Should be part of cs100 :)
Forgetting history etc
Or ghosts of Christmas Past?
The crucial part about (biological) evolution is successful reproduction. The individuals best adapted to their prevailing environment are a dead end if their progeny (if any) don't themselves successfully reproduce.
Probably why the world is full of rapidly reproducing vermin and there aren't any libido-less superwombles.
The obviously analogy in the technological world is - if your product doesn't sell with enough volume or sufficient margins begetting the next is problematic. In the late '70s even the least expensive digital computer was beyond most - even with the advent of the first eight bit processors (especially outside north america.)
My first machine was a tandy colorco (M6809) which then (outside US) in today's money would be the equivalent of a decent consumer notebook. The next was a consumer CP/M 2 & 3 machine because it was half the price of an AppleII and a lot of development software (free/shareware.) Later when the choice was between an Amiga and a PC/AT the price difference for the Amiga was even greater. At that time I was attracted to the Amiga hardware and software (tripos) design.
If at the beginning someone had given me an Alto my future might have been rather different (I had read and owned all the Parc SmallTalk books - Adele Goldberg(?).) Eventually ran Digitalk Smalltalk-V on a PC/XT and mucked about with XLisp and Scheme (and Forth and later Actor.) All eventually the impotent superwombles.
Concurrently I ran various Unix clones Minix 1.0, Coherent-3 etc and worked in the Unix Minicomputer bofhland where ethernet was starting to become the norm but serial (rs232) terminals were still the desktop standard. Even then it was pretty clear the network was "the thing" and the future.
I would not underestimate the evolutinary fitness of open source (and free ~ libre) software. The technological explosion from the 1990s into the last few decades has deep roots in that fertile soil. Even of the highest quality, all proprietary closed source software eventually dies and becomes extinct and not even belated open sourcing can normally rescue it.
As an afterthought ETH's Niklaus Wirth's Oberon system and other work probably also deserves an honourable mention here.