- The New York Times suffered a breach of its GitHub repositories in January 2024, leading to the theft and leak of sensitive personal information of freelancers.
- Attackers accessed the repos using exposed credentials, but the breach did not impact the newspaper’s internal systems or operations.
- The stolen data, amounting to 273GB, was leaked on 4chan and included various personal details of contributors as well as information related to assignments and source code, including the viral Wordle game.
Hot take: GitLab is sluggish, buggy, crap. It is the “Mega Blocks” of source control management.
If you have source files that are more than a few hundred lines and you try to load them on the web interface, forget about it.
They can’t even implement 2FA in such a way that it isn’t a huge pain to interact with. There’s been an open issue for over 7 years now to implement 2FA like it is everywhere else, where you can be signed in to more than one device at a time if you have 2FA enabled (https://gitlab.com/gitlab-org/gitlab/-/issues/16656).
Not to mention this was not a GitHub failure, this was a failure by the NYTimes to secure their developer’s credentials. This “just in house/self host everything and magically get security” mentality that’s so prevalent on Lemmy is also just wrong. Self hosting is not a security thing, especially when you’re as large of a target as NYTimes. That one little misconfiguration in your self hosted GitLab instance … the critical patch that’s still sitting in your queue … that might be the difference between a breach like this and protecting your data.
Forgejo?
Haven’t used it first hand, but I think it’s more promising.