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.lang.c Subject: Re: How idiot-proof must CPP be? Message-ID: <4428@utzoo.UUCP> Date: Sat, 6-Oct-84 18:53:08 EDT Article-I.D.: utzoo.4428 Posted: Sat Oct 6 18:53:08 1984 Date-Received: Sat, 6-Oct-84 18:53:08 EDT References: <9225@watmath.UUCP> <522@wjh12.UUCP>, <9297@watmath.UUCP> Organization: U of Toronto Zoology Lines: 13 > What more justification do you need than a good use for it? Basically, it should be useful enough, to enough people, to justify forcing all C implementors to provide it. It also shouldn't break things, and it should be clean and simple enough that its prospective customers can understand it. Finally, it should be sufficiently well understood (this normally implies substantial experience with one or more experimental implementations) that its consequences and side effects are known. Implementing something just because it looks useful is the way you get kitchen-sink systems like PL/I and OS/360. -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,linus,decvax}!utzoo!henry