Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!apple!sun-barr!newstop!texsun!letni!dms3b1!caleb!jdp
From: jdp@caleb.UUCP (Jim Pritchett)
Newsgroups: comp.sys.amiga
Subject: Re: Spreadsheet recommendations
Message-ID: <9711.AA9711@caleb>
Date: 3 Jun 90 02:00:46 GMT
References: <136468@sun.Eng.Sun.COM>
Followup-To: comp.sys.amiga
Organization: is sometimes desirable
Lines: 56
Keywords:
Distribution:

[ ]

In the above referenced article, Chuck McManis writes:

> I wish to God that someone would take this
> "prototype" for a spreadsheet and give it to a professional developer
> and say "Here is the prototype, why don't you write a commercial grade
> version for us."

I have had several MaxiPlan versions over the past few years.  I like
MaxiPlan, but at least up to my version (1.9 something - NOT the latest)
it still has some obvious and somewhat annoying bugs (especially until
you figure out how to avoid them - the major thing to remember is scroll
as little and as slowly as possible!)  I use it regularly, however.

My observation about MaxiPlan is really about it's author (correct me
if I err here.)  I believe that he is the same person who wrote MaxiComm
before MaxiPlan.  Recently, I believe that he produced Ultracard.  The
common thread in all of these programs, is that the customers are really
the debuggers for the programs.  All were released in an extremely (IMHO)
poor state.  The subsequent releases (many were needed!) usually improved
things some and added new features (and sometimes new bugs.)  Each release
usually cost a significant amount to "upgrade."  After several revisions,
the products seem to end up useful, but at a heavy cost to the consumer
in time, aggravation, and cost.  As always, caveat emptor.  Since I don't
know the author personally, I can't really evaluate why these trends
appear.  It could be simply greed (it costs more in the short term to
produce a solid product.)  It is more likely that he simply lacks the
"stamina" to finish debugging a program.  Too many programmers today
lose interest after the program "mostly works."  It takes quite a bit of
determination to finish the job after this point.  This is one of the
reasons why having a good "outside" set of Beta testers is extremely
important.  I wish that more of the Amiga developers would use outside
Beta testers to help them debug their products.  (It always amazes the
programmers when the users actually try to do something that is documented
to work, but is a little different from what the programmer had in mind.
The usual result is a bug discovered.  The usual programmer reaction is
"But you shouldn't do that!"  The correct reaction is "I'll fix it right
away.")

[ Note:  in case you haven't figured it out yet, I work on a large (non-Amiga)
software project.  I see the above problems daily.  ]

... I am now leaving SoapBox mode.  If you are lucky (or your kill file is
set appropriately), you may not have to read another such tirade from me
for a while.



--

                                                Jim Pritchett


UUCP:  {attctc|texbell}!letni!dms3b1!caleb!jdp
 or    texbell!rwsys!caleb!jdp