Sebastian Thiel

As far as I know, `gitbutler.gerritMode = true` is the only flag to determine whether Gerrit support is enabled or not. If it doesn't show up consistently, this *might* be...

Thanks a lot for reporting! I agree that in an effort to become a good Gitizen, GB shouldn't avoid breaking with standard Git behaviour unless there is a very good...

Thanks a lot for making me aware! > I do not know what distinguishes my local Windows system from the Windows virtual machine on which the test was found to...

Thanks so much for gathering everything there is to know about this issue! I'd hope that one of the existing tests for delayed process filters can be used to also...

Regarding possible fixes, and knowing that I merely skimmed the issue after the first paragraph to not spend my time budget on this issue alone, I can share that I...

I am extremely impressed by this analysis! It seems more exhaustive than I could ever have imagined, and it feels I learned a lot and can only hope to recall...

Since all global RNG calls are automatically done per thread, I'd think that arranging one call per thread to `seed()` should overall be simpler. However, I wouldn't know what offers...

Maybe it's also changes in the random number generator that lead to 'less randomness' overall. Maybe it's a timing issue because CI got faster - reasons for this can probably...

> As far as changes in `fastrand` itself, I don't think any of those have affected the changes in my observations, since the most recent release of `fastrand` was [about...

Thanks a lot for reporting, the video makes the issue crystal clear! Indeed, in the plain-text editor one would least expect such effect, apparently some smartness is still going on...