I was thinking about modeling business processes in a prototype, and I 
realized that there are several "levels of modeling risk" as I am calling 
them.  They reflect client familiarity with their own processes, and how 
much effort needs to go into prototyping in order to get clear and 
sufficient feedback from the client.  It strikes me that processes that 
have higher risk should be budgeted much more time during prototyping.

I was wondering if anybody here would be interested in a conversation about 
how they fit into prototyping and FLiP.  I would be interested in any 
philosophical and real-life observations.

1) Lowest Risk.  The business process is already modeled in an existing 
computer system that is being converted (to fusebox, for example).  The 
client is already familiar with computer screens and steps they need to 
take to get from A to B.  They can easily give meaningful and accurate 
advise to the programmer on exactly how they they think the new system 
should operate.  That does not mean that the old system did things the most 
efficient way.  You may very well want to convince the client to do things 
slightly differently.  However, it does mean that the client has a clear 
understanding of the process as it is implemented by the software, and 
won't need much education and guidance.

2) Medium Risk.  The business process is already modeled in non-software 
office processes and paper work, but is not modeled in any computerized 
system or software, or is partially computerized, in spreadsheets, for 
example.  The customer knows what data they want to get out of the system, 
and they often know what they want their reports to look like, but they are 
more at the mercy of the programmer when it comes to getting that data into 
the system in a user-friendly way.

3) Highest Risk.  The business process not directly modeled in the client's 
office environment.   The customer knows of some similar system that does 
what they want, or knows about some of the potential types of data that 
could be extracted from the system, but cannot give clear and immediate 
feedback on what they really want.  The client is absolutely at the mercy 
of the programmer and the prototyping process to produce something that is 
meaningful.  In my opinion, you should educate the client that these types 
of system will *almost always* require follow up releases of the software 
down the road, as since they can't know all the limitations or 
possibilities of the system until they actually start using it.  In other 
words, even with all the amazing things that prototyping can do, it may 
very well not solve all their problems.  Also, maybe role-playing with a 
paper or manual solution, simulating the data flow and output,  would be a 
way to do "real world prototyping" before any software-based prototyping is 
done.

Of course, actual software system are often contain a mixture of elements 
with different levels of modeling.  My main point is that the higher the 
risk, the more you need to make sure the client is intensively involved in 
the prototyping process.  Also, you will need to to budget or estimate more 
time/money up front for the higher risk items.







================================================
Douglas M. Smith - Application Architect
TeraTech - Tools for Programmers(tm) - www.TeraTech.com
[EMAIL PROTECTED], ICQ: 41044319
================================================

==^================================================================
This email was sent to: [email protected]

EASY UNSUBSCRIBE click here: http://topica.com/u/?bUrFMa.bV0Kx9
Or send an email to: [EMAIL PROTECTED]

T O P I C A -- Register now to manage your mail!
http://www.topica.com/partner/tag02/register
==^================================================================

Reply via email to