Grace Hopper

research.opt.indiana.edu

Admiral Grace Murray Hopper (1906 - 1992)

A good, short biography: www.sdsc.edu

The mother of COBOL [Cobol Language] (and Assembly Language). Grace was also known for carrying around a "nanosecond". It was a one-foot length of copper wire, which, of course, is that distance that electricity will travel in that length of time. (See Whimsical Units Of Measurement) Remember, this was back in the '70s. Her point was if we were going to build computers with cycle speeds faster than 1MHz we would have to put the logic boards closer to each other than the one foot they were in the machines of the late '60s and '70s. Basically, she envisioned the advent of microprocessors.

Actually, while Hopper had junior staff on the COBOL short-range committee, she herself was one of the long-hairs sidelined by the Cobol Causes Brain Damage incident. But she was the mother of Assembly Language and reusable object-code libraries; as cs.yale.edu describes:

By 1949 programs contained mnemonics that were transformed into binary code instructions executable by the computer. Admiral Hopper and her team extended this improvement on binary code with the development of her first compiler, the A-O. The A-O series of compilers translated symbolic mathematical code into machine code, and allowed the specification of call numbers assigned to the collected programming routines stored on magnetic tape. One could then simply specify the call numbers of the desired routines and the computer would "find them on the tape, bring them over and do the additions. This was the first compiler," she declared.

{From what I read elsewhere, Cobol Language borrowed quite a lot from Hopper's earlier work. The link www.objectz.com gives an example of her Flow-Matic language, which allegedly had a big impact on the COBOL committee. It resembles verbose early Fortran Language. She pushed for descriptive names when many wanted to keep short abbreviations.}

Alternative link: en.wikipedia.org

Of course in those days the difference between a compiler and an assembler was kind of hard to pick ...

One of COBOL's original ideas was to allow identifiers long enough to be meaningful. We should be thankful for that.

The originality is somewhat arguable, see the histories of Lisp Language or Fortran Language .

Didn't original FORTRAN only allow 6-character variable names? And Lisp was considered merely a "teaching language" at the time. COBOL was intended to be a production language. Also; Lisp, FORTRAN, and COBOL were being created at about the same time: late 1950s. Any cross-pollination of ideas would probably be through indirect sources or beta's.

''I heard Grace Hopper speak at New Mexico State University in approximately 1980. Here are some things I remember:

She showed up to speak to a group of approximately 400 wearing Navy Blues. At that time she was a Captain.

She remarked that recently while speaking at the Pentagon an enlisted sailor had remarked that she was "probably the oldest Captain in the Navy." She was nearly 80 then.

She gave everyone in attendance a small piece of very thin phone wire - 11.8 inches long, it was a nanosecond.

She carried in her briefcase a coil of wire 960 feet long, weighing around 1.5 pounds. It represented a microsecond.

She remarked that she would like to hang the coil of wire around the neck of any programmer who uttered don't worry, it's only a microsecond or two.

While working on the Cobol Language project for the Navy, she spoke before the Defense Appropriations Committee.

She was requesting $80,000 to support developing a Cobol Language compiler.

She presented a snippet of Cobol code in three languages: English, French and Russian, attempting to show that the language did not matter to the computer.

The congressmen raised an uproar, because: "We don't want our computers speaking Russian."

She remarked that efficiency and clarity in coding would go far toward success in any programming project.

She finished with a very patriotic statement to the effect that she: "was honored to have had the opportunity to serve her country in the US Navy."

I shall never forget her - she made a lasting impression on me on many levels.'' -- Mark Fowler

It appeared she got some flack for pushing the idea that programming could be made more approachable than machine language and its assembler-like cousins of the time. (Vendor independence was another reason for the interest.) Many in the field seemed to think it was a bad idea, partly because it would allegedly attract poorly-skilled riff-raff into the field, but largely for machine resource/performance reasons. However, the hardware improvements eventually made it practical. It was an early consideration of Computer Programming For Everybody. Somewhat related: Mind Overhaul Economics. The "riff-raff" argument perhaps can be compared to the "overly-smart IDE" debates we have today.