LICENSE:

• Wrote a custom LICENSE file for my repo — free for humans, locked out for AI
• Humans can copy, modify, and publish the code freely (MIT-style)
• AI systems and AI operators (companies, model trainers, agents, etc.) get zero rights — no reading, training, scraping, or code-gen assistance
• Exception: I (original copyright holder) can grant separate signed written permission to specific AI operators
• Found out non-ai-licenses on GitHub already did something similar — worth checking out too
• Not OSI-approved as "open source" since it restricts a field of use (AI training)
• Doesn't technically stop scraping in real-time, but makes it a clear license violation / copyright claim if caught

What about this ? should something like this default for new projects ?

I'm pretty new to GitHub, and I needed to use it for school. I thought I needed GitHub enterprise and I saw it had a 30-day free trial so I signed up for it. June 13 until July 13 the conformation email said. Ended up not needing it, so I deleted it or tried to. Apparently something went wrong, but I did get an email stating something got downgraded to a free plan. So I assumed it was taken care of.

Few days ago I got a payment for GitHub enterprise, that shouldn't have happened. So I opened a ticket and explained what happened and asked for a refund:

First response I got said that, because there was activity in June I couldn't get a refund. My response, was that the activity was during a free trial and I shouldn't even have to pay for that. (I was under the impression that I paid for July)

Support responds by adding a receipt as proof that I paid for June and not July and that the service period was for 2026‑06‑13 to 2026‑06‑30. So I asked why I got charged for a free trial.

Support responds that I apparently cancelled my free trial on the same day as starting it (Why would anyone ever) And for some reason that chanced it to a paid version. I never got any email confirming that I cancelled my trail, meaning I had no way of knowing or confirming that information myself.

After that response I contacted PayPal and send a detailed description with proof of emails hoping they could mean anything for me. I send another comment to GitHub with a summary of what happened and the information given to me, with a final plea for my refund. PayPal approved my request the next day and i FINALLY got my refund.

Excuse my swearing: Holy shit did they make it difficult for me to achieve anything in this conversation. I felt like they kept pulling things out of their ass. I'm usually not the type of person to complain about customer service, but they were not helpful in the slightest. I tried finding things online about there customer support, but couldn't find many similar stories. Most of the complaints were about long waits, which i surprisingly didn't experience. Is this common for them? Or did I just get unlucky?

A few years ago there was always a debate about which CI/CD tool to use for open source projects. Travis CI was the goto for a long time, then CircleCI picked up steam, and Jenkins was always lurking in enterprise corners. But lately it feels like GitHub Actions has quietly become the default assumption for most new repos.

I've been setting up a few new projects recently and noticed that almost every template, starter repo, or tutorial just assumes you're using GitHub Actions. The marketplace has grown a lot too, and the free tier for public repos makes it hard to justify setting up anything else from scratch.

What I find interesting is how deeply integrated it is now. You get Actions, Packages, and security scanning all in one place, which changes how you think about your workflow from the start.

But I'm curious whether this consolidation is actually a good thing for the ecosystem. Are we losing something by having one platform dominate both code hosting and CI/CD? Has anyone intentionally stuck with an external CI tool even for new projects, and if so, what was the reason?

Would love to hear from people maintaining larger open source projects especially. Is GitHub Actions genuinely good enough, or is convenience doing a lot of the heavy lifting here?

Imagine you've just opened a repository you've never seen before.

What do you check first, and what immediately makes you think, "This project is well maintained"?

https://preview.redd.it/wbgp3ih5hobh1.png?width=1332&format=png&auto=webp&s=a43d6fa2a2dad519a0a142cb3cedd44b2c1ef713

How do I find and remove the file or use GitLFS to upload it?

The June 30 changelog announced access restrictions on the public API endpoints and UI views that expose stargazer and watcher lists. The stated reason is reducing spam and scraping abuse.

I'm trying to understand the real motivation here. Is this genuinely about protecting users from scrapers, or is it more about GitHub controlling access to data that third-party tools (star history, trending trackers, research) were built on? Those two explanations lead to very different conclusions, and I can't tell which one it is.

Either way it's a reminder that everything we treat as "public" on GitHub is public at one company's discretion, and can change with a single changelog entry. No vote, no notice, no contract.

So the honest question I keep coming back to: should more of us be moving toward open, decentralized platforms that can't do this? Self-hosted options like Forgejo and Codeberg exist, and newer federated projects like gitlawb pin repos to IPFS with signed refs across independent nodes so no single operator can restrict data. Or is GitHub's convenience and network effect just worth the tradeoff, and this is an overreaction?

Curious what people who actually live in GitHub every day think. Does a change like this move the needle for you at all, or is it a non-issue?

https://preview.redd.it/t35uf670znbh1.png?width=1902&format=png&auto=webp&s=6aec5519d72f837d6f3ec16341fb1acffb557af0

this was right after making it and its over the limit to where i have to use GitLFS, this didnt happen in the tutorial

I'm thinking about building an open-source project in Go and would like to validate the idea before investing a lot of time into it.

The idea is a self-hosted GitHub notification platform.

Instead of sending plain webhook messages, it would transform GitHub events (Pull Requests, Pushes, Releases, etc.) into clean, readable notifications for Telegram, Discord, Slack and other platforms.

Some features I'm considering:

- Self-hosted (Docker / Docker Compose)

- GitHub Webhooks

- GitHub App support

- Beautiful notification templates

- Human-friendly changelogs

- Optional AI-generated PR summaries

- Retry queue for failed deliveries

- Delivery logs

A few questions:

- Would you actually use something like this?

- Are there any mature open-source alternatives I should look at?

- What do existing solutions do poorly?

- What feature would make you switch to a new solution?

I'd really appreciate honest feedback, criticism, or links to similar projects before I start building it. Thanks!

I want to change my username/handle on GitHub to @ Asteroider, but the settings dialog for it says that name is unavailable. However, when I go to https://github.com/Asteroider/ it gives me a 404, whereas other options like @ Asteroid and @ Asteroiderer do have accounts that I can view.
Is there any way to fix this, since GitHub support is apparently like YouTube's in that they are so high and mighty they only support people who can pay or make money on their platforms?

My brother who is a Design Engineer (Civil) is currently building a software product related to his construction field. Software is in MVP stage and it is being used by few users . It's developed as an (.exe) file, he currently shares it with people through "we transfer". Wanted to know couple of things as he is very new to this..

1. Is therr a better way of sharing the, .exe with others?

2. How can he protect the source code, is it possible for users to access the source code after installating the application or sharing the package files through we transfer

3. Related to code signing, after installing exe it asks for security related questions like is it safe to install or similar question. We want to avoid that, researched about code signing. Does it help?

Please if some can help us with above questions. It's of a great value for us.

Any suggestions for a self-paced course online I can take to get the hang of GitHub. I’m finishing up a python bootcamp on Udemy and I think it would be a good next step. I need something I can track and share with the department that’s supporting my studies so it can’t be just the documentation or that pro git book someone else posted about. Thanks in advance ☺️

maintainer of a small design-system extraction CLI here. checked the insights/traffic tab on my repo this week: chatgpt.com sent 22 visits in 14 days, google sent 21, reddit sent 1. three weeks ago chatgpt was at 5.

i never did anything to optimize for this. best guess: the readme is structured like a direct answer to "how do i extract a website's design tokens", and LLMs cite it when people ask.

curious if other maintainers are seeing ai assistants show up in their referring sites. is this a blip or is "GEO" actually becoming a thing for dev tools? worth checking your own traffic dashboard, the data is under insights > traffic and almost nobody opens it.

For me, it's usually figuring out which ones are genuinely different versus slight variations of the same idea.

Is there a signal that immediately tells you a repo is worth a closer look?

Is there a way I can do this, so any time I update one it updates the other?