Path: utzoo!utgpu!watmath!att!mtuxo!mtgzz!drutx!druhi!terrell
From: terrell@druhi.ATT.COM (TerrellE)
Newsgroups: comp.sys.atari.st
Subject: Re: Sell or PD -- some thoughts.
Message-ID: <4432@druhi.ATT.COM>
Date: 24 Jul 89 15:06:35 GMT
References: <15273@watdragon.waterloo.edu>
Organization: AT&T, Denver, CO
Lines: 56

In article <15273@watdragon.waterloo.edu>, achowe@tiger.waterloo.edu (CrackerJack) writes:
> 
> I'm also currently in the process of writing a game (actually on hold
> do to summer job as a programmer) but I have an answer to your question
> about selling and PD (not necessarily the right one).
> 
>    b) My last point is the idea of releasing early versions into the PD
> 	  to:
>   
>       i) sample the market
>       ii) beta test software
>       iii) cheap promotion and advertising
> 
>       From here you can decide whether you would like to become an
>       independent, sell your product to a companies, or just carry
>       on releasing into the PD updates because its more a fun casual
>       past time than a money making venture.
> 

Alternative (b) is a tempting one - beta testing is EXPENSIVE.  When I
beta tested a program that I developed, I had to involve testers all across
the country (none of them were local).  That meant a hefty phone bill, and
expensive postage.  But it was worth it:  the "world" didn't see the problems
with the program that were fixed through the testing effort.  Furthermore
I don't lose sales to customers who are satisfied with a buggy public
domain version, and have little motivation to go through an expensive
upgrade.

I'm surprised that you suggest a preliminary PD release for beta-testing
when you mention elsewhere in the article how important your reputation
as a software developer is.  Making the public your beta test crew can
be hazerdous to your reputation.

I used alternative (b) once to do beta testing on a PD program that I had
no plans of "commercializing".  The program was a spelling correction program,
and one user discovered a bug that would effectively delete big portions
of the file being corrected!!!  One the one hand I bug was reported (and fixed)
that might have gone undetected with a smaller test crew, but also an 
embarrasing bug was reported worldwide on the net...  

Also, with a formal beta-test arrangement, your testers are more aware of
your expectations.  They will go out of their way to test things that
ordinary users wouldn't bother with.  My advice:  if you have plans of
developing a commercial application, choose a beta test team (~7 people),
and pay the price.  If the testing is properly managed, your investment
will be returned tenfold!  If you have no plans of commercializing a PD
program, then try a testing strategy that is less expensive...

I did follow alternative (b) more or less by accident.  A PD application 
kept growing and maturing.  Eventually I stopped issuing PD updates, and
a year later released a commercial version of the product...  NOW I wish
that I could get the names and addresses of all of the PD users so that
I could offer an upgrade...  So if you can, find some way to inexpensively
get the names and addresses of your PD users.  This information can be
very useful.  If one of the PD software houses sells your product, I suppose
you might be able to get the names and addresses from them...