Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!henry From: henry@utzoo.UUCP (Henry Spencer) Newsgroups: net.unix-wizards Subject: Re: ls holy wars Message-ID: <5892@utzoo.UUCP> Date: Tue, 20-Aug-85 19:30:07 EDT Article-I.D.: utzoo.5892 Posted: Tue Aug 20 19:30:07 1985 Date-Received: Tue, 20-Aug-85 19:30:07 EDT References: <1303@utcsri.UUCP> <93600010@siemens.UUCP>, <834@ncoast.UUCP> Organization: U of Toronto Zoology Lines: 32 > Our TRS-80 Model 16 can NOT NOT NOT run a shell script ANYWHERE as fast as > a C program. ls | pr -t -5 takes FOREVER to run! ... > > Now before some well-meaning computer junkie tells me that we should get > Vaxen, please look at the price tag on yours. > > Now before you suggest piped stuff for ls or any other program where columnar > output is COMMONLY USED to a terminal, remember that if DEC went out of busi- > ness, the Unix community would live on through the 68000. The business world > cannot afford Vaxen! Repeat after me: "There is no such thing as a free lunch." We can't afford a VAX either. But our lousy little 11/44 runs shell scripts briskly. Partly because we have a high-performance shell; partly because we gritted our teeth, recited "you get what you pay for", and spent some money on good disk drives. Which are available for many, many 68000 boxes too. It really should come as no surprise that you have to bend your software out of shape if it has to run quickly on slow hardware. This sort of nasty pragmatic necessity should not be confused with fundamental principles of good software design. Actually, I have no objection to writing a C program that implements the equivalent of "ls | pr -t -5"; it is a fairly common wish. I'll probably do it here someday. But I won't call the result "ls", and it won't try to guess where its output is headed and alter the output on that basis. -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,linus,decvax}!utzoo!henry