I was running the MIT-EECS department computing machinery back when Symbolics and LMI (Lisp Machines, Inc--Greenblatt's competitor) were just getting going, and we had around 8 or 10 of the original LispMs for undergrads to use, wire-wrapped and all. Even those, as slow as they were, were impressive machines for the time, given the 'turtles all the way' down architecture. (Everything written in Lisp, except the microcode.)
Back then, we were using MIT's CHAOSnet, a home-grown competitor to PARC's original 3Mb Ethernet. Unhappy memories of big fat yellow coax everywhere, easily dented. We used a TDR a lot to find kinks in the cable when the network wasn't working well.
The LispMs were years ahead of their time in terms of sheer architectural genius, but were also incredibly baroque--many brilliant hackers adding ornament upon ornament for years...
Comments
Boy, that brought back happy memories!
I was running the MIT-EECS department computing machinery back when Symbolics and LMI (Lisp Machines, Inc--Greenblatt's competitor) were just getting going, and we had around 8 or 10 of the original LispMs for undergrads to use, wire-wrapped and all. Even those, as slow as they were, were impressive machines for the time, given the 'turtles all the way' down architecture. (Everything written in Lisp, except the microcode.)
Back then, we were using MIT's CHAOSnet, a home-grown competitor to PARC's original 3Mb Ethernet. Unhappy memories of big fat yellow coax everywhere, easily dented. We used a TDR a lot to find kinks in the cable when the network wasn't working well.
The LispMs were years ahead of their time in terms of sheer architectural genius, but were also incredibly baroque--many brilliant hackers adding ornament upon ornament for years...