> Well, Redhat has to make money somehow, so this decision is understandable.

RedHat has been making money for decades. The Centos rug pull was just a way to make more money at the expense of their users and the larger open source community.

It is impossible to justify a rug pull. By definition, it is a hostile, one-sided, trust-destroying, and underhanded move.

Apparently not an equivalent amount of money to the value they feel they’ve provided. And honestly given the communities response, I think they’re probably right. I mean if you think about it there’s no reason someone using Centos for a proper use case wouldn’t be fine moving to Centos Stream, unless they were getting some value out of Centos production-friendly releases. Vendor compatibility and LTS patches and updates take a big team to maintain, someone has to pay for that and I feel like many were just using Centos when really they should be using a supported enterprise distro.

I found it notable that the linked RHEL post had zero mention of the why of it. Seems like a glaring omission.

Can you define a rug pull?

Promising to support an OS for 10 years then cutting that support to just 2 out of the blue?

The question I’m trying to answer is who promised what, where was that published, and when was it published. If a squirrel promised me 10 years of support, that’s different than a CentOS maintainer making that same promise on the wiki.

The 10 years of support on CentOS 8 was published on centos.org's wiki:

https://web.archive.org/web/20201101131417/https://wiki.cent...

Red Hat's explanation was something along the lines that it was a wiki and not guaranteed to be accurate? I don't think editing was open to to people outside of the CentOS org.

I could completely see how one could make decisions based on that date, but I can also see how doing that could backfire. As always, a plan b is valuable.

Frankly, I think this kind of thinking is short-sighted quarterly corporate bullshit almost certainly done by IBM and Redhat itself would never have done it (and most of the people who were Redhat in those days left to work somewhere else). How many corporate environments the world over are running on Redhat, paid Redhat, because the engineers there came up in the Linux world learning free Redhat and were able to continue using CentOS at home or in their dev environments? That sort of cradle-to-grave brand identity with people who eventually become high-level decision-makers and money controllers is invaluable, but the value is also impossible to quantify, accountants never se it, and instead just see people trying to use all of their hard work for free.

Yet somehow even Pablo Escobar understood this and threw away billions just giving money to poor people to keep cities on his side. I'm not even going to say IBM's basic business model of making shit that works, doing at least some complex things reliably and well back when that was very difficult, is invalid. But it is fundamentally different to Redhat's business model, which was simply cultivating a lifelong love of their product in the minds of its users. It's like the Army acquiring the Red Cross. They work in a lot of the same places doing overlapping work and no doubt a lot of corporate synergy could be achieved and administrative overhead eliminated, but it would irreparably destroy a century and a half of goodwill and brand identity. Exactly the kind of strategy you'd expect from people whose decision consequence horizon extends no further than the next three reporting periods.

"Acquire competitor, just to shut down competitor" is a pretty gross (and monopolistic) business practice. So that's a relatively crass way of looking at it.

>"Acquire competitor, just to shut down competitor"

There was 6 years in between those two events, and by all accounts CentOS got much better post-acquisition.

There is no "just" here. Originally it was considered a good idea, and (years) later they changed their minds.

"Not acquire competitor, and have to shut down your own firm" is a more crass decision.

Perhaps if you need to acquire and shutdown your competitors, you should shut down your own firm, or at least pursue a different strategy. Monopolistic business strategies are crass.

Leave out the "cr" and I agree

RH did great financially for many years while CentOS was around; do you have some reason to claim that it was an existential threat?

because having CentOS lead to Oracle Linux being a thing which allowed big competitor to offer what was essentially 100% compatible product to your own while undercutting your support costs.

I don't think RHEL/IBM cares about the small losses on CentOS/Rocky/Alma/etc what they found was hurting them was companies like Oracle

I don't think that's correct; AFAIK OEL has always pulled directly from RHEL sources. That is, they were a sibling, not a downstream of CentOS. I am open to the idea that CentOS was basically collateral damage as Red Hat attempted to block Oracle, but that's not the same thing.

That's what I'm suggesting - the only way RHEL could block OEL is to also kill off CentOS as we knew it.

Agree and wish there was something that could be done against such business practices, but it is indeed difficult to regulate.

Over the years I became aware that the acquire-extinguish pattern was definitely not just a coincidence, specifically i remember disney bought up some 'IP' just to litigate against some adjacently-related online fan game in the UK in order to shut it down and 'acquire' its captives, over to a similar product they already owned. Sorry I looked for the exact reference but cant find. There are countless other stories if you look though.

This all came full circle when I was 'enlightening' a young relative who had just finished business school, or rather she 'learned' me. Turns out this acquire/extinguish game is like a whole 100-level course in <pick-your-business-school-of-america>. Who knew.