|
|
|
|
|
by DisgracePlacard
1025 days ago
|
|
> If we're going to be trusting some random guy's binaries, I think we are in the right to demand that it is byte-for-byte reproducible on commodity hardware I don't think anyone has a right to demand anything of the project. The MIT license specifically has the whole "THIS SOFTWARE IS PROVIDED AS IS" spiel for a reason. Thinking that you can demand anything of an open source developer who, afaik, has no responsibilities in relationship towards you, is a rather toxic mindset that should be kept out of open source. |
|
What a tiring argument. Why complain about anything? Why do anything differently? Maybe it's because people have invested time, money, and effort into supporting and using the project and because of your stupid decision they now have to do more work. There were a lot of comments about people showing up that monday to pin+vendor the old SerDe. Others planned to remove it entirely in favor of other libraries. Almost like actions have consequences even if your "license agreement" says otherwise.
> is a rather toxic mindset that should be kept out of open source.
I wish I lived in your echo chamber where everyone who uses your stuff is just totally, like, accepting of your garbage ideas. The precompiled binary idea was a classic garbage idea. Refusing to make it reproducible was just the corn on top of the cowpie. The author made a major screw up, doubled down, tripled down, and then finally gave in. The author has no idea what scale and scope of project this is used in. They also clearly did no evaluation on potential damage OR ask for feedback before moving it into mainline. For a library with 3M+ downloads this was the what third? Fourth? Classic ego-driven folly.
You know what else shouldn't exist in open source? Ego tripping morons.