The lost cause of the Lisp machines

[Paolo Amoroso]

Tim Bradshaw published this blog post that's especially interesting to me as I never used Lisp hardware (except the Cardputer 😀): The lost cause of the Lisp machines.

--

Paolo Amoroso | Blog | Mastodon

[Tim Daly]

As someone who suffered for 30 years from GAS (gear acquisition syndrom) for music

I bought a full set of recording studio gear, studio reference speakers, a few of almost

every instrument ever made, and a 3ft by 6ft book of music books. So I'm aware of GAS.

On the other hand I started as a lisper in 1971 and have used it ever since. I reviewed the

Common Lisp definitions and even have my name enshrined in the archives. I have

written programs in almost every lisp ever released. Bill Schelter (AKCL) worked in my

office every time he came to visit at IBM Research. I have contributed patches to the

garbage collector and tail recursion. So I'm pretty sure I have enough background to comment.

I had a Symbolics lisp machine in my office for about 5 years. I developed an Expert System

product, a Computer Algebra product, and researched a "Design-to-Build" system for 

planning automated robot assembly algorithms from 3D solid drawings.

I LOVED that machine. You could hit a bug, travel down the stack, patch the code live,

rewind the stack and continue. Code that depended on the change was automatically

recompiled.

Lisp isn't a programming language. It is a way of thinking. The Symbolics machine was

a mind-expression machine. I know it sounds "over the top" but it isn't. Just like lisp is

an "epiphany event" (you don't get it until you "get it"), the Symbolics machine has the

same character. If you haven't lived with one you won't "get it".

His criticisms, while rationally valid, miss the fundamental point.

Tim Daly

--
You received this message because you are subscribed to the Google Groups "Medley Interlisp core" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lispcore+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/lispcore/CAGi1hztjgQkT2KbRAH5dr3Je8ss31RS%3DdnEUs_E9BqE93oe%3DdA%40mail.gmail.com.

[pixel...@gmail.com]

My 2 cents:

I think what truly resonates with a lot of Lisp folks is the notion of a Lisp first OS.
So much so that it's a goal to implement such a thing in FPGAs for many hobby folks.

In many modern circles Lisp is wrongly thought to be totally inappropriate for building "systems" in.
It's usually just assumed to be as slow as Python, only interpreted, and have weak typing, by people who never cross the Algol family lines.

Right now as somebody roughly involved in the current landscape I have to endure endless arguments for C, C++, Rust, Zig, Go , Odin, Jai etc being the ONLY things you should be writing "systems" (whatever that means this week) in.
They are so worried about speed that they won't consider anything else.
Having Lisp machines as part of history is a nice tool to bonk em with so they settle down. (Even if these systems were slower back in the day)

I think a lot of wrong assumptions are made about Lisp machines in general - I can understand his frustration.
It's not truly "lisp machines" that captivate people, it's the notion of Lisp being first class that captivates people.
Right now we're back to the Stone Age with compile/run languages and shell languages that are inconsistent and deplorable.

[Paolo Amoroso]

There's now also a Hacker News thread.