I was looking through the different data types, that were specified in the IEEE 754 documents and I found these weird ones called things like "decimal32" and "decimal64".

I had never heard of them before, I know how the standard "binary32/64" floats work but never have I come across these weird ones
and now that I am getting curious about them I am having a hard time finding good explanations on how these work,
there was an explanation on Wikipedia and the official document but those were very dense so I had trouble figuring them out.
Also, at least Wikipedia, mentioned,
that they are encoded differently on different systems?

So can someone please explain clearly how this weird "decimal32" float format works?

And also is there a reason,
why you come across it so rarely? Does it suck or something,
is it irrelevant or mostly unimplemented or for highly specialized tasks?

Would love to know, how this stuff works, thanks.