how does this compare to WillAbides/setup-go-faster?

https://github.com/WillAbides/setup-go-faster

They target different parts of actions/setup-go for optimization:

- WillAbides/setup-go-faster speeds up the Go toolchain setup (literally installing Go)

- cloudx-io/setup-go uses the slower actions/setup-go toolchain setup, but changes cache strategy so your `go test` and `go build` steps do less work

Those strategies are compatible. I hadn't heard of setup-go-faster — thanks for putting me on to it.

If you're deciding between one or the other, it'll probably come down to which inefficiency predominates in your codebase (i.e. how many tests you have, how quickly they run, and how much real churn there is in your test package build graph).

which rockets have you worked on? are you in the aerospace industry?

Your takes are hotter than a launch pad with all 33 raptors firing!

and costs taxpayers billions compared to our working models 60 years ago. Great innovation

I remember the Saturn V being caught by chopsticks in mid air too.

Starship is fantastic value for money compared to Artemis!

Saturn V didn't take 8 launches and 9 orbital fueling operations to get to the moon.

>> Stainless steel is just such a terrible choice for a rocket this large.

Honest question, but what do they usually use?

Aluminum alloy and carbon fiber, because both have high strength to weight ratios. Steel is denser, so its specific strength isn't as high. But it retains strength at both lower & higher temperatures, and it's much cheaper. It's also cheaper/quicker to modify a steel design, as welding is easier than aluminum (at least for low carbon steels used in Starship). And if the same structure were made of carbon fiber, it would likely have to be fabricated from scratch.

I think choosing steel has helped Starship, as it allowed the ship to survive reentry and land in the ocean despite heat shield failures. Had the ship been made of aluminum, I'm certain earlier missions would have broken up when the aluminum heated up and lost its strength.

Does this mean you think Starship HLS will fail? It will use a steel structure and is planned to be part of future Artemis missions.

Also are you willing to stake money on that claim? How about this: If before July 25th, 2036, any version of Starship made primarily of steel (HLS, future Mars versions, etc) has an apogee above 2,000km, you pay me $1,000. If not, I pay you $1,000. If you like, we can bet some other amount up to $10k. And if you prefer, the loser could donate the amount to a non-profit/charity of the winner's choosing.

> We'd all be using WinXP SE and IE6.

the reason we have linux is because a bunch of geeks took a stand. i understand that you can't fight every battle, but that doesn't mean you shouldn't fight any battles.

you're making a rationalization because taking a stand is usually inconvenient, and you don't want to be inconvenienced by anything.

Supposing you know someone's motivation as if you have access to their psyche is a very unproductive way to participate in a discussion.

The reason we have a Linux is because a bunch of geeks took a stand on an issue they cared about and _focused on_. We wouldn't have Linux if the project also took on matters like teenage pregnancy, surveillance, and gerrymandering.

That's a very good point. What if codeberg decides tomorrow that they got weary of all worldly affairs, and from then on will only host dev tools?

Yet those same Linux geeks have now decided not to take a "stand" against AI, but rather embrace it. Twenty years ago when the Afghanistan and Iraqi wars raged on, did they oppose working with the military contractors at Red Hat or the NSA? No, they focused on the mission.

at least it's still got a bunch of Kiwi engineers building the Rutherford engine.

pack it up. we're going to codeberg.

they were cooked the minute they chose to write it in typescript

i'm sure it is, but it's written in Rust so it should be a little harder to pwn

absolutely no one is smart or productive enough to justify $692m in pay. they could hire thousands of engineers for that money.

ya i could hire 500 min wage baseball players to replace ohtani then.

Isn't that the plot of Moneyball? You'll almost never find a single player with 30 stolen bases and 30 HR both, but you can find two players with 30 stolen bases and 30 HR alone for cheaper.

sure, and it doesn't work when everyone knows how to value stats that contribute to wins. Back then, they didn't do so equally. you could go bargain hunting and come out on top.

That’s not for you to decide. It’s for the people with the $692m to decide.

He runs the 2nd most valuable company in the world, and the company is most profitable company ever.

Giving him 700m in stock to keep him around is worth it for both the investors and employees.

Isn't Google famous for the investors and the employees both being entirely powerless?

They spend 25b a year on stock compensation. Giving your CEO ~230m of that is probably okay.

The market disagrees.

The best alternative to Docker Desktop on macOS is to abandon the GUI entirely use colima to create the linux VM.

    brew install colima docker docker-buildx docker-completion docker-compose
    export DOCKER_HOST="$HOME/.colima/docker.sock" >> ~/.zshrc
    source ~/.zshrc
    colima start --cpus 8 --vz-rosetta --ssh-agent --mount $HOME:w

then add this line to your $HOME/.docker/config.json

    "cliPluginsExtraDirs": ["/opt/homebrew/lib/docker/cli-plugins"],

that will get you a fast virtiofs VM with the latest docker, including compose and buildx. it may seem scary to replace an officially blessed tool like Docker Desktop, but i have had zero issues with colima. it isn't "docker compatible". it's docker. just need to run `brew upgrade` and `colima update` every once in a while to keep it up to date.

I've been using Finch (https://runfinch.com/) for the last few years, it basically is a managed Lima, nerdctl, containerd and BuildKit.

I just alias docker to finch and it just works.

If you're going full CLI on macOS, I've had the best experience with:

    brew install podman

Podman manages the linux vm for you automatically.

I've come to enjoy podman more than docker on my linux hosts anway; the default runtime (crun) is lighter than docker (runc), podman-kube-play is great for managing multi-container pods and is compatible with kubernetes. It also integrates very neatly with systemd. Of course there is the whole daemon-less and rootless side of the things as well..

I’ve been using OrbStack instead of Docker Desktop and gotta say, I’d not replace it with anything else. So if anyones looking for a more automated alternative, check out OrbStack.

I hear great things about OrbStack; unfortunately the licensing tied to their free offering doesn't play nicely with corporate environments (and we're cheap!).

I switched to Colima instead and couldn't be happier.

Thank you!!! Been struggling with time skew on Podman desktop for around a year now with no fix in sight. At least in the initial test since I saw your comment a few hours ago, this is working great!

Does this support volume bind mounts and port forwarding?

You can see the command above mounting a volume

”—mount $HOME:w”

Tried Podman about a month ago; UI was kinda meh, 'compose' was beyond unusable... then tried Colima (+ lazydocker, a lightweight, beautiful TUI providing the little overview I need) and haven't looked back for a single second.

Same, been running on colima for years now and haven't had a single issue with it

I you fancy needing a GUI, rancher desktop is decent.

every day the US strays further from the light :(