[Openmcl-devel] A personal outline of the history of Coral Common Lisp

[Bill St. Clair]

 Jeremy,

Sorry to hear that your memory is failing. Thank you for sharing what you
remember.

I remember rewriting the 6805 RAM for the Atari Star Wars Machine to
implement Turtle graphics for that screen, and writing 3D Turtle Graphics
for the Symbolics Lisp Machine. Both at the Atari Cambridge Research Lab,
in 1984.

I remember Jeremy giving me a 1 megabyte floppy, around 1988, plugging it
into my Mac Plus, and being amazed that there was a real Common Lisp that
ran on that little toy computer.

I remember expanding Gail’s CLOS beginnings to a full implementation, and
am sad that Gregor’s Art of the MetaObject Protocol book was not yet
available in 1990, or MCL and CCL would have a complete MOP today.

I remember helping the Dylan team with WOOD (William’s Object-Oriented
Database, a persistent heap), to store compiler products, and general MCL
maintenance. Our layoff party, in October of 2015, featuring Halloween
colors and snacks of Apple turnovers, sticks in my memory.

I remember helping, at Digitool, with the port of MCL from 68000 to
PowerPC, and writing the C code to implement segmented stacks. I think I
left Digitool just before the financial break-down, so, unlike Alice and
Gary, Hakim didn’t owe me any money.

Clozure provided interesting work for the last 14 years of my working life
(2006-2020), writing application software in CCL. Thank you to all, Jeremy,
Andrew, Gary, Gail, Gary Palter, Matt.

Bill St. Clair

On Jul 21, 2024 at 11:32:02 AM, Jeremy Jones <jaj at pobox.com> wrote:

> Disclaimer: This is an outline of the history of CCL only.  There is a lot
> of important history missing because I wanted to focus on CCL. I've also
> written more about the early years because it is less well known.
> TONS of details and people have been left out. I've left out descriptions
> of things like ObjectLisp and Qlogo, because I wanted to keep it short and
> there's Google.
>
> This is written from a failing memory, so please send me any corrections
> and suggestions.  I apologize in advance for leaving people out and getting
> things wrong. If anyone wants a more detailed history and/or wants to help,
> let me know. After feedback and corrections, I'll probably turn this into a
> Medium article.
>
> ‐--------
>
> A personal outline of the history of Coral Common Lisp
>
>
> I was hired by the Atari Cambridge Research Lab in 9/83 to write a
> compiler for Qlogo. They gave me a Symbolics 3600 and told me to model the
> compiler on Guy Steele's Rabbit compiler.
>
> The Atari lab closed in 4/84. I was one of three founders of Coral
> Software, along with Steve Hain and Glenn Forrester in 9/84. Glenn handled
> the business while Steve and I did the hacking.  Steve wrote the initial
> runtime and I ported the Qlogo compiler to C for the Macintosh.
>
> I was friends with Gary Byers while we both attended The Evergreen State
> College. I convinced him to become Coral's first employee in 1/85. A month
> or two later we hired Chris Fry and Gail Zacharias.
>
> The technical team consisted of myself,  Steve, Gary, Gail, and Fry. We
> focused for two and a half years to produce Object Logo and CCL 1.0. We all
> contributed to all parts of the system but Gary took over the compiler,
> Steve the GC and runtime,  Gail the editor Fred (Fred Resembles EMACS
> Deliberately) and the Common Lisp Reader (written largely in LAP, Lisp
> Assembler Programming!), Fry filled out Common Lisp functionality and did
> QA, I did the IDE and the integration with Mac toolbox.
>
> CCL would have totally failed if it wasn't for the non-technical people at
> Coral.  Glenn for raising money, buying the house we used as an office, and
> paying the bills.  Andrew Shalit for the excellent documentation.  Phillipe
> Krakowski for marketing. Susan and Adrian,  administrative assistants and
> others I've forgotten.  At its peak I think Coral was eleven people.
>
> CCL 1.0 was released in 6/87. It was a complete implementation of CLtL1
> and included an IDE and ObjectLisp. It ran on 512k Macs. (It was actually
> called Macintosh Allegro Common Lisp because of a misguided marketing deal
> with Franz.) It was an incredible product, acquired many fans, and won
> awards. It kicked ass.
>
> Apple acquired Coral in 12/88 and changed the name of CCL to Macintosh
> Common Lisp (MCL). Apple hired Bill St Clair and Alice Hartley to continue
> working on MCL. Gary, Bill, and Alice continued working on MCL while the
> rest of us worked on Dylan.
>
> Apple spun off MCL to Digitool in the early 90s. Gary, Bill, and Alice
> continued to work on MCL at Digitool. Digitool failed as a business and
> owed Gary, Bill, and Alice a lot of back salary. Gary made a deal with
> Digitool to open source the non IDE portions of MCL in return for forgiving
> the debt. This became OpenMCL.
>
> Gary continued working on OpenMCL for JPL and Sandia. OpenMCL was used on
> space probes! Ron Garrett has told this story.
>
> Gary, Gail, and I founded Clozure Associates in 2001. Clozure took over
> OpenMCL and renamed it to Clozure Common Lisp.  Lots of stuff happened to
> CCL in the 2000s including ports to Intel and ARM, an excellent Cocoa
> bridge, and an IDE for Cocoa. Matt Emerson joined Clozure and ported CCL to
> Intel32.
>
> In the 2010s, I left Clozure,  Gary and Gail retired. Clozure officially
> disbanded as a company in 2022 I think. I have a deep appreciation for Matt
> and the others who are keeping CCL alive.
>
> Finally I want to say that it has been an honor and a privilege to work
> with these amazing people for so many years. Thank you!
>
> Jeremy
>
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clozure.com/pipermail/openmcl-devel/attachments/20240721/e47ce765/attachment-0001.htm>