This seemed familiar? The first few lines/commands are a straight copy and paste of a post that reached front page 8 months back: https://maurycyz.com/misc/easy_git/ (HN post [here](https://news.ycombinator.com/item?id=45710721)). The rest is just AI rewording. Even the quips about wiring this up to the blog's site generator are stolen. Even the _date format_ (which is different from the rest of the sites blogs) is stolen.
However, this does not look affiliated at all with the real author's blog. Exploring this vercel site, looks like they rip dozens of creators blog posts day for SEO spam? And for what? Their homepage (and its non-functional demo) is just for a link preview feature? Something that many browsers already provide natively (and if not, you can just use the many existing browser extensions for this)?
Is this a giant spyware/malware promotion? I'm very confused.
OP (https://news.ycombinator.com/user?id=sheelagay) seems to be affiliated with the page and seems to be only posting this for the last few months; their first post is promoting/showcasing the "product".
When I started using Git around 2006, we didn't have GitHub. We used Google Code for SVN hosting, but we wanted to try Git because it was 'truely' decentralized, so we just pulled from other developers' computers directly. It involved setting up port forwarding on the router and letting your computer on 24/7. Fun times.
You can also have an upstream git repo on a local filesystem and add that as a remote.
But when most people talk about a git server that includes collaboration and giving people SSH access just so that they can read from your repo is not reasonable.
Nobody seems to have really embraced the truly distributed model of git, where you can expose a local repository via read-only HTTP, and collaborate by pulling from each others' repositories. No pushes.
This would be unwieldy in a corporate environment and for those who don't really grok git, but for a small cadre of experienced developers, this may be a workable model.
Sourcehut _kind of_ pushes for this model. Folks publish their own repositories, and email patches to others. But you can also just email a patch to the author without a Sourcehut account, because email is descentralised.
The Linux kernel and u-boot also follow similar flows.
Honestly, it's incredibly convenient to send a patch via `git send-email -1`, instead of having to create a for, add a remote, push, and then navigate some web-based wizard.
Actual collaboration does not require write access on the repo. You need to share your patch to the repo owner via another mechanism (email is very suitable for that, but anything that can share text file is ok). The repo owner handle merging patches.
As a species, I think we're past the point of thinking headlines are for conveying accurate information. The primary purpose is algorithm manipulation and attention grabbing. Every decent content creator openly admits as much. Direct complaints to the Lords of Engagement. They won't listen, but no one else wants to hear it either.
I didn't know that was possible to allow pushing the checked out branch - but for completeness, the alternative would be to make a bare repo where simply no branch is checked out.
This is indeed a Git server. It is not a Git Forge though. Pull Requests, Issues, CI/CD, access management, etc. are about as important as a putting my code somewhere.
I find it strange the websites some of these HN articles are hosted on... going through the effort to write a post before getting a domain outside of .vercel.app is interesting.
That is it. No patience. No care. No attention to detail. No scratchpad with multiple edits. Just Claude and a prompt — the only two things you can tolerate.
I think it's just a general jab at the unwieldiness of yaml (because the model was in the "talk about yaml files" and "be witty" parts of the latent space I imagine), not some particular file.
Check the home page - it mentions 47 tabs. Tabs is obviously in a similar semantic space to indents, so the number 47 carried over. Or maybe this AI slopmachine just likes the number 47.
This seemed familiar? The first few lines/commands are a straight copy and paste of a post that reached front page 8 months back: https://maurycyz.com/misc/easy_git/ (HN post [here](https://news.ycombinator.com/item?id=45710721)). The rest is just AI rewording. Even the quips about wiring this up to the blog's site generator are stolen. Even the _date format_ (which is different from the rest of the sites blogs) is stolen.
However, this does not look affiliated at all with the real author's blog. Exploring this vercel site, looks like they rip dozens of creators blog posts day for SEO spam? And for what? Their homepage (and its non-functional demo) is just for a link preview feature? Something that many browsers already provide natively (and if not, you can just use the many existing browser extensions for this)?
Is this a giant spyware/malware promotion? I'm very confused.
OP (https://news.ycombinator.com/user?id=sheelagay) seems to be affiliated with the page and seems to be only posting this for the last few months; their first post is promoting/showcasing the "product".
When I started using Git around 2006, we didn't have GitHub. We used Google Code for SVN hosting, but we wanted to try Git because it was 'truely' decentralized, so we just pulled from other developers' computers directly. It involved setting up port forwarding on the router and letting your computer on 24/7. Fun times.
We did something similar, but used a dedicated server we already had on the network instead of another dev's machine.
You can also have an upstream git repo on a local filesystem and add that as a remote.
But when most people talk about a git server that includes collaboration and giving people SSH access just so that they can read from your repo is not reasonable.
Nobody seems to have really embraced the truly distributed model of git, where you can expose a local repository via read-only HTTP, and collaborate by pulling from each others' repositories. No pushes.
This would be unwieldy in a corporate environment and for those who don't really grok git, but for a small cadre of experienced developers, this may be a workable model.
Sourcehut _kind of_ pushes for this model. Folks publish their own repositories, and email patches to others. But you can also just email a patch to the author without a Sourcehut account, because email is descentralised.
The Linux kernel and u-boot also follow similar flows.
Honestly, it's incredibly convenient to send a patch via `git send-email -1`, instead of having to create a for, add a remote, push, and then navigate some web-based wizard.
That's how I learned git before GitHub, but it was a pain to configure DNS and port-forwarding when ISP didn't provide static IP.
Radicle doesn’t get enough credit. They’ve created a really excellent take on modern distributed git.
Why isn't it reasonable? Ssh works fine for the big repo providers.
Are you just saying that managing account access at the personal machine level is a lot of work?
I'm nodding along.
Git support http as a transport.
Actual collaboration does not require write access on the repo. You need to share your patch to the repo owner via another mechanism (email is very suitable for that, but anything that can share text file is ok). The repo owner handle merging patches.
As a species, I think we're past the point of thinking headlines are for conveying accurate information. The primary purpose is algorithm manipulation and attention grabbing. Every decent content creator openly admits as much. Direct complaints to the Lords of Engagement. They won't listen, but no one else wants to hear it either.
or use git-shell so users do not need a full ssh account on the server due to security concerns:
> git config receive.denyCurrentBranch updateInstead
I didn't know that was possible to allow pushing the checked out branch - but for completeness, the alternative would be to make a bare repo where simply no branch is checked out.
Huh, I've always set up bare repos as my "local" (on a desktop somewhere in my house) destination. I didn't realize you could just use regular repos.
You can but any edit to the working directory will render the repo ineffectual for pushing.
This is indeed a Git server. It is not a Git Forge though. Pull Requests, Issues, CI/CD, access management, etc. are about as important as a putting my code somewhere.
For access management (sharing the Git-over-ssh hosting for a centralized repo), gitolite adds the little layer.
https://gitolite.com
> This is a nice way to work on server-side files without SSH lag or error-prone copying.
Are people really copying files to the servers then committing there?
I find it strange the websites some of these HN articles are hosted on... going through the effort to write a post before getting a domain outside of .vercel.app is interesting.
> the effort to write a post
This post was pretty obviously written by llm. You can assume the llm also picked the hosting platform.
It’s a slop article based on a tweet.
> That is it. No Docker. No CI pipeline. No serverless function. No YAML file with 47 indents. Just SSH and git — two tools you already have.
Sigh.
That is it. No patience. No care. No attention to detail. No scratchpad with multiple edits. Just Claude and a prompt — the only two things you can tolerate.
Can you rewrite this by hand, please? I was so confused what this article was supposed to tell me.
> No YAML file with 47 indents.
What is this referencing? what yaml file?
The front-end also screams AI generated
I think it's just a general jab at the unwieldiness of yaml (because the model was in the "talk about yaml files" and "be witty" parts of the latent space I imagine), not some particular file.
Check the home page - it mentions 47 tabs. Tabs is obviously in a similar semantic space to indents, so the number 47 carried over. Or maybe this AI slopmachine just likes the number 47.
maybe just ask your chat ai as usual?
Looks like a bot doing self-promotion.
This is an AI generated article