From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID autolearn=no autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 103376,d3432b5ee2e7ebcf X-Google-Attributes: gid103376,public From: bobduff@world.std.com (Robert A Duff) Subject: Re: ****************** Ada vs C++ Help ******************** Date: 1996/09/03 Message-ID: #1/1 X-Deja-AN: 178196801 references: <3226aa7e.7480949@news.redstone.army.mil> <508a18$1af@mtinsc01-mgt.ops.worldnet.att.net> organization: The World Public Access UNIX, Brookline, MA newsgroups: comp.lang.ada Date: 1996-09-03T00:00:00+00:00 List-Id: In article <508a18$1af@mtinsc01-mgt.ops.worldnet.att.net>, John M. Miano wrote: >1. Code size and execution speed are a function of the compiler and not the >language. Partly true. However, it's not very useful to someone choosing a language, to tell them "it is theoretically possible to implement so-and-so language efficiently". Existing implementations are all that matter, and anything else is vaporware. Furthermore, language design decisions can have a strong effect on whether it is feasible, or even possible, to generate efficient code. For example, a dispatching function call can be done in Ada or C++ by calling indirect through a table of pointers to functions. It is *possible* to do the same in Smalltalk in many (perhaps most) cases, but it's not *feasible*, since it requires global analysis of the whole program at link time, which is takes too long for large programs. So most implementations of Smalltalk use a rather slow method lookup mechanism. I think it's fair to blame that on the language, and not on the implementations. - Bob