Pre-PC and Early Programming Languages - Memories of the Dawn of the Personal Computer

The pioneers who battled with "codes of 0s and 1s" in rooms filled with vacuum tubes

In the era when computers were still called "electronic calculators" and belonged only to certain government agencies and massive corporations, programming was a task entirely different from what it is today.

Early computers had neither keyboards nor displays. For humans to input programs, they had to flip physical switches one by one or feed "punch cards"—cardboard with holes punched in them—into the machine.

All that the computers of that time could understand were combinations of "0s" and "1s" representing voltage on and off—in other words, machine language. Humans would construct complex mathematical formulas in their heads and manually translate them into mind-numbing machine language codes. This was the daily routine back then.
If you made even a single mistake punching a hole along the way, the massive machine would throw a tantrum, and you would have to start all over from the beginning.

Why did humans have to accommodate the machine's needs to such an extent?
Amidst such intense stress and a sense of futility, a groundbreaking idea began to sprout.

"Wouldn't it be better if humans could write programs in a language they could easily understand, and have the computer automatically translate that into machine language?"
The translation program born from this philosophy was the "compiler," and this marked the beginning of the history of high-level languages.

Putting formulas directly into the machine—The challenge of FORTRAN and the genius Backus

In 1957, a team led by IBM's John Backus released the world's first full-scale high-level language.
Its name was FORTRAN, an abbreviation for "Formula Translation." As the name suggests, it was a revolutionary tool that allowed mathematicians and scientists to input the formulas they wrote in their notebooks into the computer almost exactly as they were.

The prototypes for variable assignment and loop processing, which are familiar to anyone who enjoys modern programming, are found here.
However, when this groundbreaking language first appeared, the reaction from veteran programmers in the field was cold. There was fierce opposition, with people claiming, "If you use a soft, easy-to-understand language, the translated program will run too slowly to be of any use."
Since the memory and processing power of computers at the time were extremely precious, that concern was only natural.

Backus and his team became determined. To ensure they wouldn't lose to the efficiency of experts who wrote machine language directly, they developed "optimizing compiler" technology with sheer tenacity, pushing the internal code to its absolute limit.
As a result, FORTRAN achieved overwhelming calculation speeds, silencing the skeptics around them.

Because the design at that time was so excellent, FORTRAN is not dead even today, nearly 70 years after its birth.
In the fields of weather forecasting, space exploration, and structural analysis where the world's supercomputers compete, fully optimized FORTRAN libraries are still running today as the fastest tools available.

COBOL: The "common language of business" that doesn't allow for even a one-yen error

While FORTRAN reigned in the world of scientific calculation, a major transformation was also taking place in the business world.

Administrative calculations such as corporate payroll, sales management, and bank statement processing do not require complex calculus. Instead, they required the ability to accurately sort and process large amounts of text data and numbers.
That is how COBOL was born in 1959.

Among the members who led the development of COBOL was Grace Hopper, a female mathematician who was a rear admiral in the U.S. Navy and is also famous for the episode of popularizing the term "bug" in computing.
She believed, "It must be a language that even businessmen and clerks who hate math can understand at a glance."

For this reason, COBOL's syntax was thoroughly brought closer to natural English sentences. The descriptive method, which used verbs and objects instead of mathematical formulas, was warmly welcomed by the business society of the time.
Furthermore, by being adopted by the U.S. Department of Defense as a "common language that runs on any manufacturer's computer," it instantly established its status as a global standard.

In the modern IT industry, COBOL is often spoken of as a symbol of legacy systems.
However, looking at it from another perspective, that is also proof that its reliability for "decimal calculations that don't allow for a one-yen error" was second to none.
Even at this very moment, deep within bank accounting systems and government core infrastructure, astronomical lines of COBOL code are supporting the backbone of society.

A blooming ecosystem leading to modern code

While the two giants, FORTRAN and COBOL, were paving the way for practical application, important seeds that would lead directly to the modern era were also being sown in the academic world.

One was LISP, developed by John McCarthy in 1958.
Specializing in 'list processing,' this language was exceptionally adept at manipulating symbols and concepts, so much so that, surprisingly, it became the 'standard language for artificial intelligence (AI) research' as early as this era.

Its advanced design, which treats data and programs in the same way, is now being re-evaluated as the origin of modern functional languages.

Another language that, while not a commercial success, played an extremely important historical role was ALGOL.
Developed jointly by researchers in Europe and the United States, this language was the first in the world to clearly define the concept of 'structuring,' which divides programs into specific blocks.

The genes of this ALGOL were passed on to the subsequent Pascal, eventually giving birth to the C language, and connecting continuously to modern Java, Python, and JavaScript.

If ALGOL had not existed, the code we use today might look completely different.

The independence of software and the evolution of the 'vessel'

Thus, from the 1950s to the 1960s, computers evolved from mere 'calculating machines' into 'versatile tools that could become anything' through the replacement of software.

It is not that humans moved closer to machines, but that machines moved closer to human language. This paradigm shift can be said to be the starting point of today's rich IT society.

The reason we can easily manipulate the cloud and AI today is because of the accumulation of the tenacity of our predecessors who once dove into the sea of 0s and 1s and tried to reclaim 'language' for humans.

As humans gained free language, the possibilities for software expanded infinitely.
However, the hardware, the 'vessel' to receive it, was still as huge as a room and belonged to a limited, privileged class.

The language was ready. So, how would this powerful tool shrink to a size that fits on our desks and in our pockets?

The era finally advances to a dramatic tectonic shift in hardware, namely the birth of the personal computer.

[Random Memo] Characteristics of major early languages

1. FORTRAN

A pioneer in scientific and technical calculation

It is the world's first full-scale high-level language (a language close to English or mathematical formulas that is easy for humans to understand, rather than machine language).

【Background and Purpose of Development】
・The "Formula Translator"
Developed in 1957 by John Backus and others at IBM. The name comes from "FORmula TRANslation."
・Liberation from Cryptic Code
It was created to free scientists and engineers, who had previously been writing mind-numbing steps in machine code like "0101..." or assembly languages close to it, from the pain of complex coding.

【Language Features】
・Write Mathematical Formulas Directly
Overwhelming ease of use, allowing scientists to write the mathematical formulas they were accustomed to almost exactly as they were into programs.
・Blazing Fast Processing
It overturned the common wisdom of the time that "languages easy for humans to use result in slow execution speeds," achieving super-fast computational processing comparable to what skilled programmers wrote in machine code.

【Legacy Continuing to the Present】
・Still Active Today
Even now, nearly 70 years after its birth, optimized FORTRAN libraries continue to run in large-scale simulation fields such as weather forecasting with supercomputers, space development, and structural analysis.

2. COBOL

The Absolute King of Business Computing

A business-specialized language born to automate corporate administrative tasks (slip processing, payroll calculation, accounting).

【Background and Purpose of Development】
・A Language Even Clerks Can Read
Developed in 1959 under the leadership of the U.S. Department of Defense, through the efforts of Grace Hopper (known as the mother of compilers and famous for the episode that became the origin of the term "bug").
・The Need for Standardization
At the time, the goal was to unify languages that were scattered across different computer manufacturers and create a "COmmon Business Oriented Language" that would run on any company's machine.

【Language Features】
・Just Like "English Sentences"
Adopted syntax that is thoroughly close to natural English sentences (such as ADD A TO B GIVING C) so that even business people who dislike math could understand it.
・Strong in Financial Calculations
Specialized in accurately processing large amounts of data (forms) rather than complex functions like scientific calculations, and in precise decimal (fixed-point) calculations where "not even a one-yen error is allowed."

【Legacy Continuing to the Present】
・The Hidden Backbone of Society
Behind the world's payment infrastructure, such as bank accounting systems and government core systems, astronomical lines of COBOL assets are still running today. It is a theme very often taken up in business columns as a symbol of legacy systems in the context of "DX (Digital Transformation)."

3. Others

Languages that Expanded the Ecosystem on the Eve of the PC

Important languages that, alongside the two giants FORTRAN and COBOL, sowed the seeds for later PC culture (BASIC, C language, AI, etc.).

(1) LISP (1958-)
・The Progenitor of AI (Artificial Intelligence)
Developed by John McCarthy, it is the second oldest high-level language after FORTRAN. Because it specialized in "LISt Processing" and was good at handling symbols and concepts, it became the standard language for early artificial intelligence research. It is also the source of modern functional languages.

(2) ALGOL (1958-)
・The Entity that Created the "Grammar" of Modern Programming
A language (ALGOrithmic Language) jointly developed by a community of Western researchers. Although not a commercial success, it established the concept of structuring by enclosing programs in blocks (begin ... end). This became the direct ancestor of almost all major modern languages, including Pascal, C, Java, and Python.

(3) PL/I (1964-)
・A Giant Language Aiming to Include Everything
Developed by IBM with the goal of being a universal language that could do anything by combining "FORTRAN for science and technology" and "COBOL for business." However, it left a lesson for posterity that its functions became too bloated and complex, making it difficult to port to small environments like personal computers.

Related Articles

・History and Technological Evolution of Calculators and Computers

Book Introduction

・Our 30-Year PC History: Nippon PC Chronicle Kindle Edition
　Edited by SE Editorial Department
　Shoeisha (2012/12/20)
　A textbook on the history of Japanese PCs, looking back through photographs. It explains the major changes in PCs over more than 30 years in a two-part structure, using many photos and touching on the social conditions of the time to make the evolution clear as time passed. The first part is dedicated to explanations of topics by era, incorporating "testimonies (turning points)" from developers and people involved at the time to reveal behind-the-scenes stories of development. The second part summarizes PC architecture and other topics by theme.
(* Book descriptions, etc., are generally based on the descriptions on the linked site (Amazon). Please refer to the link for detailed content.)

*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-