|
In the late 1980’s at March AFB, there was a single DHCP (now called VistA) instance that supported BOTH VA and DOD functionality. That the user’s selection of their division set the DUZ(“AG”) has been part of the architecture all along.
-----Original Message-----
I was referring to the parameterization in the Kernel in that the sign-on division for the user over-rides the default devision for the system in the initial setting of the DUZ("AG") variable.
I agree that the way that DAOU implemented the changes is not a true parameterization. It was a local modification to support VOE and VOE only. It was done in a manner that was not consistent with the public statements made by VA personnel. But that statement could be considered inflammatory in that it presumes a certain definition of parameterization. I do not consider the DAOU changes to have been an implementation of parameters.
When would a single VistA implementation need to to accommodate multiple
agencies especially if those agencies may have slightly different file
structures and business rules? Now, if the FOIA release contained all the
M source code for the VA and FOIA-VA and IHS and OTHER and possibly CHCS
AND all the underlying M code was appropriately parameterized, then I could see
this. But until then I would think that a single VistA implementation
must only perform according to one "agency". On 1/18/06, Cameron Schlehuber <[EMAIL PROTECTED]> wrote: The "additional parameterization" is dependent on the enterprises represented in the Agency Code field in the Kernel System Parameters file. The kind of behavior is also dependent on sometimes needing to accommodate multiple agencies default behavior as well as being able to "switch" on or off some behaviors. E.g. while the default behavior for VA is to use an MPI, the default for IHS and VOE is to not do so. However, should an enterprise (like IHS) eventually implement an MPI, that behavior will need to be ready to be switched "ON".
-----Original
Message-----
What additional parameterization of this variable is needed over and above what the Kernel already provides? On 1/18/06, James Gray < [EMAIL PROTECTED]> wrote:
|
