The role of users

[Mike or Penny Novack]

Paula Hendricks wrote:

>i wasn't going to jump into this conversation, but i think i will. i am a "user" and a non-techie. there is no way i can "fix" anything, but i am also a great beta tester... so if help is needed along the way for this, let me know.
>  
>
No, I wasn't trying to flame.

THIS is the sort of response I was trying to elicit (beta testing 
requires USERS).

I have worn all the hats (at various times) in the software development 
process. What folks who have never done this before might not realize is 
that it is important to always keep in mind which hat on at the moment.

During the "requirements" phase you DON'T want the people doing that to 
be those who will eventually be doing the coding unless there is no 
alternative. It's not for the coders decide what the program SHOULD be 
doing. That's up to other folks (wearing user hats) and then up to the 
code designers to make it so and then for the coders to implement that 
design.

Users generally need help with their part of the process. They usually 
know what they want in broad general terms but not the specifics and 
those details must be specified too ("if the user forgot to enter some 
required field, what happens next?"). That's the job of an analyst 
assisting the users come up with the formal definition of what the 
program is supposed to do. Not be telling them but by asking them. In 
that example I gave, the users can probably say what they want to happen 
in that case once they are asked but they won't imagine "exception 
conditions" until asked.

Understand? If I were to design/code almost ANY of the user requests I 
have seen I guarantee the results would not satisfy the user making the 
request. Not because I am a bad designer/coder (far from it)  but 
because I'm not a mind reader. We need USER TEAMS as part of the 
process, people who will volunteer to take a vague request and turn that 
into a formal set of requirements.

Michael