On 24.12.2020 09:16, Mike McGrath wrote:
On Wed, Dec 23, 2020 at 7:50 PM Konstantin Boyandin via CentOS-devel <centos-devel@centos.org mailto:centos-devel@centos.org> wrote: > Many people on this list are looking to find a way forward here -
to
> understand what is going on and provide suggestions where they see it > might help. It is forgivable that you and others would jump to > conclusions that equate to IBM, or greed, or revenue because you don't > have all the information we do. But just know that when you again try > to boil this all down to just a revenue discussion, you're missing a lot > of detail that actually went into this. Oh, thank you so much, I didn't hope it could be forgivable. Now that we can actually be absolved, it's so much simpler to breathe. Translating your last statements into English: you (those outside of RH) have not that information, you won't have that information, thus
you
may not come to any conclusions. Again, this is extremely simple. The fundamental flaw of CentOS Linux was it didn't bring profits directly. On the contrary, the majority of sysadmins were using CentOS instead of RHEL, since CentOS was stable enough, was quite simple to maintain - and unless any specific requirements were rpesent, didn't require paying expensive bills from RH. That simple.It didn't have to bring in profits, but it did have to be mutually beneficial. In the beginning it was, but back then OpenStack was the future of computing. Things change. Take a look at Fedora - it doesn't
bring in profits. But it is extremely useful to us and to the community. As Matthew can tell you, it takes a lot of hard work to run a community like that and actively balance the needs of Red Hat and the community. We run several communities that don't bring in profits. Red
Hat also belongs to several foundations and other open-source groups, some of which we don't have any productization plans but we do them because we think they're good for the industry.
"Mutually beneficial" - for RH and whom, exactly?
CentOS Board, with all due respect, definitely doesn't speak for the entire community.
And really all I'm getting at here is that there is more to Red Hat than profits, we've proven that over and over. If this decision has caused you to lose that trust, to make you think that Red Hat cares only about profits, that's your right.
I am not your customer (although I worked closely with many of your customers, so I have impression about your tech. support and the rest), so obviously that proof evades me.
Since the exact reasoning of CentOS dismantling won't be a matter of public knowledge, I have no reasons to change the viewpoint. It was/is all about profit.
Of course I appreciate RH caring about open source (no irony here), but again, for any corporation profit is the first priority. All the rest is secondary.
The missed lot of details doesn't change the above simple assumption. CentOS Linux was too good an alternative for majority of systems, thus that "Carthago delenda est". That simple. No need to permeate an aura of mystery. What's really funny is the constant flow of BMWs (bovine metabolic wastes) from RH, trying to "prove" that CentOS Stream is a drop-on replacement for CentOS Linux (which it isn't) and that it fits 95% of CentOS Linux use cases (which it doesn't).CentOS Stream isn't a drop-in replacement for CentOS Linux. There's not some crazy messaging to make people think they're the same. But they're also not entirely different and are far more compatible than what reporters and many were writing about on the day of the announcement (which is why we've worked so hard to correct that).
Of course they are not entirely different - they both have "CentOS" string in them.
I am glad you avoided mentioning that statement that "CentOS Stream would cover 95% of CentOS Linux use cases" (in my words). For clarification, can *you*, explicitly, either agree to that statement or disagree with it?
Let me give you one benefit of Stream that didn't exist in CentOS Linux that I *CANNOT* believe hasn't come up yet. Are you aware that there were at two months in 2020 where CentOS Linux 8 received no security updates of any kind? When the next y stream comes out, the team goes to building that y stream and stop working on the current version. People touting security and stability either weren't even aware of this, or have just decided not to discuss it. I can't imagine Stream would have that problem.
That kind of sh*t happened now and then all the time. As a sysadmin, I got used to be constantly wary and bring emergency measures in similar cases; sometimes, in a matter of hours. So, to me, it's a regular event.
That difference makes sense for those who will actually be using CentOS Stream. People on this list provided detailed explanation why CentOS Stream can't fit many typical CentOS Linux use cases.
For generic use - for generic systems, such as VPSes - CentOS Stream can be very useful, when nothing really mission critical depends on it (when a system can be safely rebuilt, when another Stream update breaks it). I suppose it can be very useful in cases similar to Fedora - for development purposes.
But no, make no mistake. Chris's blog was correct. CentOS Stream and CentOS Linux are different things in terms of deliverables. While at the same time RHEL and CentOS Stream are extremely closely related and the nuance between them will take a while for people to understand.
Yes. And CentOS Linux users, meanwhile, just migrate to other distributions (Rocky, Lenix, Springdale, whatever), lifting a burden of unnecessary expenses off RH.
Apart from RH's reputation losses (I suppose RH just doesn't care about that), there remains not much to grieve upon.