Kudos for at least trying to address this, MS should hang their head in shame, this is not the hardest problem to solve these days. If we could do it in 1995 they should be able to do it 30 years later.
No, in Teams' case, they somehow managed to take a trivial problem that was solved quite well 30-40-odd years ago (albeit in a slightly different skin - IRC) and completely botch it in every way imaginable, and then a few more ways not even the most creative of QA engineer could have possibly imagined a team messing up such a basic problem set.
It's finally a little bit less bad than it was 2-3 years ago, so the trend line is slightly angling upwards out of hell now, where the bar has been, but that's really not saying much.