Talk:Spice Lisp

Latest comment: 2 years ago by 74.111.96.172 in topic Relationship to CommonLisp


I infer from the existing description that it's a compiler, not an interpreter, but nowhere is this actually stated. Perhaps this fact should be added?—greenrd (talk) 22:02, 10 February 2008 (UTC)Reply

Lisp is, by definition, an interpreted programming language. Spice Lisp included a compiler that could reduce a lisp function definition to a sequence of low-level, stack-based instructions (much like Java byte codes). The 'eval' function—the heart of the Read-eval-print loop—could directly interpret lisp source code, and it would allow the underlying virtual machine to handle any pre-compiled functions that it encountered along the way.
This was not a distinguishing feature of Spice Lisp. The architecture was similar to other lisp implementations of its day (e.g., to Symbolics Lisp Machine lisp).71.199.121.113 (talk) 19:48, 29 June 2010 (UTC)Reply

The infobox says, "Host PDP-10, OS TOPS-10." That's not entirely accurate. A TOPS-10 system and Maclisp were used to bootstrap Spice Lisp development, but that platform was quickly abandoned once the PERQ implementation was able to stand on its own. The TOPS-10 version never was complete, and it never was intended to perform well. — Preceding unsigned comment added by 75.149.30.179 (talk) 22:16, 16 April 2019 (UTC)Reply

Relationship to CommonLisp

edit

Spice Lisp adhered to the CommonLisp specification before either it or the CommonLisp spec was released to the public. 75.149.30.179 (talk) 17:37, 10 September 2019 (UTC)Reply

Indeed! Guy L. Steele Jr., Author of the original "Common Lisp Reference Manual," sat in on the weekly meetings of the team that developed Spice Lisp while he was drafting the manual. The Spice Lisp team contributed somewhat to the manual, including feedback about what was or was not reasonable to expect from a practical implementation. 74.111.96.172 (talk) 19:56, 9 June 2022 (UTC)Reply