Path: g2news1.google.com!news2.google.com!postnews.google.com!f17g2000prh.googlegroups.com!not-for-mail
From: Hugh Aguilar <hughaguila...@yahoo.com>
Newsgroups: comp.lang.forth
Subject: Re: New Forth Website
Date: Wed, 10 Feb 2010 15:12:03 -0800 (PST)
Organization: http://groups.google.com
Lines: 68
Message-ID: <753dd863-5bfe-45e2-983b-775b7200b9f5@f17g2000prh.googlegroups.com>
References: <ly%an.81018$BV.28816@newsfe07.iad> <4fcd0237-e217-42bc-85e4-370bbf272ba2@o3g2000yqb.googlegroups.com> 
	<43f938fe-7cb2-4e15-84dd-b2596206c91e@s25g2000prd.googlegroups.com> 
	<28fc0612-457d-4d04-be9f-e34975787cea@f15g2000yqe.googlegroups.com>
NNTP-Posting-Host: 63.238.70.158
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
X-Trace: posting.google.com 1265843523 9097 127.0.0.1 (10 Feb 2010 23:12:03 GMT)
X-Complaints-To: groups-abuse@google.com
NNTP-Posting-Date: Wed, 10 Feb 2010 23:12:03 +0000 (UTC)
Complaints-To: groups-abuse@google.com
Injection-Info: f17g2000prh.googlegroups.com; posting-host=63.238.70.158; 
	posting-account=hP9USgoAAADyPwMdR5qi1rinwP694_5o
User-Agent: G2/1.0
X-HTTP-UserAgent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.0.15) 
	Gecko/2009102815 Ubuntu/9.04 (jaunty) Firefox/3.0.15,gzip(gfe),gzip(gfe)

On Feb 9, 6:53=A0pm, John Passaniti <john.passan...@gmail.com> wrote:
> For professionals-- you know, people who actually derive a living off
> creating software-- programming is a relatively small and largely
> mechanical part of what we do. =A0If you're doing this as a hobby, then
> yeah, programming is what you'll focus on because programming is fun.

As a rule, I never respond to Passaniti posts --- Passaniti has been
on my twit list ever since he said that my symtab program "sucks." I
will make an exception in this case only for the purpose of correcting
a misstatement. I'm not a hobbyist, I have several years of
professional experience in Forth, various assembly-languages and C/C+
+. When I was employed at Testra, I wrote the MFX cross-compiler for
their MiniForth chip.

> Turns out that knowing the problem you're trying to solve (that
> is, requirements analysis and specification) is key to not wasting
> your time coding functionality you don't need, or having to go back
> and rework code that doesn't meet requirements. =A0Turns out that
> thinking about architecture and design can give a significant boost in
> effectiveness by factoring before you even start writing code, and
> knowing how you're going to organize it. =A0In short, *thinking* about
> the larger act of developing software verses just one part (writing
> code) makes all the difference.

In my work experience, I have found that there are often coworkers who
seem to never actually write any code, but who want to endlessly hold
meetings discussing specifications and design. Forth is actually all
about bottom-up programming, and this worthy idea has now been adopted
by the Agile movement, but these folks seem to be stuck on the
Waterfall paradigm in which everything has to be figured out in minute
detail before a single line of code can be written. These people often
use the word "team" as a verb, and they believe that they are
important team members. They also believe that programming is a purely
mechanical task devoid of any creative aspect. Any programming that
they do, which isn't much, will rely entirely upon cut-and-paste of
existing source-code. They never write code from scratch, although
they are quick to criticize other people's code for being
"unidiomatic," which essentially means that they can't figure out how
it works. I call these people "pilot fish."

My slide-rule program is, IMHO, very readable. It is possible for a
person to put together a custom slide-rule using only cut-and-paste
programming, and without actually knowing any Forth at all. This can
be as simple as taking one of my slide-rules and swapping out the
functions that generate the scales, in order to get a custom layout. A
step up would be to modify the existing scales by labeling particular
marks, or by folding them on a particular mark. There are plenty of
examples to imitate. Another step up would be to write new scales
supporting mathematical functions of your own design. Even this could
be done with only a modicum of Forth knowledge, just by imitating the
dozens scales that I have already written.

There is a place in the world for people with Passaniti's level of
programming ability. I support people like this by making my software
easy enough to work with that it can be upgraded using cut-and-paste
programming. People have to be a lot nicer than Passaniti to get my
support though. These multi-page diatribes of John's just earn him a
place on my twit list. I won't support him.

For any further discussion of computer programming, I ask that
programmers move over to www.forthwiki.com and start threads in those
forums. Note that I define the term "programmers" as: people who write
programs. For the most part, I have stopped posting messages on C.L.F.
--- if you want to respond to any of my posts, or discuss Forth
programming, you will have to move over to Alfred's site. It is a
moderated site, which means that trolls will just get their posts
deleted. You have to focus on Forth programming. Have any of you ever
tried that?