UML is a joke. It's the attempt of people who don't like programming to avoid it and do code-free software engineering. The problem is that this ultimately just removes the main difficulty of figuring out the implementation details and creates a world of make-believe where programs don't need to actually compile or execute.
The obvious exception is, of course, when discussing different implementation approaches, as a short-term thing, but not as documentation.
As you pointed out, most of these diagrams express how something is implemented (structure), not its behaviour. I hadn't considered it from this perspective yet, but you're definitely right. A good diagram would express primarily what code does, not how it does it. However, some diagrams manage to convey at least a bit of behaviour as well, but even then writing a simple markdown document can do a much better job.
Trust your tests. They are alive and well maintained.
Maybe one could use diagrams to describe these tests. At least on the surface this seems like it could work.
For further actions, you may consider blocking this person and/or reporting abuse
We're a place where coders share, stay up-to-date and grow their careers.
UML is a joke. It's the attempt of people who don't like programming to avoid it and do code-free software engineering. The problem is that this ultimately just removes the main difficulty of figuring out the implementation details and creates a world of make-believe where programs don't need to actually compile or execute.
The obvious exception is, of course, when discussing different implementation approaches, as a short-term thing, but not as documentation.
As you pointed out, most of these diagrams express how something is implemented (structure), not its behaviour. I hadn't considered it from this perspective yet, but you're definitely right. A good diagram would express primarily what code does, not how it does it. However, some diagrams manage to convey at least a bit of behaviour as well, but even then writing a simple markdown document can do a much better job.
Maybe one could use diagrams to describe these tests. At least on the surface this seems like it could work.