Path: utzoo!attcan!uunet!aplcen!uakari.primate.wisc.edu!zaphod.mps.ohio-state.edu!lavaca.uh.edu!menudo.uh.edu!sugar!ficc!karl
From: karl@ficc.ferranti.com (Karl Lehenbauer)
Newsgroups: comp.windows.ms
Subject: Re: Windows 3 criticism
Message-ID: <6I56PF3@xds2.ferranti.com>
Date: 3 Oct 90 02:03:18 GMT
References: <4297@rex.cs.tulane.edu> <57914@microsoft.UUCP>
Reply-To: karl@ficc.ferranti.com (Karl Lehenbauer)
Organization: Ferranti International Controls Corporation
Lines: 56

In article <57914@microsoft.UUCP> sunni@microsoft.UUCP (Sunni ROGERS) writes:
>I think this guy suffers from the "Real Men don't use Menus" complex.

>You are limited at the C:> to how many parameters (documented or not)
>you can remember. At least in Windows it's self explanatory.

I am completely sold that you come up to speed faster in a icon/menu/mouse 
graphical environment than on a command interpreter.  And for programs you
don't use very often, they're a lot easier to drive around.

However, there are things you can do from the Unix shell, for example, that I 
have yet to see anything like under a windowing environment, other than in a
shell within a window... powerful things like simple programs that manipulate
data in standard ways being hooked together with pipes.

I wrote a shell script (in Tcl) the other day that recursively descended our 
comp.binaries.pc archives.  For each package, it ran through all the parts
in name-sorted order, decompressing each if necessary, ran each through
sed(1) to strip off the headers and trailers, catted the parts together and
piped the result to uudecode to create the original .zoo archive, then
cleaned up after itself.  Once I got it right, I let it run through the
*hundreds* of packages in the archive.  Plus, when I want to do it again
later, I have the script.

In a windowing environment, you'd have to double click on your compressor,
your uudecoder, your editor, etc, and do it by hand, click-click, moving
around from program to program and reentering data like filenames and such
to each program.  You could record a macro, but that's a kludge, really,
and besides, there are a lot of if-type decisions to be made, like is this
file compressed or not, and so forth.

Hear me out, I do not think there is an inherent, permanent dichotomy between
these methods.  The Dynamic Data Exchange (DDE) holds forth a lot of promise
for communications between programs.  AREXX has done well on the Amiga as
a programming language and interprocess-communications mechanism for 
communicating between tools.

But for the most part, there are not yet efficient windoing equivalents to 
many of the things we are used to doing, like shell scripts and, dare I
say it, programming.  Once again, I find the edit-make-test-edit-make-test
process much easier from a command interpreter than from the file manager
or program manager.

Suit yourself.  With Windows it's nice that we can have both, plus I expect
to see, indeed we are already seeing, an evolution toward windowing programs
that can be programmed and can communicate with one another.

>I think maybe you are just intimidated by the fact that there
>will be people who can use computers and won't have to have you there
>telling them what to type in and thinkg that you're a genius because
>you can remeber every damn backslash, parameter and directory name.

It really wasn't necessary to guess at the guys motivations and then
ridicule him for it, I don't think.

...and those ! * ^ $ [ ] characters can do some pretty amazing things.
-- 
-- uunet!ficc!karl (wk), uunet!sugar!karl (hm)
"The computer programmer is a creator of universes for which he alone is 
 responsible.  Universes of virtually unlimited complexity can be created in 
 the form of computer programs."  -- Joseph Weizenbaum