In the mid-to-late 50's, the Sage systems (AN/FSQ-7/8) had rooms full of real-time, graphic terminals (both vector and storage) with light guns and button panels, but no keyboards. I didn't work close enough with the CPU to know of any access beyond punched cards and register toggle switches. I don't remember any keyboard or tty access in the main computer room, although there were 024 card punches in other rooms that were on-line, in that they had a direct feed to the system. Mainly used as slow card readers, if I remember correctly. If there was something equivalent to a CommandLineInterface, I didn't see it.
On that basis, could one argue that GraphicalUserInterfaces pre-dated CommandLineInterfaces? --JimRussell
SAGE was part of the Whirlwind project, yes? See:
http://www.nap.edu/readingroom/books/far/ch4_b1.html
From this page: In software, they pioneered use of real-time software; concepts that later evolved into assemblers, compilers, and interpreters; software diagnosis programs; time-shared operating systems; structured program modules; table-driven software; and data description techniques.