THE layoffs at Microsoft are real and they go deep. Microsoft has recently laid off about 5,000 people (including people on contract) and based on reports from media near Red Hat, a similar number of people is to be discarded (laid off) by IBM. But since then (earlier this summer) not much has been said. Like with Microsoft, many of these things happen gradually and quietly. We took note of it several times earlier this year, but now we're looking at actual messages from Red Hat insiders (or former insiders). We'll provide some evidence as we go along, accompanied by our interpretation of the present and future of Red Hat. We welcome Red Hat insiders who can provide us with further input.
First, as a little bit of background, earlier this year we researched for long periods of time to better understand whether IBM's planned (and openly announced) layoffs in NC area affect Red Hat (formerly RHAT and then RHT, now IBM). We looked for some rumours online, but came up with too little; almost empty-handed, but that was months ago. Remember that Red Hat has its own operations in NC (the headquarters and founding place); corporate media said IBM planned to lay off in NC and 4 other states, the total being -- reportedly -- about 5,000 people (IBM is still hiring in India by the way). The tricky thing is, IBM and Red Hat both have NC-based operations and a rather large number of workers there. It's somewhat of a business hub. But we also know that IBM does not need two HR departments, two marketing departments, etc. Managers are sort of converging in duties, conflicting in terms of roles, overlapping in the workflow sense and so on.
Last year, as we noted here before, I had heard from an IBM acquisition victim (whom I cannot name, but he is a high-profile person) that they always wait 2 years before the guillotine falls. Why 2 years? Go figure. But if one studies the pattern (after IBM acquisitions), then it's always 2 years. It has now been almost exactly two years since the acquisition was announced (a couple of months from now).
Has IBM begun axing staff of Red Hat? Well, nobody has explored or covered that subject (which we know of...) and it is unlikely that IBM or Red Hat will just spoon-feed this kind of information. It needs to be 'pulled', as they won't 'push' out such information. Citing a recent press report, Ryan saw signs of impact for Red Hat. "They said it would "make it difficult to hire more people" with the pandemic raging," he quoted. "No comment when asked about whether that meant layoffs from Red Hat."
So deeper we go into anonymous posts like this one. It doesn't take long to 'get' what's going on. To quote a couple of messages: "If it's like the red layoffs from last fall, they are given a month severance and their access to anything internal is immediately cut off. All the while, management will claim "these people were given the chance to find new positions". They were? No they weren't. That's like being figuratively walked to the curb and being told you can't talk to anyone inside anymore. Funniest thing is that even after being gone 6 months, there are still working links that many have and some logins that IT forgot to close. I guess that's the obstacle when you let engineers build their own doors for a project."
Another one says: "12+ people recently made redundant by Red Hat Management. As new VP at EMEA level took over the charge, his first action as to known to IBM management was to lay off people and ceased the department where the money isn't showering as IBM/RH would have expected. Irrespective of How much time and talent those 12 people have put forward to do the branding of the new team. So, it has started here at RH as well and will continue to grow, I mean lay-offs ;) because IBM wants to use their 60% non-productive resources to be consumed in a brand like RH. As IBMers says, A good engineer is replaceable in 3 months But a chicky [sic] manager, hard to find. :)"
"Culture has definitely changed," said another person, "and I can agree that the new management from especially the most recent company (big blue) has changed things. There is definite drinking of the koolaid that is necessary and if you dare to state the obvious, you'll be quickly labeled and dismissed as no longer needed. As for "kingdoms", there is only one now and containers is it. Work even on the OS itself is pushed down to the bottom of priority. Long time managers are scared and running. This is actually good new though for other companies who need good people."
This one says a lot about Jim Whitehurst, who recently became President at IBM: "I was at RH for more than a decade, from the Matthew through Jim, and I saw the changes coming when Jim came on board. He never really seemed to grasp was the core value was that Red Hat offered. He didn't use our products (used a Mac, along with other members of the executive management team), allowed RH to dump open source solutions for our own business to move things to Google's services (which is a huge message to our very clients of "we don't value or trust open source, so why should you?") and the top-down view shifted away from what made Red Hat special and valuable and more towards what would make us value to another company who would want to buy us.
"IOW, something that would benefit the CEO, EMTs and shareholders.
"But would ultimately cause the company itself to crash and burn.
"The buyout by IBM, the culture change, and now the layoffs were all things I had expected. They came a year or two later than I expected. I'm sorry to see the company circling the drain now as I absolutely LOVED my time there, the people I worked with and what the company was at the time."
"Not only has Red Hat been getting lay offs from IBM," Ryan noted, "but they're sloppy about how they do it and forget to cut off employee access to things apparently. After security proverbially walks them to the door. Red Hat sounds like it's falling apart from the inside and their CEO doesn't even understand what the product is, or use it himself. Other comments suggest that IBM is gutting Red Hat of anything it doesn't expect to immediately turn a huge profit. Firing engineers without seeing if they could even be tasked elsewhere."
MinceR then joked that "they don't care about security, they just want to ruin yours and mine..."
"IBM is trying to stuff its own nonsense that isn't making money into Red Hat products," Ryan continued. "I guess the logic is that if they can shove it into a Red Hat product people are buying, they can say it's "value added" and justify it."
Containers hype is mentioned there along or between the lines. In many cases a container is just a binary blob, usually with proprietary core inside (no code), laced with an 'OS' somewhere 'around it' (not a very good design).
"They said that OS development was knocked down to lowest priority," Ryan noted. It says it right there. That may explain quite a lot, not just about RHEL's direction but also Fedora's. We see more vendor tie-in/lock-in, more software patents (monopoly) and not much of real value.
IBM's new CEO loves containers; we recently decided to see his videos, including fairly recent interviews where he talks about "Watson everywhere" (proprietary) and "cloud everywhere" (yes, those are the slogans).
"Too many bad things are happening in Fedora at once," Ryan said. He recently used and participated in it. He saw things he disliked, including censorship of dissenting voices. "I'm not about to stick around and wait for "Silverblue" [Big Blue]," he said, alluding to what some might label "vapourware". Planet Fedora has been mostly dead lately (I've followed it closely for years).
"This whole thing is getting comically bad," Ryan said. "It's no longer modern to install an RPM and just have the program there a few seconds later without REBOOTING. So that tells you it's going to be one of those things that just breaks the entire world, and their answer to what about software is "Either reboot every time you install an RPM or just use Flatpaks!".
Flatpak used to depend strictly on systemd (they apparently fixed that, at least temporarily). "Fedora seems to be increasingly out of the loop," Ryan concluded. "I think at some point they might just stop pretending that they even care about testers. The push for BtrFS came from Facebook of all places. That guy from Facebook that doesn't use an @facebook email because he doesn't want to make it obvious. Looks like management they brought in from other companies is forcing reorgs on Red Hat. Sounds like morale in general among employees is low."
If Red Hat isn't part of GAFAM now, it's certainly part of what figosdev calls "GIAFAM". And judging by lack of commitment from IBM (e.g. to Fedora), it doesn't look too encouraging, at least not for Red Hat. Thankfully we still have Debian and other large distros. There's far more than one point of failure when it comes to GNU/Linux.
Going back to commitment issues, Ryan believes "that a Red Hat Enterprise Linux with no Fedora is a matter of perhaps a couple of years off. They're far more interested in what other big companies want in RHEL, and you could just as easily spin up RHEL 9 Technical Preview releases and see if that works out for them. IBM has managed plenty of operating systems that didn't get any outside input at all and doesn't seem to really care about Fedora. It's just a development structure that it inherited. The BtrFS discussion had no talk from Red Hat. Some people kept saying "Let's talk to Red Hat and see why they dropped it from RHEL 8.". Nobody from Red Hat ever offered any input, at least in public, on the BtrFS feature. The file system situation in Fedora and Red Hat has already been quite different with RHEL defaulting to XFS and Fedora defaulting to Ext4 for a long time now. Red Hat seems to have no interest in actively developing Ext4 or a potential successor. They support it in the sense that you could install to it with non-default options, but they strongly discourage that, especially with large volume sizes. Linux distributions are by no means on the same page anymore regarding what a sane default should be for the file system or what, if anything, should replace Ext4."
One sure thing is, almost all distros sooner or later move to adopt systemd, sometimes because there's no other choice (too many dependencies upon it). ⬆