← Back to home

History Of Software

Honestly, you want me to dissect this dry chronicle of silicon and code? Fine. But don't expect enthusiasm. It’s just… more data. More of the same predictable progression.

This article needs additional citations for verification. Honestly, who cares about proving every single tedious detail? It's not like the world hinges on whether someone remembers the exact date a certain line of code was committed. But, fine. If you insist on the pretense of absolute accuracy, help improve this article by adding citations to reliable sources. If this unsourced material is so offensive, it may be challenged and removed. Just don’t expect me to break a nail hunting for your precious "reliable sources." Find sources: "History of software" – news, newspapers, books, scholar, JSTOR (August 2016). Or don't. ( Learn how and when to remove this message ).

History of computing

Hardware

Software

Computer science

Modern concepts

By country

Timeline of computing

Glossary of computer science

Software is, in essence, a collection of meticulously crafted instructions, etched into the memory of those stored-program digital computers you seem so fascinated by, waiting to be executed by the processor. It’s a rather recent invention in the grand, weary scheme of human history, but undeniably fundamental to this so-called Information Age.

Now, the genesis. Some credit Ada Lovelace and her hypothetical programs for Charles Babbage's analytical engine back in the 19th century. They call her the "founder." Cute. But her efforts, like so many grand ideas, remained purely theoretical. The technology of their era was simply too… primitive to bring such a vision to life. Then there’s Alan Turing, who, in 1935, supposedly laid down the first theoretical groundwork for software. This, they say, birthed the twin academic fields of computer science and software engineering. Fascinating.

The earliest software, for those first stored-program digital computers in the late 1940s, was a brutal, direct affair: instructions written straight into binary code, usually for those colossal mainframe computers. But as modern programming languages evolved, hand-in-hand with the rise of the home computer, the scope of what software could achieve exploded. It started with the rudimentary assembly language and continued through the more sophisticated paradigms of functional programming and object-oriented programming.

Before stored-program digital computers

Origins of computer science

See also: History of computer science

The concept of "computing" itself is ancient, stretching back to devices like the abacus, the intricate Antikythera mechanism, astrolabes, those elaborate mechanical astronomical clocks, and early mechanical calculators. [1] The Antikythera mechanism, in particular, stands as a testament to an astonishingly complex ancient mechanical astronomical device. [2]

However, these were purely hardware affairs. They possessed no "software" in the modern sense; their computational capabilities were intrinsically bound to their physical form and engineering.

Software, as we understand it, necessitates the abstract notion of a general-purpose processor – what is now conceptualized as a Turing machine – coupled with computer memory. This memory is where reusable sets of routines and mathematical functions, the very essence of programs, can be stored, initiated, and terminated independently. Such a concept is a relatively recent arrival in the human narrative.

The first documented computer algorithm is attributed to Ada Lovelace in the 19th century. It was designed for the analytical engine and aimed to translate Luigi Menabrea's work on Bernoulli numbers into machine instructions. [3] Yet, this remained a theoretical exercise. The state of engineering during the lifetimes of these two brilliant minds was simply insufficient [ citation needed ] to construct the ambitious analytical engine.

The first truly modern theory of software emerged from Alan Turing in his seminal 1935 essay, "Computable numbers with an application to the Entscheidungsproblem (decision problem)." [4]

This theoretical foundation eventually paved the way for the establishment of two distinct academic disciplines: computer science, which delves into the more theoretical aspects (Turing's essay being a prime example), and software engineering, which focuses on the practical application and creation of software.

However, it's crucial to note that before 1946, software as we conceive it today – programs residing within the memory of stored-program digital computers – simply did not exist. The earliest electronic computing devices were "reprogrammed" through laborious rewiring. The ENIAC, one of the pioneers in electronic computing, was largely programmed by women who had previously functioned as human computers. [5] [6] Engineers would present these programmers with blueprints of the ENIAC's wiring, expecting them to decipher the logic for programming the machine. [7] These women were instrumental in preparing the ENIAC for its public debut, meticulously wiring the patch panels for demonstrations. [8] [9] [10] In 1950, [Kathleen Booth] developed assembly language to simplify the programming process for the computers she worked on at Birkbeck College. [11]

Grace Hopper and UNIVAC

Grace Hopper was among the first programmers for the Harvard Mark I. [12] She later compiled a comprehensive 500-page manual for the computer. [13] While Hopper is often anecdotally credited with coining the terms "bug" and "debugging" after finding a moth causing a malfunction in the Mark II, the term "bug" was actually in use prior to that incident. [14] Hopper's groundbreaking work led to the development of the first compiler, a concept she carried from her work on the Mark computers to her endeavors with UNIVAC in the 1950s. [15] She also developed the FLOW-MATIC programming language for the UNIVAC. [14] Concurrently, Frances E. Holberton, also at UNIVAC, developed a code [ clarification needed ] known as C-10, which enabled programmers to use keyboard inputs, and in 1951, she created the Sort-Merge Generator. [16] [17] Adele Mildred Koss, alongside Hopper, was also involved in the creation of an early precursor to a report generator. [16]

Early days of computer software (1948–1979)

In his influential 1948 manuscript, "A Mathematical Theory of Communication," Claude Shannon (1916–2001) laid out the theoretical framework for implementing binary logic in computer programming. Following this, the very first computer programmers relied on binary code to direct computers. This was, predictably, an arduous process. Programmers had to meticulously construct lengthy strings of binary code, essentially dictating every operation. Loading these instructions and data onto computers was a tedious affair, involving flicking switches or punching precise holes in punched cards and then feeding them into the machine. A single error in this process could necessitate reloading the entire program from scratch.

The landmark achievement of the first stored-program computer successfully executing software held within its electronic memory occurred at 11 am on June 21, 1948, at the University of Manchester, using the Manchester Baby computer. This program, authored by Tom Kilburn, was designed to calculate the highest factor of the integer 2^18, which is 262,144. It employed a large trial divisor, repeatedly subtracting it from 262,144 and checking for a zero remainder. If the remainder was not zero, the divisor was decremented by one, and the process repeated. Google even released a tribute to the Manchester Baby, recognizing it as the "birth of software."

The development of FORTRAN was spearheaded by a team at IBM led by John Backus in the 1950s. The initial compiler was released in 1957. The language's efficacy for scientific and technical computing was so profound that by 1963, all major manufacturers had either implemented or announced FORTRAN support for their machines. [18] [19]

COBOL's conception began in 1959 when Mary K. Hawes convened a meeting, including Grace Hopper, to discuss the creation of a standardized business-oriented computer language. [16] Hopper's contribution to COBOL was the introduction of a novel symbolic notation for programming, making the code more self-documenting. [20] [Betty Holberton] played a role in editing the language, which was submitted to the Government Printing Office in 1960. [21] In the 1960s, [Jean E. Sammet] developed FORMAC. Her influential book, Programming Languages: History and Fundamentals (1969), became a standard text. [21] [22]

Apollo Mission

Margaret Hamilton is pictured here next to a stack of code she and her team developed for the Apollo Mission computers.

The ambitious Apollo Mission to the moon was critically dependent on software to operate the computers within the landing modules. [23] [24] The programming language used was called "Basic" (distinct from the BASIC programming language developed concurrently at Dartmouth). [25] The software suite also incorporated an interpreter, comprising a series of routines, and an executive system (akin to a modern operating system), which managed program execution and scheduling. [25] Both were designed by Hal Laning. [25] Margaret Hamilton, who had previously grappled with software reliability challenges during her work on the US SAGE air defense system, was also a key member of the Apollo software team. [23] [26] Hamilton's primary responsibility was the onboard flight software for the Apollo computers. [23] She held a profound belief that software operations were not merely mechanical functions but were intricately intertwined with the human operators. [25] It was also during her tenure at NASA that Hamilton coined the term "software engineering". [27]

The actual "software" for the Apollo missions' computers was physically implemented using wires threaded through magnetic cores. [28] A wire passing through a magnetic core represented a binary "1," while a wire looping around the core represented a "0." [28] Each core was capable of storing 64 bits of information. [28] Hamilton and her colleagues would develop the software by punching instructions onto punch cards, which were then processed on a Honeywell mainframe for simulation. [23] Once the code was deemed "solid," it was sent to Raytheon for implementation, where women known as "Little Old Ladies" meticulously wove the wires through the magnetic cores. [23] This method resulted in programs that were remarkably "indestructible," capable of withstanding even lightning strikes, as famously occurred during the Apollo 12 mission. [28] The process of wiring these computers was time-consuming, often taking several weeks, which effectively halted software development during that period. [29]

During testing on the simulators, Hamilton identified potential critical errors that could arise from human mistakes during operation. [23] NASA, confident in the astronauts' extensive training, did not initially prioritize preventing such errors. [30] Hamilton's proposals to incorporate error-checking code were rejected as "excessive." [23] Tragically, her predictions proved accurate during the Apollo 8 flight, when human error led to the computer erasing all navigational data. [23]

Bundling of software with hardware and its legal issues

In subsequent years, software was often sold alongside hardware by original equipment manufacturers (OEMs) like Data General, Digital Equipment, and IBM. Customers purchasing a minicomputer, then the smallest available computer, would find it lacked pre-installed software and required installation by OEM-employed engineers. [ citation needed ]

This practice of bundling attracted the scrutiny of US antitrust regulators. In 1969, IBM faced a lawsuit from the US government alleging improper "tying" practices, arguing that forcing customers to buy or lease hardware to obtain software constituted an antitrust violation. However, after years of legal proceedings, the US Justice Department ultimately dropped the case, deeming it "without merit." [31]

Data General also faced legal challenges related to bundling, though in this instance, it stemmed from a civil suit initiated by a potential competitor. When Data General launched the Data General Nova, the company Digidyne sought to utilize its RDOS operating system on its own hardware clone. Data General refused to license their software, asserting their "bundling rights." In 1985, the US Supreme Court, by allowing a 9th circuit appeal court's decision to stand, established a precedent in Digidyne v. Data General. This ruling eventually compelled Data General to license its operating system, as restricting its use to DG hardware was deemed an illegal tying arrangement. [32] While the District Court had initially noted that "no reasonable juror could find that within this large and dynamic market with much larger competitors," Data General "had the market power to restrain trade through an illegal tie-in arrangement," the appellate court ruled the tying of the operating system to the hardware as per se illegal. [33]

More recently, in 2008, Apple Inc. sued Psystar Corporation for distributing unauthorized Macintosh clones with OS X pre-installed, leading Psystar to countersue. One of Psystar's arguments, referencing the Data General case, was that Apple maintained a dominant position in the market for OS X-compatible computers through an illegal tying arrangement between the operating system and Apple hardware. District Court Judge William Alsup dismissed this claim, stating that, similar to the District Court's ruling in Data General, the relevant market encompassed all PC operating systems, not just Mac OS, and that Mac OS did not hold a dominant position within this broader market. Judge Alsup also pointed out that the precedent set in Data General, which suggested that the tying of copyrighted products was always illegal, had been "implicitly overruled" by the verdict in the Illinois Tool Works Inc. v. Independent Ink, Inc. case. [34]

Packaged software (Late 1960s-present)

  • This section needs expansion. You can help by adding to it. (March 2019)

The late 1960s witnessed the emergence of an industry dedicated to independently packaged software – that is, software not created as a unique, one-off product for a specific client, nor bundled as a standard offering with computer hardware. [35]

Unix (1970s–present)

Unix stands as an early and profoundly influential operating system that continues to exist today. The most prevalent descendant of Unix in contemporary use is macOS (formerly known as OS X and Mac OS X), while Linux shares a close lineage.

The rise of Microcomputers

In January 1975, Micro Instrumentation and Telemetry Systems began offering its Altair 8800 microcomputer kit through mail order. Later that year, Microsoft released its inaugural product, Altair BASIC, and the burgeoning community of hobbyists started developing programs for these kits. Tiny BASIC was published as a type-in program in Dr. Dobb's Journal, with its development becoming a collaborative effort.

In 1976, Peter R. Jennings, for example, created his Microchess program for MOS Technology's KIM-1 kit. However, lacking a tape drive, Jennings distributed the source code in a small booklet to his mail-order customers, who then had to painstakingly type the entire program by hand. In 1978, Kathe and Dan Spracklen released the source code for their Sargon (chess) program within a computer magazine. Jennings eventually transitioned to selling paper tape and, later, compact cassettes containing the program.

The process of manually typing source code from a computer magazine was inconvenient and time-consuming. A single mistyped—or, worse, misprinted—character could render the entire program inoperable, yet individuals persisted with this method. (Optical character recognition technology, which could have potentially automated the input of listings, was not yet widely accessible.)

Even with the proliferation of cartridges and cassette tapes in the 1980s for distributing commercial software, free programs (such as simple educational tools designed to teach programming concepts) were still often printed, as it was more economical than producing and attaching cassette tapes to magazines.

Ultimately, a confluence of four factors brought an end to the practice of printing complete source code listings of entire programs in computer magazines:

  • Programs began to grow significantly in size.
  • Floppy discs became a viable medium for software distribution and their cost decreased.
  • The user base expanded to include ordinary individuals seeking a straightforward way to run programs.
  • Computer magazines started to include cassette tapes or floppy discs containing free or trial versions of software.

Commercial software quickly became subject to piracy, much to the chagrin of its producers. Bill Gates, co-founder of Microsoft, was an early proponent of combating software piracy, famously articulating his stance in his 1976 Open Letter to Hobbyists. [36]

1980s–present

  • This section needs expansion. You can help by adding to it. (September 2013)

Prior to the advent of the microcomputer, a successful software program might sell up to 1,000 units at a price point of 50,000–60,000each.Bythemid−1980s,personalcomputersoftwarewassellingthousandsofcopiesfor50,000–60,000 each. By the mid-1980s, personal computer software was selling thousands of copies for 50–700 each. Companies such as Microsoft, MicroPro, and Lotus Development achieved tens of millions of dollars in annual sales. [37] They also established dominance in the European market by providing localized versions of their successful products. [38]

A critical turning point in computing history was the publication of the specifications for the IBM Personal Computer by IBM employee Philip Don Estridge in the 1980s. This move rapidly led to the PC's ascendance and subsequent dominance in the worldwide desktop and later laptop markets – a position it largely maintains to this day. Microsoft, through its shrewd negotiation with IBM to develop the initial operating system for the PC (MS-DOS), reaped immense profits from the PC's success over the subsequent decades, driven by the widespread adoption of MS-DOS and its successor, Microsoft Windows. Securing that negotiation was a pivotal moment in Microsoft's trajectory.

Free and open source software

Free and Open Source Software (FOSS) encompasses software that is not only freely available for use but also distributed under licenses that grant users the liberty to access, modify, and share its source code. This model stands in contrast to proprietary software, where the source code is typically kept confidential, and usage is governed by restrictive licensing agreements. FOSS fosters an environment of collaboration and transparency, empowering developers and users globally to contribute to software improvements, customize it to their specific requirements, and share enhancements without encountering legal or financial impediments. Prominent examples of FOSS include operating systems like Linux, web browsers such as Mozilla Firefox, and programming languages like Python. The underlying philosophy of FOSS not only drives technological innovation but also cultivates a worldwide community dedicated to creating software that is both accessible and adaptable to a wide array of needs.

Recent developments

App stores

In recent years, applications designed for mobile devices (cellphones and tablets) have been designated as "apps." Apple strategically channeled app sales for the iPhone and iPad through its App Store, thus controlling app vetting and receiving a portion of every paid app sale. Apple prohibits apps that could be used to bypass its app store ecosystem (e.g., virtual machines like those for Java or Flash).

In contrast, the Android platform supports multiple app stores, generally allowing users to choose their preferred option (though Google Play requires a compatible or rooted device).

This model has been emulated for desktop operating systems with platforms like GNOME Software (for Linux), the Mac App Store (for macOS), and the Windows Store (for Windows). Crucially, all these platforms maintain their inherent non-exclusivity: they permit application installations from sources outside their respective app stores, and even from alternative app stores.

The explosive surge in app popularity, particularly for the iPhone but also for Android, triggered a kind of "gold rush," with numerous aspiring programmers investing considerable time and effort into creating apps in the hopes of achieving financial success. As is often the case in such speculative endeavors, not all of these ambitious entrepreneurs found the fortune they sought.

Formalization of software development

The development of curricula within computer science has significantly contributed to advancements in software development practices. Key components of these academic programs include:

How software has affected hardware

As an increasing number of programs migrate into the realm of firmware, and as hardware itself becomes progressively smaller, more affordable, and faster—a trend largely anticipated by Moore's law—a growing array of functionalities, initially executed by software, are now being integrated directly into hardware. A prime example of this is the evolution of graphics processing units. (Conversely, the pendulum has sometimes swung the other way due to cost or other considerations, as seen with softmodems and microcode.)

Today, most hardware companies employ a larger number of software programmers than hardware designers [ citation needed ], a shift facilitated by software tools that have automated many of the tasks previously performed by printed circuit board (PCB) engineers.

Computer software and programming language timeline

The following tables meticulously document the year-by-year development across various facets of computer software, encompassing:

1971–1974

| | 1971

| Programming languages | Operating systems | Computer networks | Computer graphics