Hacker Newsnew | past | comments | ask | show | jobs | submit | allyant's commentslogin

To be fair Microsoft haven’t done as much integration with their eco system; such as integrating it with Azure DevOps as they could. And long may it continue.


You missed the whole game-ification, enshitification, copilot-ification of github, and open source theft for their AI training episodes

Always like that with Microsoft apologists


Not a fan of defending GitHub, but TBH them launching GithubActions made me abandon all other CI platforms I was using in the past, I even left GitLab just for that.


What sold you on GitHub Actions? The one 9 of reliability?

https://www.githubstatus.com/


Not sure what you’re intending to show.

That page shows 99.5%+ uptime across all 11 of the services provided.

To do that at a small company world cost significant money.

To have it integrated and cheap for all those small shops is very valuable to them.


It shows 98.3% for actions for me

https://ibb.co/PsMkpJFY


99.5% uptime is nothing significant...


Yeah but it's sufficient for the largest part of CI needs. Most people are making CRUD apps, 6 9s reliability is not required there.


not really, I've ran into Github Action downtime and you can't run tests or anything so either you wait or you just let stuff merge

and again don't know how this is acceptable in 2026, for a paid CI service

and I'm not talking about 6 9s, I mean, 99.5% is terrible


A 99.5 means over one year you will need to wait about 40 hours total for the issue to be solved. What kind of high pressure crud app are you making where this is a problem?


we have technology, 99.5% is embarrassing for a company like Microsoft

it's 2026, not 2016

and that 40 hr may be that over 40 days you just have to randomly stop working if you depend on GHA for something

and why are other services not struggling like this?


I never said I paid, I was referring to my opensource repos.


Oh yes I can totally believe they use public and private repos for their model training. It is surprising they aren’t ahead of the curve for AI coding like Claude code, codex or cursor because of this.


I very much doubt that. Imagine the PR nightmare and potential liability if someone managed to extract from the model credentials in a private repository.


Explain to me how using OPEN source for training is theft? I've (admittedly skimmed) quite a few licenses, but I really dont remember anyone restricting reading/indexing source code?


MIT license includes: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

The argument is that AI models trained on open source should include the copyright info for all projects blended into the weights. And since they don’t, it violates the terms of the license.


You don't relinquish copyright by using a FOSS license though, right?


Github hasn't changed that much overall.

The copilot service and the training on open source would be the same if Microsoft hadn't bought GitHub, it would just be branded differently.


Have you seen github outages?


Well a lot of the outages are copilot which doesn't change my argument that already holds copilot separate.


This is one thing I think they need to get in-line with, and rename CLAUDE.md to AGENTS.md to follow convention.


To be fair, I think Anthropic/Claude started doing CLAUDE.md before AGENTS.md was a thing.


I'd say if we criticize that they did not think that far ahead, we are still pretty fair.


I strongly disagree. It is extremely annoying when I cannot make different agents read a different prompt file when I needed to tell them different things. Cursor currently autoloads AGENTS.md without a way to disable it and it sucks.


They ain't giving up that free marketing.


On one hand this is lovely code and seems to work well, on the other hand there are currently estimated to be 2.86 million call center staff in the US alone.


Project creator there. The biggest challenge in marketing this project to potential customers is not the fact that it does not work well, but the unions and human problems. It really marked me.


It would be good to have some example commands that can be ran right after installation, rather than having to figure out how to run the queries.


https://tivimate.com/ is about the nicest, but yes the IPTV scene doesn't have a nice UX at all, naming conventions (or lack of) especially.


On Android TV, I prefer Sparkle TV (https://play.google.com/store/apps/details?id=se.hedekonsult...) than Tivimate.


I wonder as an experiment if you are able to run an instance of the parent OS via mounting /:/ and using an image that matches the parent OS…


The situation in Iran is somewhat different. Despite being under heavy international sanctions and having a government that also exercises strict control over internet access, Iran saw the deployment of Starlink terminals following specific events, such as natural disasters or in response to certain socio-political situations. This deployment could be part of a broader strategy by the U.S. or other entities to provide uncensored internet access to the Iranian population during times of crisis. However, it's important to note that such deployments might be temporary, unofficial, or not fully sanctioned by the Iranian government.


Cannot seem to access the web UI also.


I was thinking current laser weapon tech would be ideal for this.


Not using access keys- although using the cli you could create a user which does have the ability to login to the portal.


Untrue, any valid access key pair set can call “GetSignInToken” and access the aws console.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: