Jürgen,
I submitted this patch in a slightly different form once before; it
didn't make it into GNU APL.
After having lived with the patch for a while on my own system, I've
updated it to the current form.
The gist of the patch is to acknowledge redefinition of a ]usercmd with
*identical* arguments as being successful rather than complaining about
a "BAD COMMAND". The patch does not change the effect of the command,
only the report.
With the patch, GNU APL will report
"User-defined command ]... installed"
each time a command is defined with identical arguments.
I think this is a better response than complaining
"BAD COMMAND"
in the same circumstance, given that the behavior actually succeeded.
I hope that you'll apply this patch to GNU APL.
Best wishes,
David