[2] The idea arose from the difficulty of programming the IBM SSEC machine when Backus was hired to calculate astronomical positions in early 1950.
Attendees likewise complained with issues with "scaling", or the need to religiously track the decimal point in arithmetic operations.
[5] John W. Sheldon, a supervisor of IBM's Technical Computing Bureau attending the meeting, and others felt that an "interpretive" programming system that utilized floating point operations was the best solution to this problem.
Sheldon asked John Backus, who had previously worked on a CPC to SSEC code translator, to supervise the creation of a new floating-point interpretive programming language for use internal to IBM.
Starting in 1953, Backus and five colleagues designed this new language and named it "Speedcoding", where its use soon spread outside of IBM to customer installations of the 701 system.