# If the user is supposed to enter the file (path) of the file (or files)
# to be printed, how would you imagine the server would acquire the file
# data?
I'm presuming some analog to the capabilities of RFC 1867 ("Form-based
File Upload in HTML"), plus some special processing:
A to-be-printed form has a distinguished field, "print-file". If it's
a HTML form you'd write it
<INPUT TYPE=file ACCEPT=application/postscript NAME="print-file">
for example.
A smart print client ("IPP driver") would have to know about
"print-file", and OMIT the element that requests print-file, and
instead not actually submit the form until the print-file is ready.
A dumb print client would just submit it using their browser.
(ftp://ftp.parc.xerox.com/pub/masinter/file-upload-impl.txt).