones to something that provides QA for future RHEL releases.
This move strikes me as IBM inspired since traditional RH was smarter about doing things that went this directly against FLOSS.
Re: (Score:2)
They've been really good to their customers, but they're always had a loath / hate relationship with the clones.
Those two things are the same thing. Most of the people running the clones are their customers. They just don't want to pay for every box, because they have lots of machines they don't need Redhat to be involved with. And this is the basic problem with Redhat as it is being used by IBM now — the bulk of the software doesn't belong to them, and almost all of the pieces they do hold copyright to (maybe all of it?) is FOSS.
What's weird about this to me is that it would have seemed like Redhat would have
Re: (Score:2)
They've been really good to their customers, but they're always had a loath / hate relationship with the clones.
Those two things are the same thing. Most of the people running the clones are their customers. They just don't want to pay for every box, because they have lots of machines they don't need Redhat to be involved with.
I don't think it's quite right that they don't need Red Hat. If a customer has 19 clone boxes and 1 supported Red Hat box all the issues will happen to the Red Hat box.
That's probably why they've gone to the free to 16 or whatever model. So that potential customers start out using RHEL instead of clones, and when they cross the threshold it's easier to keep using RHEL than to switch to a clone mixture.
And this is the basic problem with Redhat as it is being used by IBM now — the bulk of the software doesn't belong to them, and almost all of the pieces they do hold copyright to (maybe all of it?) is FOSS.
What's weird about this to me is that it would have seemed like Redhat would have been a good fit for an IBM which seems to be ever more focused on providing services. In that model it's a benefit to not have to do all of the software development yourself. IBM is basically trying to take control of software under licenses explicitly designed to prevent it, though to varying degrees of quality.
If IBM succeeds at locking this software up behind their licensing agreements, then we're going to need still another version of the GPL which explicitly prohibits this behavior.
Possibly, when the current revenue stream is dominated by subscription fees it might be hard to make the sw
Re: (Score:2)
I don't think it's quite right that they don't need Red Hat. If a customer has 19 clone boxes and 1 supported Red Hat box all the issues will happen to the Red Hat box.
No doubt there's some of that, but there's also a lot of uses where you just don't ask for support. If you just need a crapload of machines to spread work out across, if you have a problem with one of them you don't actually try to fix it. You just reimage it and go on with your day. If you have a ton of desktops and one user's desktop goes on the fritz, same story. If you've got your act together you only have one or two different kinds of PC to deal with at a time, a few special cases aside.
I think the biggest threat to Red Hat is the fact that Ubuntu has almost completely supplanted Fedora among home users and hobbyists and stuff like this won't help.
Ubuntu is less
IBM had nothing to do with it (Score:4, Insightful)
Multiple people in Red Hat have gone on record as saying at no time did anyone from IBM talk to anyone at Red Hat about the decision to stop pushing debranded SRPM information to git. Remember the decision to push the community away from RHEL clones to CentOS Stream was made a long time before IBM was part of the picture. Yes I'm sure there is pressure to perform, but this is not some IBM plot like recent Oracle's bizarre comments claimed.
That said, if RH's business is getting a bit lean, forcing out the clones will certainly not help with that at all. It's unlikely that any Alma Linux user would ever be a paid RH subscriber, and even with Rocky's push to monetize RH's work for themselves, that likely wouldn't eat into RH's business either.
Re: (Score:2)
Multiple people in Red Hat have gone on record as saying at no time did anyone from IBM talk to anyone at Red Hat about the decision to stop pushing debranded SRPM information to git. Remember the decision to push the community away from RHEL clones to CentOS Stream was made a long time before IBM was part of the picture. Yes I'm sure there is pressure to perform, but this is not some IBM plot like recent Oracle's bizarre comments claimed.
That said, if RH's business is getting a bit lean, forcing out the clones will certainly not help with that at all. It's unlikely that any Alma Linux user would ever be a paid RH subscriber, and even with Rocky's push to monetize RH's work for themselves, that likely wouldn't eat into RH's business either.
Problem is, alot of people are / were familiar with RHEL just cos of CentOS. So if they got to a point that they wanted paid linux support, it was always better / easier to get into RHEL, compared to SUSE or something else, cos of experience with CentOS.
Now that's not going to be the case anymore. So if someone wants to get paid linux, they may actually something other then RHEL, since they no longer have background experience with it.
Re: IBM had nothing to do with it (Score:1)
Re: IBM had nothing to do with it (Score:2)
Re: (Score:2)
Tivoli's greatest asset was their customer base, though IBM did an annual gross of literally a billion bucks selling TME10 licenses a year or two after the acquisition. But the additional sales strained the technical support structure, so they had to bring in a level 1 support team, and they hired a bunch of idiots to work for peanuts. Prior to that, if you made a support call to Tivoli, you went straight to someone who had been a sysadmin previously who answered and immediately disposed your call. Shock am
Re: (Score:2)
After all, like Oracle, IBM is where software goes to die.
Re: Bussiness model (Score:1)
Based on the timing of this,I think it has more to do with Amazon Linux ripping off red hat and not really contributing anything back than it has to do with IBM. It is not really good for open source if Amazon clones red hat and then puts Red Hat out of business.
What does the Linux Foundation have to say ? (Score:1)
Re: What does the Linux Foundation have to say ? (Score:1)
Wait... (Score:1, Insightful)
doesn't RedHat get software for free and then sell it to others?
Re: (Score:3, Insightful)
Yes, though for anything important, they tend to want to shun it if at all possible unless they 'own' the development.
For a long time, RH's commercial line about distributed storage was "Gluster is good, you may hear about this ceph thing, but it's no good"
Now that they have Ceph under their control "Ceph is great, Gluster is dumb".
When docker popularized container approach, but it happened outside of RH control, they ignored it as long as possible until finally they made a clone (podman) rather than ever f
Re: Wait... (Score:2)
Re: (Score:1)
I'd say it's a bit disingenuous to say that Red Hat just repackage software developed by others. Red Hat QA that software and pass any code fixes back upstream. Red Hat also develop a few projects around the Linux ecosystem and are one of the main contributors to the Linux kernel.
And they're not selling the software but support of said software.
Re:Wait... (Score:5, Informative)
Well, actually, they are also selling the software, point blank.
See https://www.redhat.com/en/stor... [redhat.com]
The nominal price per year per server is $349 with *ZERO* support.
It is fair to say that for most of the key software they fund the development of that software though.
Re: (Score:2)
The nominal price per year per server is $349 with *ZERO* support.
Does anyone actually pay that?
Redhat has a free tier, with up to 16 systems. Above that, one can probably negotiate a better rate.
Re: (Score:2)
Does anyone actually pay that?
Yes, quite a few do, actually. I've worked at several. It stacks up pretty quickly, especially in clustered and redundant setups.
Redhat has a free tier, with up to 16 systems. Above that, one can probably negotiate a better rate.
They have a "free" tier now, they didn't used to. They added it about the time they locked out downstream. Any system built in this free tier must be registered and attached to RH's CDN, and communicate with it or you will not get updates. It breaks yum/dnf.
Oh, also it's only for Developers. If you're a company, organization, non-profit, or anything else, you have to talk to incre
Re: (Score:1)
Re: (Score:3)
And they're not selling the software but support of said software.
That's what they *should* be doing. But make single use support easy, and stop insisting that people wire up their VCenters for license management. I think the openSuSE Leap / SLES model is pretty good, as is SuSE in general-- you can get an opensuse box up and running for free, and if you want support, you can easily do an in-place "update" that turns it into official, licensed, SLES.
Unfortunately, openSuSE (and SLES) somewhere acquired a bad reputation-- I ran SuSE from about 1998 to around 2016, and wi
Re: Wait... (Score:2)
Yes, but Red Hat contributes a lot to the community. Then Amazon clones it, does not contribute, and sells it as Amazon linux while competing with Red Hat in cloud.
Re: Wait... (Score:1)
They become less of a no-brainer (Score:2)
I think this is the likely fallout of this move, whether you have sympathy or not: Their competitors can now offer a new, significant thing that they canâ(TM)t, so Red Hat has taken on the burden the proof of value. They say their programming teamâ(TM)s contributions will still make them the best choice, but the judges will be customers and time.
Re: (Score:2)
The problem with Red Hat is they'd rather you ran openshift than docker / k8. They'd rather you run JBoss than tomcat. They're trying to push everyone from the open software that runs the internet, to their product that most people at best tolerate.
They're trying to redefine the linux standard to be the Red Hat standard, and I don't think that's a good idea, to put it mildly.
Re: They become less of a no-brainer (Score:1)
Re: (Score:2)
True. But they're also commercial offerings from Red Hat. And that's what they want you to use.
GPL Loophole (Score:2)
Re:GPL Loophole (Score:4, Informative)
Though it's not so much as a loophole, as deliberate design in the GPL.
I have always been surprised at how the GPL specifies that, ultimately, the recipient of your binaries must also be able to get your source, however, explicitly, you are given leeway to make source access a bit more obnoxious than the binary.
A scenario expressly spelled out is that you give your end-user the ability to click to download your binary, but require the end-user to submit requests for source code through snail mail, and charge them a reasonable fee to mail them DVDs of the source code. Making source a more limited PITA is expressly granted.
Re: (Score:3)
There's no such scenario spelled out in GPL 1 and 2, although they seem to indeed allow requiring a snail mail request -- but only because there was no better universally available method at the time. Ie, it's not an "express grant" but the lack of a crystal ball.
GPL 3 makes it more clear: obtaining the source must be same or less effort than the binary.
Re: (Score:2, Interesting)
Re: (Score:2)
Yes, this restriction on further distribution of the source code is explicitly against the spirit of the GPL and despite the fact that wise people have said otherwise I'm pretty sure it's against the letter. Someone with part title to the Linux kernel should sue them with the aim of getting them banned from distributing it forever as is provided under the GPLv2 (for once, I think Linus was wise not to upgrade to the v3 which has an automatic forgiveness clause). This is one of the cases where I wouldn't pro
Re: (Score:2)
It's not a loophole. When Richard Stallman started to make the code for GNU, he had to find a way to sustain himself. He was asking people who wanted to have access to the code to pay him money to get access. What I don't believe is acceptable is that IBM demands that anyone who has paid to access the code be restricted from giving it to others freely.
My thoughts as well. The loophole is being able to contractually prevent redistribution of code you received under the GPL.
Re: (Score:3)
That's not exactly the "loophole", what Red Hat is claiming is that if you pay to get the binaries, they will give you access to the sources as per the GPL, and you can, as per the GPL redistribute those sources, but if you do they will no longer give you access to future binaries (or sources). So they basically say you can exorcise your rights under the GPL but if you do they will cut you off.
Personally I think this is a violation of the GPL itself as they are punishing you for exorcising your GPL rights.
Re: (Score:2)
That's not exactly the "loophole", what Red Hat is claiming is that if you pay to get the binaries, they will give you access to the sources as per the GPL, and you can, as per the GPL redistribute those sources, but if you do they will no longer give you access to future binaries (or sources). So they basically say you can exorcise your rights under the GPL but if you do they will cut you off.
Personally I think this is a violation of the GPL itself as they are punishing you for exorcising your GPL rights. Unfortunately as end users we don't have standing to sue, you have to be a copyright holder of one or more of the GPLed software that is included in RHEL. I think the most likely candidate for this is the Free Software Foundation, who own the copyright on all of the GNU software and utilities including bash, GNU coreutils, glibc, gcc, emacs ... the list goes on and on.
Maybe loophole isn't the right term, but I agree there actions is against the spirit and intent of the GPL, even if it is permissible under the terms of the GPL. It would be interesting to see what contractual claims RH/IBM could make beyond cutting off access if someone redistributes the source despite the agreement. Could they sue for damages since redistribution may deprive them of revenue? Could FSF win a copyright claim since RH/IBM complies with the GPL but conditions future access based on a separa
Re: (Score:2)
I would not be surprised if there is a version 3.1 of the GPL that addresses this problem and all the FSF software gets relicensed scuppering RedHat's move for at least all the FSF software.
What I can't see is how RedHat are going to stop say Oracle setting up a limited company somewhere buying a RHEL subscription and redistributing the SRPM's for at least everything under GPL. If a significant number of people with subscriptions (I am game) start downloading all the SRPM's then RedHat won't be able to tell
Re: (Score:3)
> I would not be surprised if there is a version 3.1 of
> the GPL that addresses this problem and all the FSF
> software gets relicensed scuppering RedHat's move
> for at least all the FSF software.
That wouldn't do much of anything versus IBM. Plenty of GNU tools were forked when RMS went all "Thou shall not make a profit off *MY* software." with GPLv3. Some of that was by companies that had added GNU tools to their own products and didn't care for the choice between "stop making money" or "re-to
Re: (Score:2)
I don't think Stallman has ever had an issue with people making money off of Free software, after all he explicitly said as much either in the license itself or in the GNU Manifesto. The sole purpose of his life's work is to make Free software and it's source available to people. Tivo implemented a loophole where they distributed the source code they modified (as required by the GPL2), but restricted their hardware from running any modified software. This is what the GPL3 sought to change, it really has not
This is concerning (Score:5, Insightful)
. "Considering IBM's gross profit for [fiscal 2022] was $32.863 billion, this certainly wasn't a make-or-break decision for IBM's profitability," said Kadan Stadelmann, chief technology officer at Komodo, developer of a cryptocurrency and blockchain platform.
CTO of a crypto company, but doesn't seem to understand the difference between gross profit (~$33billion) [macrotrends.net] and net income ($1.8billion) [macrotrends.net]....
I guess this is a clue explaining what's wrong with the crypto industry...
Re: (Score:2)
Oh no, they only net $1.8 billion, clearly they are unprofitable.
Yes, it is a bit disingenuous to quote gross profit (here presumably for the enhanced dramatic effect), but they still have a pretty big net income.
If RH is 'unprofitable' by any accounting it's IBM's fault. All the financials ever shown directly or cited by IBM have clearly shown relatively healthy net profit. If there's a boat anchor tanking that in any balance sheet, it would be IBM's accounting of the 34 Billion they pissed away acquirin
Re: (Score:2)
Oh no, they only net $1.8 billion, clearly they are unprofitable.
Yes, it is a bit disingenuous to quote gross profit (here presumably for the enhanced dramatic effect), but they still have a pretty big net income.
You're just as short sighted as the cryptobro. $1.8bn sounds like a lot of money until you realise they invested $34bn in buying Red Hat. That is a fucking horrendous ROI.
Re: (Score:2)
Risking blowing up the profit you do get is not a wise move. I'm skeptical this will lead to appreciably more revenue, but it will absolutely erode mind-share and relevance to the packaged experience. Trying to get money from clone users is like trying to get blood for a stone. The last thing they want to do is make it even *less* likely that RH offerings align with most of the packaged ecosystem.
If the profit over time means too long to get the desired ROI, then they made an unwise purchase for too much
All I know is (Score:2)
IBM's move prompted me to start examining what would be required to support and manage Ubuntu Server within our existing "RHEL"-centric ansible setup*. I considered SUSE, but I have to think about the next person doing my job - and it's a lot more likely they'll walk in knowing Ubuntu.
It seems to mostly just come down to the different package names and file paths; although I haven't yet looked at any of our more complicated server setups. But I'm thinking the path differences at least can be handled with so
Re: (Score:2)
but they'd have a much harder time walling that off
Oh how you underestimate IBM if it decides to maximize profit of a project, or kill it in the attempt.
Looking at the active contributors to the project since 2022, nearly all the activity has been RH employees exclusively. Which speaks to RH's point about being the financial backbone of so many projects, but also speaks to how much they have the ability to control open source projects.
Re: (Score:2)
There is truth in that. But it's also true that ansible is rather mature / feature complete from the point of view of the average user. Plus it mostly depends on external stuff (e.g. openssh, python) for security. So having to fork it at some point from today onward would probably not impact that many users of the free version.
Re: (Score:2)
IBM's move prompted me to start examining what would be required to support and manage Ubuntu Server within our existing "RHEL"-centric ansible setup*. I considered SUSE, but I have to think about the next person doing my job - and it's a lot more likely they'll walk in knowing Ubuntu.
Well, maybe. SUSE is RPM-based, and there's a lot of historical cruft that's going to be very similar to RH/Fedora/Rawhide. I think they're actually the ones who are going to become winners here. They already have a profitable business model and a community that likes them, and they have much to gain from additional community growth.
Re: (Score:2)
Those are good points. And the situation seems to be rapidly evolving - the broader Linux community is still processing the fallout and how to address it. Additionally, I saw that AlmaLinux teased some new partnership announcements coming in the next few weeks - it's not a given I'll walk away from it.
But in any case I'm still just exploring and testing - and the stuff I figure out vis a vis supporting Ubuntu will be useful with any distro... if in the end I decide to make a move. Or, for that matter, with
Re: (Score:2)
IBM's move prompted me to start examining what would be required to support and manage Ubuntu Server within our existing "RHEL"-centric ansible setup*. I considered SUSE, but I have to think about the next person doing my job - and it's a lot more likely they'll walk in knowing Ubuntu.
It would probably be smarter to do Debian. Anyone who knows Ubuntu can figure out Debian, which has excellent and straightforward documentation, and more importantly doesn't use snap. It has systemd so the same installation and configuration instructions will work. But Ubuntu seems to be going towards using snap as the basis for big pieces of the OS, and snap sucks — both on its lack of technical merit, and its closed server nature.
It seems to mostly just come down to the different package names and file paths; although I haven't yet looked at any of our more complicated server setups.
And obviously down to using apt instead of rpm. rpm actually has one a
I use Debian everywhere. (Score:2)
Re: I use Debian everywhere. (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Somebody doesn't get open source (Score:2)
"Copying a company's product isn't what open source is about,"
The first line of the Wikipedia article [wikipedia.org] says:
"Open-source software (OSS) is computer software that is released under a license in which the copyright holder grants users the rights to use, study, change, and distribute the software and its source code to anyone and for any purpose."
OK, so Red Hat doesn't want to do that anymore. They want to release "their product" under different terms. That is, they want to stop releasing open source software. That's fine as long as "their product" isn't a derivative work
Origins of Free Software (Score:5, Interesting)
Perhaps it is time to look at what originally caused the Free Software movement.
Richard Stallman worked at MIT Artificial Intelligence Laboratory which bought a new laser printer. This printer was shared across the laboratory, and so sometimes queues could be long. With the previous printer, Richard Stallman had modified the printer firmware to send a mail to the print job owner when the job is finished, and also to send messages when the printer jammed. However, the manufacturer of the new printer kept the firmware proprietary, so he could not make the change. Even if he had been able to do so, he would have been prohibited by copyright from sharing the improved firmware.
If the printer had been running Red Hat Enterprise Linux, Richard Stallman would have been able to get the source for printer firmware (but he would have needed to get the support login). He would also have been able to change the software, and he would have been able to share the fix.
https://amogh.medium.com/the-s... [medium.com]
Re: (Score:3, Informative)
He would also have been able to change the software, and he would have been able to share the fix.
Except for Red Hat's standing threat to cut off support to those who exercise their rights to distribute source code.
Re: (Score:2)
I cannot find that threat in their terms anywhere. It seems unlikely that Red Hat would stop the subscription in the scenario I outlined. There would be no reason to.
Re: (Score:3)
It's been openly discussed on mailing lists. In fact, that's the point that RockyLinux devs are trying to get around -- finding a way to obtain SRPMS without having to agree to the RedHat.com user license, which specifies that their access can be withdrawn for any/no reason.
Re: (Score:2)
I do not have a direct source on that threat. Smarter people than me have gone over it. It is my understanding that all source code will be customer only and not to be shared.
So in this example of the printer, they would be in violation of GPLv2 from my understanding.
An article quote:
Benny Vasquez, the AlmaLinux Foundation's board chair, fears that AlmaLniux must follow Red Hat's licensing and agreements in addition to following the software source code licenses. That would also mean, as AlmaLinux understan
Re: (Score:2)
("They" - being the person that shared any source code under the terms. Richard Stalman in that example.)
They could share the patch though, from my understanding. In the printer example.
But I'm not a lawyer. Nor do I play one on TV.
Re: (Score:2)
Minix was also around at the time (or a bit after). When you bought the Minix book, your got the binaries and all the source code. But you were not allowed to share that source.
What people did was share context diffs. You'd download them and patch your copies of the source. I went through patching v1.3 to 1.5 to upgrade my 286. The only thing you didn't have source for was the compiler so you could not upgrade it.
Back then, most people did not have CDs on their computers. You had floppies and a 2400 ba
Re: Origins of Free Software (Score:1)
Well, for this use case, he would only need to distribute the patch file and other RHEL users could apply it to their own copy of the source and recompile. No need to redistribute the original source, itself. In fact, even for upstream users, a patch file might be sufficient.
I don't agree with what Redhat are doing, but this example does highlight why, in fact, out might not be totally against the spirit of FOSS (at least, in terms of its origin story).
Re: (Score:3, Informative)
It's not against the spirit of Open Source software, or at least, not as a blanket term. That means a lot of things. But it's definitely against the spirit of the GPL. The reason why we have a GPL version 3 today is that it was believed that the GPL needed an anti-Tivoization clause in order to fulfill its mission of making sure that users can actually make, share, and use changes.
Re: (Score:2)
If the printer had been running Red Hat Enterprise Linux, Richard Stallman would have been able to get the source for printer firmware (but he would have needed to get the support login). He would also have been able to change the software, and he would have been able to share the fix
yes, BUT in this analogy RH would have then cut off their support contract with Stallman's lab/university (per the new TOS), which means no more support, no more valid login to the support server that contains the firmware+sources, no more updated firmware in the future, no future printer purchases, possibly no more toner/drums/parts for that particular printer, etc. etc.
"You are totally free to request+distribute the source to this binary that we just gave you as required of us by the GPL, but the contract
Re: (Score:2)
Hopefully a court eventually sees through this BS.
I'd rather Red Hat come to their senses rather than be forced by the court to do the right thing.
After 10 Plus Years Of Using CentOS (Score:2)
As soon as IBM bought Redhat and then started changing crap with CentOS I moved to OpenSUSE and have been quite happy with it.
Now if we can get ARM up to snuff, we can kick IBM out of most everything. Including their own IBM Compatible world.
Re: (Score:2)
"IBM Compatible World?" That hasn't existed for decades now. IBM doesn't even care about PCs; they sold their PC division to Lenovo years ago. IBM may play in all the games but they certainly aren't setting any standards. Windows on AMD and Intel define the PC platform.
ARM will never replace Wintel. For one, the ARM consortium doesn't care at all about a standard platform. They don't care about fragmentation of implementations. They only care about getting their license fees. The licensees only car
RHEL is getting forked out of the mainstream (Score:2)
I'm looking forward to seeing what SuSE is doing with their planned fork of RHEL into a new distribution.
If all of the other major IT players (Like Oracle and Amazon) back it, we could end up with a new Enterprise standard distribution that doesn't require forking license fees over to IBM. I think many businesses are getting sick of cutting IBM a big check annually, and it wouldn't take much to get people to switch.
okay redhat (Score:3)
Re: (Score:3)
what about all the source code redhat benefits from that they have not written themselves, i bet there is more coming into redhat than goes out of redhat,
I don't like this move but that's not really the right comparison.
Think of it this way, there's three kinds of distros, the main distros like RHEL and Debian, there's a few others [wikipedia.org] but those are the big two.
There's a reason why there's so few main groups who build their distros from the ground up, it's an incredibly difficult and time consuming tasks. You need to figure out all the different packages needed, find a compatible group of libraries, get everything built and configured, and then keep watching for
Re: (Score:2)
You're both right. But who's more right? Because this is literally what this argument is about, so it actually matters (even if you are here just proxies for the two camps.)
On the one hand you have Debian, which doesn't charge anybody for anything, and which is apt-based. On the other hand, you have Redhat which does charge people for things, and which is rpm-based. And almost everything else (as measured by installed base) is based on one of these things. And it's notable that the most popular Debian deriv
Re: okay redhat (Score:1)
Um (Score:3)
Most acknowledged that the company's explanation that it couldn't keep funding the development of software that competitors then gave away for free was reasonable.
Um .. sure, it's reasonable.
But in that case, you make your own operating system, from scratch, or from proprietary sources you own the rights to - you don't use one that has a license that says you have to make the source available.
Re: (Score:2)
Most acknowledged that the company's explanation that it couldn't keep funding the development of software that competitors then gave away for free was reasonable.
Um .. sure, it's reasonable.
But in that case, you make your own operating system, from scratch, or from proprietary sources you own the rights to - you don't use one that has a license that says you have to make the source available.
As bad as this move by Red Hat is I think it's important to remember two things:
1) Red Hat is still one of the largest, possibly the largest, contributors to Open Source.
2) Among distros, they're one of the best for trying to push their patches upstream to make them available for everyone.
3) All of the code is still GPL'd, so even stuff they didn't push upstream it's trivial for competitors to take.
To play the devil's advocate, what I suspect their thinking is that the source code and the OS are different e
Re: Um (Score:1)
RHEL sucks (Score:1)
These are the same people that randomly deprecate packages that are critical to many applications, break backwards compatibility between _minor_ releases, and retire certifications without a good reason.
I don't know why anyone (except of course for the incompetent people that constantly need someone else to do their job) would use RHEL.
Just wait (Score:4, Funny)
Their transition to the dark side will be complete when they change the color of the hat...
Re: (Score:2)
IBM'ers haven't worn suits for decades now. Maybe they should call it "Blue IBM logo polo with Khakis Linux" instead :)
The playbook: (Score:2)
Embrace, expand, extinguish. They are now ready to start their proprietary, closed-source "extend" phase.
MAANGs and Fortune 500's exploit CentOS (Score:2)
Re: (Score:2)
I'm curious as to which distribution Facebook will switch to. At their size and scale, they could easily roll their own RHEL-compatible distro.
If they used and supported to SuSE RHEL fork, that might also help to spur adoption for that distro.
Re: (Score:2)
and 99% of a Red Hat distribution is open source code that Red Hat and IBM are profiting from.
They need to quit their whining.
I had always advocated for Redhat... (Score:1)
Red Hat can't have it both ways (Score:2)
RHEL and RedHat are Dead (Score:1)
The company I work for has pretty much banned all new RHEL / RedHat software and is requiring us to migrate away from using any RedHat software.
In addition, RedHat as truly wrecked CEPT.
Kick the bums out. (Score:1)
Re: (Score:3)
If he wanted to do that he probably would have also wanted to promote GPL3 instead of advocating sticking with GPL2.
By stating he wanted to stick with GPL2, he was saying that he wanted to let companies pull shenanigans
A reminder that Red Hat/IBM isn't just RHEL... (Score:2)
IBM/Red Hat doesn't just make/support RHEL. Ansible, Ansible Tower (and AWX project), Cockpit... That is just off the top of my head.
So many of us have helped contribute to bug fixes from using products like CentOS. But I guess now we are just seen as pirates.
I really hope Ansbile doesn't go down this same road in the future. I know that is a different example, but it wasn't so different from Red Hat Linux previously.
IBM, common... Removing the source from public a GIT repo was one thing. Now also only cust
oracle oracle oracle (Score:2)
this is about oracle nothing else nothing more...
prevent oracle shipping and supporting customers without it actually doing some work = win
do hard work for oracle = fail
bullcrap (Score:2)
IBM knew what Red Hat did and the position it held in the industry before they decided to buy it. They intentionally bought Red Hat with a plan to change the way it does business and to turn it into a cash cow.
IBM is evil
As a Slackware user.... (Score:2)
...I'm just sitting back, munching on pop-corn, watching the spectacle.
Their spoken logic belies their motive. (Score:2)
Redhat has been profitable for years. Their success is what prompted the large purchase price by IBM. This is not even in question.
IBM likely told Redhat execs that they needed to increase profitability -- without laying out a plan. Redhat likely decided to go after the low hanging fruit by cutting off anyone that was costing them the sale of a support contract, such as Oracle. The smaller entities of Rocky and Alma are simpky collateral damage.
Redhat had been disingenuous with their response when they s
Are SRPM spec files considered derivative source? (Score:1)
Re: (Score:2)
Overplayed and Overreacted (Score:1)
With all the amount of whining around this issue, it seems to be really overplayed.
RedHat has not limited access to source RPMs or SPEC files. They are still going to be available in Centos git repos, with a revision history for each package.
What is going away is two things:
a) access to source RPMs for a particular RHEL release or update level (I believe it has been already gone since RHEL 9)
b) the list of package versions that a particular RHEL update or RHEL release includes.
That is, all source revisions
Joe didn't say IBM was pulling the strings (Score:1)