home *** CD-ROM | disk | FTP | other *** search
- Path: solon.com!not-for-mail
- From: seebs@solutions.solon.com (Peter Seebach)
- Newsgroups: comp.lang.c
- Subject: Re: Interest in comments on the C language.
- Date: 23 Mar 1996 21:30:12 -0600
- Organization: Usenet Fact Police (Undercover)
- Message-ID: <4j2fk4$hr3@solutions.solon.com>
- References: <4inp40$kj2@ogre.cs.waikato.ac.nz> <4j1mbj$j5h@news.interpath.net>
- NNTP-Posting-Host: solutions.solon.com
-
- In article <4j1mbj$j5h@news.interpath.net>,
- Scott McMahan - Softbase Systems <softbase@mercury.interpath.com> wrote:
- >Oliver R Jones (orj@cs.waikato.ac.nz) wrote:
- >: 1: Is C inheriantly efficient (speed and code space wise)?
-
- >C is efficient on 32-bit UNIX machines. If you deviate from the
- >types of machines C was developed for, it becomes a matter of
- >how good the compiler is.
-
- C was designed and developed for 16-bit machines. C is elegant on a 64-bit
- system, but much poorly considered code will perform poorly in one way or
- another on such a system.
-
- >: 2: Is C inheriantly very readable and writable?
-
- >C is very writeable as-is, it takes extreme effort on the part of the
- >writer and reader to make it readable.
-
- I wouldn't call it extreme; I can produce what I believe to be legible
- code without significant effort.
-
- >C is as easy for non-coders to read as any other language.
- >Infer from that whatever you want to.
-
- :) This is not entirely true; there are languages which are harder for
- non-coders to get information out of. As a *definitively* non-coding
- child, I was able to read the source to `hack' quite comfortably, simply
- because it was written in terms of symbols I knew ("+", for instance)
- and I understood the logic it was implementing. I could read it well
- enough to decipher what the monster data would look like and edit that
- data with a binary editor, and I could read it well enough to predict
- the results of actions before trying them.
-
- >: 3: Is C a very simple language?
-
- >The core language is one of the simplest ever. It eliminates
- >almost every special case and exception to the rules other
- >languages have, and is very minimal. But, the library is
- >a monster because of this! It picks up all the slack from the
- >language! It has to provide all the stuff the language lacks.
-
- This is basically true. The Unix syscall interface is nearly
- simple and elegant, and Plan 9 is very close to having a
- programming interface that meshes well with C.
-
- You can write libraries for C which are much simpler and more
- elegant to the user than the standard one, and I frequently do.
-
- >: 4: Is it consistant with accepted mathematical notations?
-
- >No, that's APL.
-
- *smirk*
-
- >: 5: Why was C designed and for what purpose?
-
- >C was designed for implementing the UNIX operating system and
- >its utilities. Since then, its purpose has been to provide
- >a general purpose language for multiple platforms. Its purpose
- >became far beyond its original design!
-
- This is generally the case of elegant designs. C was designed to
- allow programmers to write the whole range of code, from low-level
- bit twiddling through abstractions and libraries, in a single
- language. It scales remarkably well to either end of the spectrum;
- I've seen tightly written C that compiles to a provably optimal
- sequence of instructions for the task at hand, and I've seen
- abstracted libraries that provide a simple interface to some truly
- hairy work.
-
- I would think that the purpose was more general than you would
- imply, but I'm in no position to prove it.
-
- This is the real reason I write C; I can do proper, modern SE, and
- if I need speed, I can tweak the piece of code that needs it, *without*
- having to switch environments.
-
- -s
- --
- Peter Seebach - seebs@solon.com - Copyright 1996 Peter Seebach.
- C/Unix wizard -- C/Unix questions? Send mail for help. No, really!
- FUCK the communications decency act. Goddamned government. [literally.]
- The *other* C FAQ - http://www.solon.com/~seebs/c/c-iaq.html
-