I remain surprised/impressed that so many people from that era simply
decided to write their own Unix clone, no small undertaking at the
time.
+1
I remember Leor bringing BDS C and his Unix emulation for CP/M to a Boston USENIX and showing it to a number of us in his hotel room. At the time we were all always worried about what IP was 'open' and what was not (remember this is before Judge Green, much less the AT&T/BSDi case). Dennis' comment at the time was it reminded him of early UNIX in look and feel from the 11/20 days. Which considering the Z80 was a 64K address space is not surprising. IIRC: Leor's system had a bank switched memory scheme so he could get more physical memory in it was a trick that was in vogue for the S-100 bus systems. But it was an 8" floppy based machine.
I looked at it longingly because I could not afford to own an S-100 based system of my own!
A year or two later, Onyx brought their Z8000 based system to the UDel USENIX and wowed us again.