Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!seismo!mcvax!ukc!dcl-cs!bath63!pes
From: pes@bath63.UUCP
Newsgroups: comp.sys.atari.st
Subject: Documentation (or, NAG 2 followup)
Message-ID: <886@bath63.ux63.bath.ac.uk>
Date: Tue, 31-Mar-87 05:46:08 EST
Article-I.D.: bath63.886
Posted: Tue Mar 31 05:46:08 1987
Date-Received: Sat, 4-Apr-87 15:01:37 EST
Reply-To: pes@ux63.bath.ac.uk (Paul Smee)
Distribution: world
Organization: AUCC c/o University of Bath
Lines: 49

The following was meant to be for Richard Sansom, as a followup to
the NAG 2 questionnaire which I sent him.  However, it got bounced
by a mailer somewhere along the line.  Since I think it ought to
be brought to the attention of Atari, I've put it here.  (I've
already mentioned this to an Atari(UK) contact as well.)  I'd just
add the note from my original mailing to the effect that, for the
user population I'm thinking of, 'buy the developer's kit' is not
the answer.  (And, given the response I got when I contacted
Atari(UK) about it about a year ago, it would be like pulling teeth
to get them to sell you one anyway -- and with some fairly obnoxious
strings attached.)

                 **** Original message follows ****

If you got my NAG 2 answers, you'll have noticed my flame about
documentation.  While I mean every word of it, I thought a bit more
background might help put it into perspective.

The problem is really one of intense personal frustration.  I *like* the
ST.  I'd like to see more people using it.  Because of my position in
our Computer Center, people often ask me for advice when they are
contemplating buying a personal computer of some sort.  At present,
though, and primarily because of the lack of proper documentation, I can
only honestly recommend the ST to two classes of user:  the incredibly
naive ones, who want to buy a machine and a software package, load and
go, and never want to go beyond that; and the dedicated hackers, who are
willing to spend hours puzzling things out for themselves, and probably
even prefer it that way as a matter of pride.

However, most people I come into contact with are not in either of those
extreme groups.  They want to write their own programs (at least to some
extent), but they don't want to have to spend millenia trying to find
the hooks and handles which will allow them to make use of the support
which is potentially provided by the system.  I cannot in honesty or
good faith recommend the ST to anyone in this large category; and
everytime I have to suggest that someone would be better off with an
IBMclone or a BBC Master because they'll be able to find out how to use
it, it TICKS ME OFF, because I know the ST would be a better machine,
but there's no document I can recommend to them which would let them
make use of it.

I've seen most of (and bought many of) the 'third party' documents, and
so far they all seem either to be trivial, or (while having the right
idea) inaccurate (usually in the same ways -- must have sprung from a
common source), or incomplete.  (Or, even worse, some combination of the
three.)  (The new Sybex book on GEM *may* be an exception -- but as it
is about 'proper' GEM, and Atari GEM is a subset, I'm not sure yet.)

Cheers, Paul