AUTHOR REBUTTAL

This seven-page paper discusses a novel way of implementing coroutines which are well-known constructs. The paper refers in its first sentence to two previous papers and devotes most of its one-page introduction to examining the reasons which led to restricting the discussion to languages that should be used for the class of tools for the simulation and implementation of computer operating systems.

The paper is meant to share the experience gained by the author in implementing such a tool with other designers/implementers. It does not mention that PL/1 or C have no multitasking facilities as the reviewer seems to think]

--John C. Cavouras, Glasgow, Scotland REVIEWER RESPONSE

Indeed, the author does discuss his reasons for not using built-in multitasking facilities of certain languages. One conclusion of his introduction is “. . . such languages (sic: PL/1, ALGOL 68, SIMULA) may not provide a flexible policy for the scheduling of parallel activities . . . it might be argued that the best choice for implementing a simulation-implementation tool is a general-purpose high-level language with no simulation or multitasking facilities.” A confusion arises, however, because his concluding (short) paragraph touts ADA, including its tasking features along with others, as a desirable language for such purposes. (This is the first mention of ADA in the paper.)