On Tue, 2025-11-11 at 15:13 -0500, Paul Koning via cctalk wrote: > Via one of the articles about that Unix V4 tape I found this > https://eylenburg.github.io/os_familytree.htm which is quite an > impressive chart of operating systems old and new. > > A few bits are clearly incorrect or missing; for example, P/OS and > TRAX are both not there. I was going to mention them to the author > of that chart but I don't actually know the correct placement. (Is > P/OS an RSX-11/M or M+ derivative? I should know but don't.) > > Neat to see RSX-15 mentioned as the ancestor of RSX-11/D. Did the > earlier RSX-11 versions ever ship? I know I have seen traces of /A > and/or /C... > > The DOS-11 and RSTS entries aren't quite right either, I can send > feedback for those. > > Also, while VMS is reasonably listed as RSX-11/M derivative, I always > thought that VAXElan was unrelated. Along the same lines, where > would MicroPower/Pascal fit? > > paul
Erstwhile Bell Labs denizen Stu Feldman, the author of "make" and the first (and slowest) FORTRAN 77 ciompiler, told me decades ago that when he was at MIT he and several others decided Multics was too big and too complicated, so UNICS was (originally) designed in reaction as a simpler single-user system — hence UNI instead of MULTI. Dessis Richie was a grad student at Harvard at the time, but he had a part time job in Project MAC. Ken Thompson also worked on Project MAC even though he had studied at Berkeley. I think it was Dennis who said about BSD "never, ever, give your source code to a grad student." He had given UNIX source code to Bill Joy, who used BSD to found Sun Microsystems.
