UUtils has, until now, been a niche project with few users. Putting it in the most popular desktop distro is going to expose it to many new users and use cases. Some of those are bound to find differences in behaviour between UUtils and GNU which should be considered bugs. No doubt.

But this “not-very-well tested” mantra is just silly. UUtils itself uses the exact same test suite as GNU does. They have been testing against this suite for years:

https://github.com/uutils/coreutils-tracking/blob/main/gnu-results.svg

While not all tests pass yet, the subset of functionality that people are likely to actually use is very well tested.

And the reason some bugs were found recently is precisely because UUtils were put through the normal test cycle for Ubuntu. A small number of bugs were caught which is the goal of that process. These are things that were previously not in the test suite. I see there are some new tests. UUtils may have contributed to that as new use cases were encountered that showed differences in behaviour between GNU and UUtils. The issues discovered were quickly fixed.

Think of what is involved in creating a distribution like Ubuntu and building the tens of thousands of packages that they ship in their repos—all with build scripts written for GNU Coreutils. This is all working with UUtils unmodified.

With the distro live now, the number of users will have already exploded. Where are the bug reports and articles about all the problems encountered? Crickets.

That does not mean there will not be any such cases. That is not my point at all. My point is that “not very well tested” does not jive with how well things are going considering what a massive change this is.

UUtils is much better tested than much of the software I use.