Zsh Mailing List Archive
Messages sorted by: Reverse Date, Date, Thread, Author

Re: o'reilly zsh book?

On Mon, Jan 25, 1999 at 09:32:22AM -0800, Bart Schaefer wrote:
> The main reason I've been dragging my feet about making a proposal is that
> I can't decide what really ought to be in the book.  Basic shell usage and
> the important aspects of scripting are already covered in agonizing detail
> by ORA's bash and ksh books; aside from a few oddities like "emulate", it
> just isn't that much different to do those things in zsh.  (Intentionally
> similar, in fact.)
> So the book would have to be largely about the interactive zsh experience,
> and would have to progress pretty quickly into the real esoterica.  That'd
> certainly be fun to write, but isn't the kind of thing that sells to a big
> audience; even Bob's emacs book isn't doing spectacular business, and there
> are a lot more emacs users than zsh users.
	i can see lots of space in between what's discussed in the ksh book
(don't have the bash book, but the one time i paged through it, it seemed
mostly the same) and a book of pure esoterica.  i've spent the last month
or so, for example, writing ksh scripts to do all sorts of neat post-install
tasks on a bunch of sun machines i'm going to be bringing up soon, and the
entire time, i was thinking to myself that it would have been so much easier
and prettier to write them in zsh.  (since one of the last ones to run was
to install zsh, however, that was not very feasible.)  examples of how to
do things like that would, i think, be the bread-and-butter of a book on
zsh--like bruce suggested, what zsh needs is more examples of things that
your average "power" user (if that isn't an oxymoron) might want to do,
if only they knew that it were possible.
	another way of looking at it is how i often explain why i use zsh:
zsh is to ksh what perl is to sed/awk--in many ways, the same thing, but
with so many new features and additions that at times you wonder if there
really is any connection between the two.  even though the sed/awk book
and ora's mastering regexp books cover much of the "core" functionality
of perl, there's so much else out there that there's now five perl books
from ora alone.
	actually, now that i think about it, maybe what i'm thinking about
is less like "learning ksh", and more like the perl cookbook that gnat and
tom just put out--as much a "shell scripting for power users" as an intro
to zsh, though it would, in the process of showing all of the cool shell
tricks, provide a good description of zsh qua zsh as well.

> An additional detail is that 3.1.x is a fast-moving target at the moment.
> Several major new features have appeared since roughly early November, and
> are still mutating towards their final form.  Until things stabalize a bit
> again, a book about zsh esoterica would practically be obsolete before it
> was even finished.
	this was another concern that i had, especially given the apparent
paradigm of "ksh-compatibility over backawrds compat with earlier versions
of zsh".  still, i think that by the time a book were to really get under
way, things might have settled down some, or at least a better idea of
that "final form" might be available.

> Nevertheless, it could be a worthwhile thing to talk about.  Bob's offered
> to help me put the proposal together if I work out the details.  Anybody
> have any ideas how we might estimate the size of the audience?
	in light of the recent tcp_wrappers trojan, i suggest an undocumented 
block of code that stores the output of uname -a and whoami, and mails
a digest copy to the dev list once a month.  :)

	-- sweth.

Sweth Chandramouli
IS Coordinator, The George Washington University
<sweth@xxxxxxx> / (202) 994 - 8521 (V) / (202) 994 - 0458 (F)
<a href="http://astaroth.nit.gwu.edu/~sweth/disc.html";>*</a>

Messages sorted by: Reverse Date, Date, Thread, Author