As part of my research, I do know a lot of jargon about nuclear physics and computing. When I speak
to non-experts i do make an attempt to not assume that the person I am talking to knows my jargon.
When I was a graduate students at Edinburgh, we once got a new supercomputer. A Proffesor in another research group, asked me how fast the new machine was. I told him, it could calculate 10 propagators a night, which he didn't understand.
It is in important is software project that the developers learn how to speak to the clients, so that they produce useful software.
It is particularly important to try and get the language correct, when talking to students, who are just starting on their journey to becoming experts. When I taught statistics to engineers I did try to add lots of engineering examples.
I came across a blog post about a succesful essay submission for part of their PGCAP course. I can see why the PGCAP people who marked it like it. They probably come from a sociology background, so they are happy to see a discussion of teaching in terms of the French philosopher Foucault. He didn't read Focault for his PGCAP course, and probably he had learned about it from a course he took as an undergraduate. There are a couple of references to PGCAP sources, but it is not clear that he had read them. So essentially he was passed, because he shared the cultural references of the PGCAP markers, rather for any attempt to learn their teaching methods.
I might add that it looks as though the guy, who wrote the blog post is a good teacher.
Of course, those of us, who have decided that postmodernism is rubbish, are at a disadvantage. So, as a student is expected to learn a new set of jargon, but the teachers don't want to bother.