From mboxrd@z Thu Jan  1 00:00:00 1970
From: "Brian L. Stuart" <blstuart@bellsouth.net>
To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net>
Date: Tue, 28 Oct 2008 14:51:34 +0000
Message-Id: <102820081451.27713.490726F6000519DE00006C4122230650629B0A02D2089B9A019C04040A0DBF9B9D0E9A9B9C040D@att.net>
In-Reply-To: <2261F45DDFF07CF7CB198B55@F74D39FA044AA309EAEA14B9>
Subject: Re: [9fans] non greedy regular expressions
Topicbox-Message-UUID: 29daf0b0-ead4-11e9-9d60-3106f5b1d025

> > This guy seems to blur the distinctions here.  His discussion
>
> He doesn't. If one reads the whole section part of which was quoted one
> will see that he clearly states DFA and NFA are theoretically equivalent,
> but then goes on to explain that DFA and NFA _implementations_ are not
> identical.

Actually, that's why I said "seems to."  Basically, what he
should be saying is that some people have found it easier
to add elements to the NFA model that augment the grammar.
But then he goes on to use the initials NFA when he really
means the extensions thereof.  This creates in the reader's
mind the misimpression that there's something magically
different between an NFA and a DFA, theoretically or
implementationally.  They have exactly the same expressive
power in both realms.  Besides, nondeterminism can only
be simulated when running on a deterministic computer.
That's actually the insight behind the NFA to DFA construction.
So talking about an "NFA implementation" is rather artificial.

> They learn slowly, hardly, painfully--they aren't smart. If
> possible they'll learn less rather than learn more. What the "hacker"
> denies the lowlife is the opportunity to exist free of "GNU-is-wrong" or
> "X-is-wrong" blemish.

Look at it this way.  The people here aren't trying to
create blemishes on anything.  Rather we are trying to
help your learning process.  We're trying to head off
the tendency for people to jump from "it was easy to
learn how to do X in this tool" to "how this tool does
things is better."  Instead, before asserting something
is better or asking for others to write software according
to a personal preference, it is important to broaden
your understanding so that you know what the options
really are and what the design decisions really imply.
Implementing a laundry list of features without that
perspective simply produces bad software.

> It's
> good--for the lowlife, of course--to know the wonders they see didn't
> spring into existence out of the blue.

That's why we teach the theory that everyone seems to
want to complain about.  Building on Newton, if we want
to see farther than those before us, we need to know
those before us well enough to climb onto their shoulders.
No software is based on Hogwarts' technology, and no
good software comes from the "random walk method of
programming."  It is the product of intellectual reasoning.
Criticism of the result won't be taken too seriously
if the critic shows they have not understood the reasoning
behind it.

BLS