I like the way it behaves. That makes it easy to create a duplicate
set of files by CRTDUPOBJing the physical followed by the logicals.
Well, I have found it handy myself, but I don't really think that kind
of ease is worth the conceptual load of implicit special cases. I
mean, GOTO is very handy, and it is even possible to use GOTO safely
and understandably, with enough skill and vigilance. But is it worth
it? Most people don't think so.
There are other ways to achieve copying of PFs and their LFs to other
libraries which, in my opinion, are not unreasonably
The only thing that makes sense to me would be to have an additional
parameter on the CRTDUPOBJ command to control the behavior. The
default would be to behave as it does now, with additional values to
force it to stay with the original PF and to force it to switch to a
PF in the new library. And/or maybe a way to explicitly specify the
I think having an additional parameter is a good idea. I obviously
don't like the current behavior being the default, but if it is an
available choice, then of course admins would be free to set the
command default to their liking on their own systems.
I also would not be opposed to having a completely separate CRTDUPLF
or the like.