It probably depends on your scale, but I'd suggest self-hosting a Forgejo instance, if it's within your domain expertise to run a service like that. It's not hard to operate, it will be blazing fast, it provides most of the same capabilities, and you'll be in complete control over the costs and reliability.
A people have replied to you mentioning Codeberg, but that service is intended for Open Source projects, not private commercial work.
i would imagine that's what everyone is doing instead of sitting on their hands. Setup a different remote and have your team push/pull to/from it until Github comes back up. I mean you could probably use ngrok and setup a remote on your laptop in a pinch. You shouldn't be totally blocked except for things like automated deployments or builds tied specifically to github.com
I've been using https://radicle.xyz/ + https://radicle-ci.liw.fi/ (in combination with my own ci adapter for nix flakes) for about half a year now for (almost) all my public and private repos and so far I really like it.
> What are good alternatives to GitHub for private repos + actions? I'm considering moving my company off of it because of reliablity issues.
Dunno about actions[1], but I've been using a $5/m DO droplet for the last 5 years for my private repo. If it ever runs out of disk space, an additional 100GB of mounted storage is an extra $10/m
I've put something on it (Gitea, I think) that has the web interface for submitting PRs, reviewing them, merging them, etc.
I don't think there is any extra value in paying more to a git hosting SaaS for a single user, than I pay for a DO droplet for (at peak) 20 users.
----------------------
[1] Tried using Jenkins, but alas, a $5/m DO droplet is insufficient to run Jenkins. I mashed up shell scripts + Makefiles in a loop, with a `sleep 60` between iterations.
For me it is their history of high-impact easily avoidable security bugs. I have no idea why "send a reset password link to an address from an unauthenticated source" was possible at all.
Nah at a small scale it's totally fine, and IME pretty pain-free after you've got it running. The biggest pain points are A) It's slow, B) between auth, storage, and CI runners, you have a lot of unavoidable configuration to do, and C) it has a lot of different features so the docs are MASSIVE.
Not really. About average in terms of infrastructure maintenance. Have been running our orgs instance for 5 years or so, half that time with premium and half the time with just the open source version, running on kubernetes... ran it in AWS at first, then migrated to our own infrastructure.
At my last job I ran a GitLab instance on a tiny AWS server and ran workers on old desktop PCs in the corner of the office.
It's pretty nice if you don't mind it being some of the heaviest software you've ever seen.
I also tried gitea, but uninstalled it when I encountered nonsense restrictions with the rationale "that's how GitHub does it". It was okay, pretty lightweight, but locking out features purely because "that's what GitHub does" was just utterly unacceptable to me.
One thing that always bothered me about gitea is they wouldn't even dog food for a long time. GitLab has been developing on GitLab since forever, basically.
ad hominem isn't a very convincing argument, and as someone who also enjoys forgejo it doesn't make me feel good to see as the justification for another recommender.
From [1] "Forgejo was created in October 2022 after a for profit company took over the Gitea project."
Forgejo became a hard fork in 2024, with both projects diverging. If you're using it for local hosting I don't personally see much of a difference between them, although that may change as the two projects evolve.
I'm not OP, but; Forgejo is much lighterweight than Gitlab for my usecase, and was cited as a more maintained version of Gitea, but that's just anecdote from my brain and I don't have sources, so take that with a truckload of salt.
I'd had a gitea instance before and it was appealing insofar as having the ability to mirror from or to a public repo, it had docker container registry capability, it ties into oauth, etc; I'm sure gitlab has much/all of that too, but forgejo's tiny, tiny footprint was very appealing for my resource-constrained selfhosted environment.
Having a layer of parallelisation on top of good old sequential code seems like a very neat idea. It resolves headaches of learning how to run code in parallel in languages that aren’t necessarily my primary language (e.g. short, one-off scripts). Thanks for sharing!!
I've used tRPC and Next.js for a couple of personal projects and it's been a great experience. Hard to beat on iteration speed, especially when used with a pre-configured template like Create T3 App: https://create.t3.gg/.
I think patterns and best practices are still yet to be figured out. I believe you can create caller in a server component and it should work (https://trpc.io/docs/server/server-side-calls#create-caller), but the pages router appears to be a battle-tested solution.
Thank you I only briefly read about them before and didn't consider them properly.
Will try out calling the database directly there, but I kinda liked how with tRPC I can add validation/auth checks etc/ will have to see how I develop my own strategy for that w server actions.
Author here. My intention was to show that you can use it to help you get going quickly for a very practical, one-off, and self-contained use cases. As I mentioned in other comments already, I did not trust it blindly and did not share any sensitive data with it. Definitely not an ad!
100% agreed, this is how I always treat it and working on the problem from the article was not an exception from this rule. I share minimum input, and never trust the output blindly.
It gets 50-60% of work done, and a really good basis for me to work on it. Especially when working with one-off, end-to-end relatively short scripts.
This has been my primary use case as well (usually for writing some scripts or where I need to solve an operational task quickly), and ChatGPT has saved me a ton of time with those tasks.
Ya, for Mastercard we use their Ethoca network. They are much more expensive, like $25 per resolved charge but now our chargeback rate is near 0% for Visa / MC and get incredible rates on the front end from such clean processing. Plus we never have to worry about chargebacks threatening our merchant account again.
Article author here. I carefully reviewed and tested the ChatGPT scripts before executing them. It helped me save a lot of time manually writing these scripts!
I wouldn't say I lack technical expertise in this area, I'm just trying to use my time as efficiently as possible.
Genuinely curious: How much time would you say you saved prompting for and then carefully reviewing and testing those scripts for bugs, versus writing them yourself?
And for context what's the average line count we're talking about here? Tens of lines? Hundreds?
Glad to see you active here in the comments. Apologies if my comment comes off harshly, my intent is not to tear you down. I think there is a lot of gray space when it comes to using LLMs for generating code. Your usage here is certainly interesting, and I appreciate the additional context and discussion you’ve been providing.
No worries at all! I agree that there's probably lots of people blindly copying and running code from LLMs without any reflection. Just like it often happened with StackOverflow snippets before ChatGPT (to the point it became a meme). I'm definitely not one of them.
