> This situation seems to me like the obverse of the output redirection
> choices.  If ">" opens the output each time, thereby wiping out what was
> already in the file; and ">>" appends data to the file - why not have an
> additional input redirection symbol, "<<" which would not reposition the
> pointer to the beginning of the file for each read but read the next
> available record?

The ">>" does open the file each time, it just opens it in append mode.  The
"<<" symbol already means something in the POSIX shell and I wouldn't want
it to have a different meaning in the CI, especially since the POSIX
functionality would be useful in the CI.

* To join/leave the list, search archives, change list settings, etc *
*     please visit http://raven.utc.edu/archives/hp3000-l.html       *