Rebuild a system the customer is satisfied with? Risk his process again?
Rather not.
Wim
----- Original Message -----
From: Steve Robertson <steven_j_robertson(a)hotmail.com>
To: <classiccmp(a)classiccmp.org>
Sent: Tuesday, December 18, 2001 2:07 AM
Subject: Re: "Geeks" and licensing
>
>The book by Brooks {The Mythical Man Month)should be mandatory reading
for
>every software man. It is fun too.
>I have very good memories from projects where you could first build a
>useful
>small part of the system. The client could then update his requirements
>and
>you could get all the bugs out and when all was stable you would build
the
>next part of the system. The client has a useful
system very early in the
>project and because you work together with the client (the user) in an
>early
>stage of the project, errors in the specification and the programs never
>last long. Cost control is also facilitated. You have a satisfied
customer
most of the
time during development and very much so in the end. This was
for projects for up to 1.000.000 lines of code.
IMHO, Steve McConnell's "Code Complete" should be required knowledge for
all
software engineers.
Prototyping is certainly a valuable tool for developing complex
applications. However, one must remember that the prototype is a means to
the end and not the final product. Once the application is defined, you
should throw away the prototype and build the system from scratch using
the
knowledge gained from the prototype. By doing that,
you'll have complete
requirements, a more efficient design, and happy customers.
In real life, it's pretty tough to convince management that you should
start
all over. The typical reaction is to put the prototype
in a box and sell
as
is :-(
SteveRob
_________________________________________________________________
Get your FREE download of MSN Explorer at
http://explorer.msn.com/intl.asp.