Ah there is liquidity too, but your brother comment makes my point clearer.
About liquidity, yes most people with a million net worth actually have more than half in their house, so technically it is much harder for them to throw cash than somebody with a billion and a much smaller % of their worth in illiquid assets such as property or unlisted companies. I wish I had made this point too.
A few things to note. 1 billion isn't a thousand times a million. If you make a conservative 5% let's say out of your net worth, you still need to work with a million, whereas you don't with a billion. So, technically, $400 with a million is some amount of work hours, whereas $400k with a billion is just pocket change taken out of more than most people lifetime's of earnings that is just 1 year of your interest.
Also, a lot more people (more than 1000x) have $400 to give than $400k so in a sense if people with $400 to give were all being very generous, they could amount to a lot more than what billionnaires could give.
I give a counter example to a general statement, in logic, it is enough to debunk the original statement. My point is, SOME things wouldn't run predictably inside the cluster (in that case without pinning to nodes, which the article says isn't necessary, and which in general defeats the purpose), so you'd need to run some things outside of it.
As per standardized knowledge, I can't see how somebody even proficient with kube, could jump into any app and troubleshoot bad behavior. Apps each have their quirks and subtleties, specific components that behave a certain way. The layers still exists, the kube cluster itself (which again has many component options at every layer of the stack ; hard to know them all), and the app (which will require at least some specialty knowledge).
If it's just about pushing helm charts we wouldn't need SRE anymore, just a CI.
Uniformity ? Try deploying openbao inside kube, if kube decides to restart your pods, you're in for unsealing them at 3am, waking up everybody who owns a Shamir key. So bao stays out of the cluster, or pinned to certain nodes, defeating the purpose entirely. Also, with the ultra wide variety of tools at every layer of the stack, uniformity is a joke ; there are no 2 kube cluster deployment that are the same really.
Standardized knowledge ? The operating system is standardized knowledge. Any competent SRE should be able to login into a Linux box and figure out what's running there. And if you let your previous ops shadow it all you're just a pretty bad CTO.
Tracing who does what ? First of all anybody with admin access can run one time jobs just like anybody with sudo can run one time commands. That's like chapter 01 of the kube doc. Also again at the kube layer itself, below the helm chart, the ops who set that up or updates it can and will change stuff that breaks stuff.
Kube isn't necessarily bad and has it's purpose but it's not a product. It's like Linux, a complex piece of tech that requires a lot more knowledge than "just push this helm chart" to work.
I'd be happy to try some more, I used to play this seriously (even studying with solvers) and I also know and trust a couple of pros, will share with them as well.
Alpine is (was ?) a Renault brand, which is a French company, so it is a little less exotic than a Japanese police force buying a German car (Japan being such a massive car exporter themselves).
Tried daily puzzles on pokerchallenges.com ; 2nd one is "jam or fold" 5bb deep, from the big blind. But no information on what other players do, and we are last to act on this spot, so the question doesn't make sense.
Edit : another one didn't make sense, it asked for my equity with A5o on KK7 or something, but I have no information on opponent's hand or range or betting pattern. Apparently it assumed villain had any2, but pretty much nobody has any 2 here.
I work with junior ops "engineers". Even though some of them won't admit it, they want to do everything with AI, and they do not care much about testing. Even their peer review is done by AI, you can tell by the way the comments are written. They want to deploy stateful clusters, manage VM lifecycles, and even deploy our own Kube cluster, probably entirely vibe coded.
One of them tried to use tabs in yaml instead of spaces, didn't know what a virtualenv was, and didn't know why Kube jobs asked for xp in writing operators ; yet he is pushing multi-files features in less than a day that are meant to manage our backups, and is pushing for managing our own kube clusters in front of management (which finds that it is a lovely idea). At this point I am not sure what to do but it won't end very well I feel.
I functionned this way for a very long time, plain vim with 5-6 options in .vimrc, no plugin (aside for the very occasionnal syntax highlighting like jinja.vim iirc). I kept this setup for more than 10 years mainly to stay compatible with multiple systems (in my case Solaris/Illumos, Linux, MacOS and even Windows).
But I made the switch to nvim / LazyVim. And it is actually pretty good. I had in mind those endless hours of config and lua scripting. At the end of the day all I needed actually was to remove a plugin (folke, which messes with my 's' key) and learn to use the package manager to setup the languages I wanted.
Having things like GrugFAR or lazygit at the tip of my finger is actually a quality of life improvement. I could do without those for sure but they fit my workflow and muscle memory well.
Still wish there was something better for ansible ; I should have gone with pyinfra with my current job's project but I only learned about it after writing 12k LoCs of ansible :'(
No way I see the difference between 120 and 144, but it's just annoying to have to set it up that way, it's like getting a new motherboard and setting your cpu to 3.7Ghz instead of 3.8 or something. It's one more thing to figure out (because my screen would just go dark and it annoyed me for a couple days), setup, and remember next time.
For the desk I'm actually happy I got a reasonnably small one for my new place, the big one I had was just always filled with junk, now it forces me to tidy it up and throw away things. Also it's in my living room, not a dedicated room, so saving space feels better.
I got a DeLock KVM switch recommended by the shop vendor (who actually did spend a good 15 min understanding my need and doing web searches), and quickly confirmed by ChatGPT as a solution to my need (plug the work macbook alongside my personal linux tower, no exchange of data between hosts, no driver needed, and do a KVM switch).
I actually end up using 2 mice and switching the keyboard from bluetooth (for the mac) to usb (for the linux) because I can't get usb working through the HDMI. And the screen is limited to 120hz (instead of 144hz) because of the switch quality or the HDMI cables, I do not know (it lights up and works for a minute and then goes black a few seconds every few seconds, losing sync).
In the end it is more tedious and less featureful than just plugging the mac to the screen via a USB to HDMI cable, switching everything from the screen (and doing the same mice / keyboard juggle than above), because at least I would get the 144hz and save the 150€ from the switch. I can't return it anymore because the 15 day window has passed so I still use it, it's a little bit easier to switch from the remote control than from the screen anyway, but that's expensive just to switch from a different button.
Incus is pretty damn good to be fair. You can mix and match VMs and containers, the terraform provider "just works", the setup is fast and easy, it plays well with ZFS. Now I wouldn't be surprised if it still lags jails (or Illumos Zones) in robustness or some capabilities but I'm a happy user of them now.
In my new position (on a different product) I don't have enough fingers to count how many times the previous guy bullshitted the PO/PM with "that's not possible" of having some features / workflows enabled. Just because he didn't bother thinking through it or just didn't want to do it. Most of the stuff is a bit boring but just a few days of work and test. So yeah I entirely agree with you.