Right?! Like if youre going for historical use the context! So much historical plots make no sense if you assume they had social safety nets, clean water, workers rights, access to global markets, concept of universal human rights, on demand light and heating, largely managed forests, and don't get me started on language barriers!
I don't like that this was removed, but it is very ironic for a AI summary about AI hype fueling a RAM shortage. It was at least properaly labeled and attributes and an open source model.
And it's a balance too that self interest and ego gets alot done too. It's just getting over protective of ego or too self interested (very hard in an economy where a lot employers are straight up conmen) that leads to these pain points.
I do find in the rooms I've had the pleasure of being with the smartest people in a field were always full of reasonably humble people.
How much of your income do you want to give to buy alcohol for strangers? Would you donate a large amount of your money to an aid fund that spent 10%? 50%? 80%? on booze? What about meth? Guns? Nazi memorabilia? What it's only 5% on Nazi stuff, 95% on food?
I'm being a dick but they have a fair point in why people put caveats on aid. I'm a fan of UBI to some degree personally, because I think people as a rule should be trusted with making their own decisions, but I do like choosing where the value of labor goes too.
If you don't understand that development, security, and operations are all one job you will constantly make crap and probably point at some other team to make excuses about it, but it will be actually be your fault.
Programs have to run. They have to be able to change to meet needs. Implementing working security measures is one of those needs.
The amount of times I've had to slap devs hands that wanted to just disable security or remind security that just shutting it down is denial of service is crazy. If it can't deploy or is constantly down or uses stupid amount of resources it's also worthless no matter what it looked like for split second you ran on on the dev machine.
The next patch isn't going to fucking fix it if no one that writes patches knows about the damn issue. Work arounds are hidden technical debt and you have to assume that they will fucking break on some update later. If you are not updating because it breaks your unreported workarounds you will get ignored by the devs at some point, and they are right in doing so.
If you depend on something communicate with the team that works on it.
We can send a fucking petabyte of info around the world and to the moon and back before some people write a fucking Ticket, email, or even a IM. Look dumb and asking the stupid question rather than being an actual idiot and leaving something broken for the next decade. We're all dumb, it's why we built computers, get over it and just talk to people. If you really struggle with, don't just communicate, try to over communicate, say an obvious thing now and again just to keep the dialogue open and test that you really on the same page.
That's my rant/hill borne from ulcers supporting crappy IT orgs and having to overcome my own shortcomings to actually say something in channels where things can actually change and not just griping in private about it.
To be fair. SELINUX always seems like THE answer with flexibility it provides with App armor being just SELINUX light...
It would make more sense to me to have better support for leveraging SELINUX primatives to accomplish the same things. I at least, don't know of any LSM features that can't be covered user:role:type:security level:catagorey and namespaces?
The issue is always that info is hard to know sometimes and programers can barely stop ourselves from running as root with all files in 777 mode let alone conceptualize those other attributes for files and services
Bazzite and Kinote though I use distrobox and k8s alot for messing with other distros/apps. Vscodium and neovim. Vscodium is loaded up with nearly anything IaC or kubernetes related and Continue for some AI stuff (pointed to local and mistrial). Also heavy opinionated stuff for Python like black, etc (I want my ide to yell at me to make better code). Some GitHub and git lab add-ons too. Nvim is just as is.
Definitely overkill lol. But I like it. Haven't found a more complete solutions that doesn't feel like a comp sci dissertation yet.
The goal is pretty simple. Make as much as possible, helm values, k8s manifests, tofu, ansible, cloud init as possible and in that order of preference because as you go up the stack you get more state management for "free". Stick that in git and test and deploy from that source as much as possible. Everything else is just about getting to there as fast as possible, and keeping the 3-2-1 rule alive and well for it all (3 backups, 2 different media, 1 off-site).
Fleet from Rancher to deploy everything to k8s. Baremetal management with Tinkerbell and Metal3 to management my OS deployments to baremetal in k8s. Harvester is the OS/K8S platform and all of its configs can be delivered in install or as cloudinit k8s objects. Ansible for the switches (as KubeOVN gets better in Harvester default separate hardware might be removed), I'm not brave enough for cross planning that yet.
For backups I use velero, and shoot that into the cloud encrypted plus some nodes that I leave offline most of the time except to do backups and updating them. I user hauler manifests and a kube cronjob to grab images, helm charts, rpms, and ISO to local store. I use SOPS to store the secrets I need to boot strap in git. OpenTofu for application configs that are painful in helm. Ansible for everything else.
For total rebuilds I take all of that config and load it into a cloudinit script that I stick on a Rocky or sles iso that, assuming the network is up enough to configure, rebuilds from scratch, then I have a manual step to restore lost data.
That covers everything infra but physical layout in a git repo. Just got a pikvm v4 on order along with a pikvm switch, so hopefully I can get more of the junk on Metal3 for proper power control too and less IPXE shenanigans.
Next steps for me are CI/CD pipelines for deploying a mock version of the lab into Harvester as VMs, running integrations tests, and if it passes merge the staged branch into prod. I do that manually a little already but would really like to automate it. One I do that I start running Renovate to grab the latest stable for stuff for me.
As THE USB-C PD evangelist. I have to say. Fair.
Like PD EPD is definitely reaching the limits of the USB-C form factor to me, and data over copper is a dead end at some point too.
Still want ever device I have on it. Though as we scale past the 260 watt range (and I do...) or longer distances (also me) it's just going to have to be another interface and probably medium for data for the protocol. So far MPO for data and honestly pogo pins for power are the best I'm seeing.
Again for everything thats not a serious power device like well pumps, servers, AC/Heat pump, Power tools, etc or serious data server/client. Its fine, which is seriously impressive.
Rant over I also like the idea of better hardware stats reported to the OS. Its one reason I fell in love with software raid over hardware raid
Git lab CI is my goto for git repo based things (unit tests, integration tests, etc). Fleet through Rancher for real deployments (manages and maintains state because kubernetes). Tekton is my in between catchall.
Tbh the amount of cameras and microphones we have that upload to external unaudited servers ("the cloud") is insane to me. This is just further worry about it since the scheme allowed their ownership over our devices and privacy, which means they can also sell that.
This is how I use kubernetes (specifically Harvester HCI and some lighter RKE2 nodes). Just one big computer lots of nodes. Still working getting the plumbing fully figured out for virtual desktop to output video devices.
Honestly. I think private companies can have a place in public infra, but it's not in the freaking ownership. Rent seeking is the worst and most destructive aspect of private ownership and we've known and can look at countless example of that since Adam Smith!
It was open source, federated and written in Rust!
The Reddit API enshittification event was just what kept me here.