2008 Feb 06 8:13 AM
Hi,
For a new ECC implementation, what is the best strategy or approach to use in implementing roles and authorization? Is there a blueprint that I can use?
FYI, we are not implementing Virsa.
Your response is highly appreciated.
One thing to remember
Functional consultants MUST write a process design, this should be at TRX level and contain the needed restrictions from fucntional point of view.
The aforementioned should be the basis of the role design.
2008 Feb 06 8:24 AM
>
> Is there a blueprint that I can use?
I doubt you'll ever find (a usable) one for free or on such a forum for that matter. Most of us earn (part of) our money designing them tailored to the customers' needs. I for one am surely not going to give examples away.
The best strategy will always be:
1- determine the needs in the company (what are the tasks for various people and which resources do they need to achieve their goals).
2- determine which data has to be secured.
3- draw a concept based on above information and have it validated by the business.
4- design taskroles (singles) per task and functionroles (composites) to group tasks into functions.
5- test both tasks and functions. The first test can be part of a unit test while the second one will be like an integration test.
People you need:
Functional consultants per module. They know about module-specific authorization stuff.
Business consultants and/or key users who know which processes there are and how they're divided over the various jobs/functions in the company
The (internal) auditors to tell you which information needs to be secured.
As you see this is not a one person job and the outcome will differ per company.
Jurjen
2008 Feb 06 5:18 PM
2008 Feb 06 6:55 PM
One thing to remember
Functional consultants MUST write a process design, this should be at TRX level and contain the needed restrictions from fucntional point of view.
The aforementioned should be the basis of the role design.
| User | Count |
|---|---|
| 6 | |
| 2 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 |