Path: utzoo!attcan!uunet!cs.utexas.edu!evax!umbaugh From: umbaugh@evax.arl.utexas.edu (David Umbaugh) Newsgroups: comp.os.minix Subject: Re: Efficiency vs Effectiveness Message-ID: <1990Jun11.154831.558@evax.arl.utexas.edu> Date: 11 Jun 90 15:48:31 GMT References: <21640@nigel.udel.EDU> Reply-To: umbaugh@evax.arl.utexas.edu Organization: UT-Arlington-CSE Lines: 27 In-Reply-To: V2057A%TEMPLEVM.BITNET@cornellc.cit.cornell.edu's message of 11 Jun 90 04:07:11 GMT On 11 Jun 90 04:07:11 GMT, V2057A%TEMPLEVM.BITNET@cornellc.cit.cornell.edu (Juan Jose Noyles) said:Juan> Sender: mmdf@udel.EDU [Lots of blabber about effectiveness, efficiency and kernel omitted.] Juan> just an old junior grade hack. They're the only ones that claim that it's okay Juan> to waste code because the machine's more powerful, or bitch because it's not. I think it was that old junior grade hack, Harlan Mills, who said, "allways use the highest level language that will do the job, even if its JCL." [Quoted from memory, certainly inaccuratly.] Then there was that "Programming Pearls" column in CACM in which Jon Bentley suggested that one might sort a file using a one line shell command rather than to write a sort routine optomized for the specifics of the problem at hand. What a horrible waste of machine cycles. Execution efficiency is only one concern in program design. Often, it is one of the less important considerations. -- L. David (Dave) Umbaugh Computer Science Engineering University of Texas at Arlington BITNET PO Box 19015 Arlington, TX 76012