Some Non-standard Issues on Lisp Standardization

Lisp was born about 25 years ago as an A1 language with a precise operational semantics. Since then many Lisp dialects have been proposed, implemented and used. In 1960's Lisp 1.5 was a kind of Lisp standard, although there were many Lisp 1.5 dialects which depend on 1/0 and computer systems. In 1970's various Lisp dialects were spawned to respond to the need of more powerful Lisp systems for A1 research and symbolic computation. Among them we know that Lisp 1.6, Interlisp and Maclisp had big influences in the development of Lisp; especially, Maclisp brought us various interesting successors, including Franzlisp, Scheme, Zet alisp and Common Lisp. Common Lisp [STEELE] may be taken as a result of standardization activities of Maclisp and its successors, and Scheme is one of the first steps to try to settle and resolve some syntactic and semantic incompatibilities in Maclisp families. Eulisp [PADGET] may be taken to be another attempt along with this line with greater ambition for Lisp standardization, employing the design philosophy of extensible languages based on "leveling of languages". Various design and standardization issues have been considered and examined in the course of designing Common Lisp and Eulisp. The current standardization activities in US and Europe are mainly placed on "clean-upn of Common Lisp and "refinement and development" of Eulisp, in addition to efforts of designing object-oriented features in Lisp systems. In a sense major (standard) issues in Lisp standardization have been considered in these efforts of clean-up of Common Lisp and design of Eulisp. However, we think that there are some other important issues which may be called "nonstandard issues" according to the current state of arts of Lisp systems.