News for hackers. Stuff that matters.
Re:GitHub Source Code Leak by jfrunyon on Wednesday November 04, 2020 @10:49PM (#24995486)
They... do? Do you see the Verified note anywhere? It's not their fault if people don't understand how the tool they're using works at an extremely basic level...
Re:GitHub Source Code Leak by hvdijk on Thursday November 05, 2020 @03:11AM (#24996480)
They don't. The comment was on labelling. There is nothing in the presentation of unsigned commits to indicate that they are unsigned. The presentation is indistinguishable from that of a hypothetical GitHub that never shows commit signatures, you need to have seen a signed commit on GitHub at some point to know that the absence of that Verified note is significant.
Re:GitHub Source Code Leak by jfrunyon on Sunday November 08, 2020 @11:00PM (#25031335)
Correct. Which is also how git works. And there is a very good, very simple, reason for why both work this way: very few users sign their commits, even fewer want to/care about signing their commits, and even fewer verify those signatures.
Re:GitHub Source Code Leak by kelnos on Thursday November 05, 2020 @01:33AM (#24996112)
So few people sign their commits that the default is to assume commits are from where they say they are, even if they're not verified.
Re:GitHub Source Code Leak by jfrunyon on Sunday November 08, 2020 @10:09PM (#25031103)
Please let me know when you figure out what GitHub could reasonably do to make enough people sign their commits to change that default.
Re:GitHub Source Code Leak by jason_zig on Thursday November 05, 2020 @10:06AM (#24998848)
That's absolutely their fault from a product perspective.
Re:GitHub Source Code Leak by jfrunyon on Sunday November 08, 2020 @11:01PM (#25031339)
How do you figure? I don't consider "making the product which users want" to be a bad thing.
Re:GitHub Source Code Leak by scintill76 on Thursday November 05, 2020 @04:34AM (#24996821)
That’s the opposite, labeling signed commits.