How is it licensed, Jigsaw? Eh? What distro is it from? Is that a fucking Snap wheel?
Snap
Ok, this set me off.
What I hadn’t anticipated in my 20 years away from Linux was not only had teams of unpaid volunteers been beavering away behind the scenes to make everything work better, other much more enthusiastic teams have been thinking up new and exciting ways to break it again.
I’d just like to interject for a moment. What you’re refering to as wheel, is in fact, GNU/Wheel, or as I’ve recently taken to calling it, GNU plus wheel.
The Wheel weaves as The Wheel wills.
Use the
-w
flag and the wheel will weave as you will.Wtf did not expect a Wheel of Time reference lmao
Reinventing the wheel leads to a profound understanding of why wheels are round.
That’s what documentation is for.
Documentation is written exclusively for people with PhDs.
Well, it would have been if people updated it when making changes; now it’s just all an incorrect snapshot of an older version of wheel that no longer reflects reality.
That involves knowing how to read
Not necessarily.
“I WOULDN’T BE REINVENTING IT IF THEY DIDN’T FORCE
systemdAXLES ON EVERY WHEEL!!!”The wheel has had a number of innovations over the years. The earliest wheels were flat disks of wood that were heavy and slow turning. The Romans invented spokes and metal rims which made them faster, more durable, and gave them more traction. Questions we need answered: What is this wheel in particular designed to do? Is there any way we could make it work more efficiently at its task? Do we value performance over reliability, or vice versa? Etc. Etc.
I think we need to take a bit of a step back and consider what kind of shed we might use to store this wheel…
To answer that, we’ll need to do a deep dive into foundation technology (to determine if it is lacking and needs some improvements) (because we don’t want our wheelshed to sink).
I don’t know, that’s not really in my wheel house
What is this wheel in particular designed to do? Is there any way we could make it work more efficiently at its task? Do we value performance over reliability, or vice versa?
It works fine. It’s a perfectly good wheel.
Hey where is Underwaterbob?
He’s trapped in that Jigsaw room.
The door is unlocked though?
Yeah, but there is a wheel in there and UWB won’t leave until he figures out if there is a way to improve it.
Has any one asked him to?
No
Will he get paid to improve it?
No
What does the wheel do?
You roll it out of the way so you can exit the room.
Will he get paid to improve it?
No
Well, now I’m clearly going to have to find a way to monetize the wheel as well.
Put a lock on the wheel and charge people $0.99 to temporarily unlock it.
Evil. I like it! Maybe some mandatory ad viewing somehow shoehorned into the unlocking process as well.
Put a dynamo on it and sell the electricity
Use the electricity to power a screen and speakers and sell ad space!
Please stop giving BMW ideas.
Sounds like proprietary blobs.
Yeah but neither the romans, or medieval peasants had computers, so why are you on about?
Now we do have computers! Think of the models of wheels that could help us improve wheels!
Spent months setting up my home server with Docker containers while learning Linux. Everything worked perfectly fine.
Then I realised Ubuntu Server is just a Debian-flavored landfill. Switched to EndeavourOS. Everything worked perfectly fine.
Then I made NixOS my daily driver and thought, “Hey, let’s ruin my weekend.” Migrated the server. Everything worked perfectly fine.
Found out I could run containers as systemd services. Replaced Docker out of sheer spite using compose2nix. Everything worked perfectly fine.
Then I heard btrfs was the bee’s knees. Reformatted my drives, migrated again, and spent a week learning why subvolumes are better than sex. Everything worked perfectly fine.
Got a free MacBook. Slight hardware bump. Migrated again. Spent hours fighting T2 drivers while deepthroating Tim Apple’s cock. Everything worked perfectly fine.
Rewrote every systemd service as NixOS modules. Why? Something something George Mallory. Everything still works perfectly fine.
Did I ever notice a difference from the frontend? Nope.
Was this a good use of my time? Fuck no.
Did it need to happen? Does the pope compile from source in the woods?
I mean it sounds like you just enjoy spending your time doing that sort of thing. I’d say that was a good use of your time if you wanted to do it, no?
Im at the compose2nix phase of this pipeline. Ive got a bunch or sevices in Docker compose files and all of my systems have been running Nix for over a year now. Ive gotten the hang of my repo and made a couple modules for my specific uses and im hooked.
What would you suggest to migrate all my compose files into a nix friendly environment? I use flakes as well.
NixOS Chad
sudo systemctl stop sawtrapd
And don’t forget
systemctl disable sawtrapd
so it won’t restart again.sysyemctl disable --now sawtrapd to do both in one command.
This is a poorly designed horror trap. Here, let me help you!
but is it written in 6510 assembly, with cool graphics and catchy music with fast arpeggios?
“Or as I’ve recently taken to calling it, saw plus trap”
We’d rather re-create reality where we know everything rather than taking the time to learn how to use a system someone else wrote.
IT and DevOPS does this too.
I worked with a group once that re-invented XML so that non-technical people could create text-based rules instead of writing code. But it ended up with a somewhat rigid naming structure with control characters and delimiters. The non technical people hated it more the actual XML they had used prior.
You’re talking about YAML? /s
LOL. not far off
They started out with something close to YAML. As the project moved forward, they found out they needed to represent logic with interlinked sections. They needed section 3, point a to link back to section 1 point 3, sub point 2. So they toyed with some assembly-like operations. Then they needed some inheritance. They really just slowly re-implemented the common applications of xml one at a time, it just had less brackets and <> symbols when they were done.
it just had less brackets and <> symbols when they were done.
Hence making the parser more inefficient than XML?
It wasn’t without some advantage. The client hating it didn’t bode well though
The client hating it just means you’re smarter than them and should press on to help them outgrow their ignorance. It’s a good sign.
YAML definitely felt less intimidating to me than XML, when I first saw them.
But the YAML examples also had much less information in them than the XML ones.
But not having to type all those brackets definitely helps. In case of XML, I am always looking to just get a GUI going for it instead, because typing it out feels cumbersome (I’m from C++)
Re: the not-XML-instead-of-code thing. Eventually, this sort of thing turns into a programming language. It’s just like carcinisation. Or you wind up writing ever-more code to support the original design. The environment inevitably creates evolutionary pressure that only if/else and iteration logic can solve, forcing the design ever closer to being Turing-complete.
I woulda tried them on JSON. As long as they use an editor that keeps track of nested brackets I think it’s much more natural than XML.
I switched to TOML for my stuff.
Wow, I never even heard of TOML. Very interesting - thanks!
edit: after looking at it a bit I think I’ll actually try using it. But I find it ironic that the website for something billed as “for humans” and “easy to read” is done in light gray text on a white background. The CSS class they chose is even called “light gray” LOL.
One of the worst parts about this is that I would never have thought about reinventing it until he told me not to.
Bloody reverse psychology still working on me. >:(
I’ll just steal the wheel and reinvent it later
gasp! You wouldn’t download a wheel!
Tech bro strat.
Do you work for Apple?
Better make sure the wheel isn’t under copyright tho!
Removed by mod
Is the wheel FOSS? No? Guess I have to then.