I failed to find an example of a practical problem in the article. What I did find was a load of circling around the subject from a distant philosophical perspective, all boiling down to essentially "you know what, there exist problems that ownership semantics aren't suitable for" (but using lots of wisely-sounding language), forgetting to present the problems.

The article also suffers from the syndrome of taking a feature of a language and pretending that everything has to be seen through the lens of that feature. Well, newsflash: ownership semantics apply to only some things in Rust. They apply to values and bindings. But they don't apply to e.g. indices. I can take as create as many i's, where i=5, all referring to the same Vec, which later are going to be used to mutate the Vec, as I like, and Rust won't stop me. Similarly, there is Rc and Arc.

It also included one of those "wise" thoughts along the lines of "You know what? The true nature of computers/economy/life/love is..." - one of the templates of meaningless sentences. In this case it was "however virtually all computers are fundamentally mutable things". Hmm... Did we suddenly forget about pipelining, instruction ordering, caching, instruction-level parallelism, C/C++'s problems with memory aliasing? C is not a low-level language[1].

[1]: https://queue.acm.org/detail.cfm?id=3212479

If we were to write a statement analogous to the one advocated for in the article, but talking about mathematics instead, it could be "the fatal flaw of mathematics is that a lot of it is about reducing problems to linear algebra". How ridiculous does that sound?