One of GitHub's most staple contributors announced they are abandoning ship due to constant outages. GitHub's COO responds, promising change, but is it all too little too late?
Github has not even one-nine of uptime. Normally you want three-nines or four-nines, they have ZERO-nines. A server in your basement is worlds more reliable.
I wonder what exactly they screwed up to bork it like this. It would seem like a no brainer to leave all the git stuff alone and add all the random fancy AI stuff in a separated manner.
Yeah, and the worst thing about this is that Github is critical infrastructure. If Github goes down the drain, so many devs and projects will be affected
We already went through this with SourceForge’s enshittification back in the day, to the point that sometimes people called it “SourceForget”. We’ll survive the GitHub-pocalypse too, it will suck, but we’ll be even better on the other side, at least until the next great centralization and enshittification.
The great thing about git is, that it is pretty decentralized in principle (everyone has a full copy of all source code and commits on their machines), so it is pretty easy to move your whole repository to an alternative git hoster, like Codeberg.
I keep hearing about codeberg. Yet, when Claude CLI was leaked — I found the repo on some weird blockchain git repository with a message like “It’s here forever now…”
For OSS and personal projects, wouldn’t a blockchain solution actually be pretty good?
Not really. Blockchain technology has one use case and that is collaboration between partners who don’t trust each other. So we’re talking crypto coins, where not all nodes are really trustworthy and there is an incentive to cheat. But there’s no reason to bring this tech to your Git repository because you really do not want untrustworthy participants in your code. Only you should have access to your Git rep, and then the easier solution is to host it yourself and use a normal database.
Of course it fucking is, it runs Linux, not Winslowpes from Microslop. My basement server has 100% uptime, and I’ve got it for close to free (like ten bucks, literally). It’s an old Intel Atom powered desktop motherboard from circa early 2010s if not late 2000s. The uptime was real and literal 100%, but over time I started powering off, when I realised I don’t need it being on all the time. It still has 100% availability for when I need it. I should care more about backups, but the data is backed up, while the system … the thing is, I’ve learnt so much since I installed its system, almost a decade ago, that, I think I’d reinstall it. It’s Arch Linux, which technically doesn’t need to be reinstalled, but it uses quite a lot of actually old things I don’t bother changing.
Okay, I might be not correct, I bet Microslop runs everything of importance on Linux too. It’s rather their stack is very heavily slopped, that’s my wild guess why it’s down all the time.
Github has not even one-nine of uptime. Normally you want three-nines or four-nines, they have ZERO-nines. A server in your basement is worlds more reliable.
I wonder what exactly they screwed up to bork it like this. It would seem like a no brainer to leave all the git stuff alone and add all the random fancy AI stuff in a separated manner.
Even the most hackiest, quickly coded with no regard for other devs systems at work have one 9, it’s fucking pathetic.
Hm, interesting, I can not remember a single time in the last 10 years where github has any issue for me.
Contrary to that I know “nine” availability services that failed a lot of time.
96 issues in the last 90 days.
There’s two nines right there! Just not the ones you need.
Yeah, and the worst thing about this is that Github is critical infrastructure. If Github goes down the drain, so many devs and projects will be affected
Our company has had fits with GitHub the past month. It feels like every day something is busted.
Our company is also drinking the AI kook aid though and can’t see the forest for the trees.
We already went through this with SourceForge’s enshittification back in the day, to the point that sometimes people called it “SourceForget”. We’ll survive the GitHub-pocalypse too, it will suck, but we’ll be even better on the other side, at least until the next great centralization and enshittification.
The great thing about git is, that it is pretty decentralized in principle (everyone has a full copy of all source code and commits on their machines), so it is pretty easy to move your whole repository to an alternative git hoster, like Codeberg.
Except all the extra stuff like CI, issues, pull requests, discussions, pages, and probably some more things.
Forgejo has options to import some of that too, but it’s not that easy. A modern repository isn’t just files in git.
I keep hearing about codeberg. Yet, when Claude CLI was leaked — I found the repo on some weird blockchain git repository with a message like “It’s here forever now…”
For OSS and personal projects, wouldn’t a blockchain solution actually be pretty good?
Edit: found it https://gitlawb.com/node/repos/z6MkgKkb/instructkr-claude-code
Not really. Blockchain technology has one use case and that is collaboration between partners who don’t trust each other. So we’re talking crypto coins, where not all nodes are really trustworthy and there is an incentive to cheat. But there’s no reason to bring this tech to your Git repository because you really do not want untrustworthy participants in your code. Only you should have access to your Git rep, and then the easier solution is to host it yourself and use a normal database.
Of course it fucking is, it runs Linux, not Winslowpes from Microslop. My basement server has 100% uptime, and I’ve got it for close to free (like ten bucks, literally). It’s an old Intel Atom powered desktop motherboard from circa early 2010s if not late 2000s. The uptime was real and literal 100%, but over time I started powering off, when I realised I don’t need it being on all the time. It still has 100% availability for when I need it. I should care more about backups, but the data is backed up, while the system … the thing is, I’ve learnt so much since I installed its system, almost a decade ago, that, I think I’d reinstall it. It’s Arch Linux, which technically doesn’t need to be reinstalled, but it uses quite a lot of actually old things I don’t bother changing.
Okay, I might be not correct, I bet Microslop runs everything of importance on Linux too. It’s rather their stack is very heavily slopped, that’s my wild guess why it’s down all the time.