Oh shoot - I read your post, but then I just skimmed it again after checking fedidb and lost track of the context.
I think the point still hold that kbin is not at a point in its development where it makes too much sense for people to open independent kbin instances. By the way I understood it, mbin seeks to patch some of these problems, while the kbin development is focusing on improving the infrastructure on a deeper level. The drama is a bit silly, but I think forking is not always a bad thing and the differences in development between kbin (deep changes to the code base) and mbin (patches to make things work) are enough to justify parallell developments for the time being.
It's also worth noting that not a single new kbin instance has been created since September this year - I think for now you're unfortunately not likely to find an actively maintained instance beside the two run by Ernest.