The worst thing that happened to Unix was the MBAs at AT&T and then Novell wanting to extract the maximum possible revenue from it. Whilst large vendors licensed the code, e.g. IBM, HP, Sun MicroSystems, sundry workstation vendors and Microsoft (Xenix!) it was off limits to universities. So Xinu, Minix and subsequently Linux emerged. BTW Plan 9 and Inferno were also subjected to the heavy handed monetization efforts.
Had AT&T taken an enlightened approach, the brilliant folks at universities would have worked to improve it (granted Berkley did contribute, but they too were sued) instead they wasted time reinventing the wheel.
People have preferences and sometimes they are irrational and even counter productive. That's what taste is. I like coffee more than tea.
I ran v7, on a pdp11 and enjoyed it immensely. I'm not blind to it's limitations. I doubt I ran a pure v7 since shebang worked, what I recall most fondly about v7 was the "learn" tutorials which allowed me to self pace my Unix birth experience.
What I disliked about v7 was mainly how "crude" it could be compared to 4.1bsd, which I was exposed to shortly afterwards. Pre-sockets, things we now take for granted in a network were less easy to negotiate and you confronted cu and tip as guardians to access over wires to other services. BSD packaged some of those away quite nicely and additionally the csh (pre tcsh) had some UX advantages in command recall, if not syntax compared to (Bourne) shell.
It probably helped that a vax was significantly faster than a pdp11. We ran the printers off the vax for a reason.
I am sure my love of v7 is nostalgia. I abhor the gatekeepers of "now you like it, it's not cool" but a distinction needs to be drawn between this, and how Bell Labs alumni act. The Bell people had to deal with fandom, and setting "you must be this high to ride" barriers weeded out a lot of stupid questions, and questioners like me. I asked about mgr and "the plumber" but missed some key points about the underlying abstraction and I'm not surprised "read the source" terminated the conversation.
Plan9 and inferno presage kubernetes in so many ways. They were more forward looking than the progression of v7 to 2.9bsd and 4.2bsd and system III/system V. It doesn't surprise me the world glued on to a different model, or that the abstractions of the Unix world took root on android. It also doesn't surprise me we see MIT X10/X11 as significant and only historians of systems point to Xerox or symbolics lisp machines, or even NeWS. I coded in sunview and X10, forth passed me by. Yet, postscript and hence pdf now dominate the world. X has frozen and Wayland struggles to push it aside where sunview and CDE quietly withered on the vine.
There's much less of VMS in windows, than POSIX these days. You wouldn't have said that was coming back in the 80s.
The worst thing that happened to Unix was the MBAs at AT&T and then Novell wanting to extract the maximum possible revenue from it. Whilst large vendors licensed the code, e.g. IBM, HP, Sun MicroSystems, sundry workstation vendors and Microsoft (Xenix!) it was off limits to universities. So Xinu, Minix and subsequently Linux emerged. BTW Plan 9 and Inferno were also subjected to the heavy handed monetization efforts.
Had AT&T taken an enlightened approach, the brilliant folks at universities would have worked to improve it (granted Berkley did contribute, but they too were sued) instead they wasted time reinventing the wheel.
People have preferences and sometimes they are irrational and even counter productive. That's what taste is. I like coffee more than tea.
I ran v7, on a pdp11 and enjoyed it immensely. I'm not blind to it's limitations. I doubt I ran a pure v7 since shebang worked, what I recall most fondly about v7 was the "learn" tutorials which allowed me to self pace my Unix birth experience.
What I disliked about v7 was mainly how "crude" it could be compared to 4.1bsd, which I was exposed to shortly afterwards. Pre-sockets, things we now take for granted in a network were less easy to negotiate and you confronted cu and tip as guardians to access over wires to other services. BSD packaged some of those away quite nicely and additionally the csh (pre tcsh) had some UX advantages in command recall, if not syntax compared to (Bourne) shell.
It probably helped that a vax was significantly faster than a pdp11. We ran the printers off the vax for a reason.
I am sure my love of v7 is nostalgia. I abhor the gatekeepers of "now you like it, it's not cool" but a distinction needs to be drawn between this, and how Bell Labs alumni act. The Bell people had to deal with fandom, and setting "you must be this high to ride" barriers weeded out a lot of stupid questions, and questioners like me. I asked about mgr and "the plumber" but missed some key points about the underlying abstraction and I'm not surprised "read the source" terminated the conversation.
Plan9 and inferno presage kubernetes in so many ways. They were more forward looking than the progression of v7 to 2.9bsd and 4.2bsd and system III/system V. It doesn't surprise me the world glued on to a different model, or that the abstractions of the Unix world took root on android. It also doesn't surprise me we see MIT X10/X11 as significant and only historians of systems point to Xerox or symbolics lisp machines, or even NeWS. I coded in sunview and X10, forth passed me by. Yet, postscript and hence pdf now dominate the world. X has frozen and Wayland struggles to push it aside where sunview and CDE quietly withered on the vine.
There's much less of VMS in windows, than POSIX these days. You wouldn't have said that was coming back in the 80s.