- Over the decades, each new programming language has made software development more accessible and usable than what came before
- Low-code tools are the latest example of this trend
- Low-code empowers “citizen developers” to solve their own business workflow challenges, freeing pro developers to focus on more complex, mission-critical projects
When Guido van Rossum created Python in the early 1990s, he aimed to replace an existing coding language, called ABC, with a less complex, faster to deploy, easier-to-use alternative.
Python is now the world’s dominant software programming language. And van Rossum’s core principles drive the current boom in low-code application development tools, which allow non-programmers to build powerful business applications and bring them online. “Python comes from that tradition where the intended user is someone whose primary responsibility is not software development or coding, but getting something done,” he says.
It’s a trend that excites tech analysts. Gartner predicts the majority of future enterprise applications will be built by citizen developers rather than professional coders.
Low code is the latest example of a trend that dates back to the early days of computing. Over the decades, each new coding language has made software development more accessible, more comprehensible, and more usable than what came before.
Breaking the code
In its earliest days, computer programming was an arcane discipline of elite practitioners. It usually required an advanced degree in mathematics and a mastery of hexadecimal instructions and binary code. That changed in 1954, when a small IBM team introduced Fortran, for Formula Translator, which required much less human input than predecessors. Where earlier systems might take as many as 1,000 instructions for a specific problem, Fortran could do the job with fewer than 50.
“The intention of [Fortran’s] design was very clearly to allow engineers to take textbook formulas and turn them into computer code without having to learn the whole hardware model,” says van Rossum.
Each new coding language has made software development more accessible, more comprehensible, and more usable than what came before.
New languages continued this trend toward greater simplicity. COBOL, for Common Business Oriented Language, followed in 1959. COBOL greatly simplified the programming of financial and business applications. In 1964, two Dartmouth College professors invented BASIC to make programming easier for students. Microsoft went on to create multiple versions of the BASIC language, which was so popular by the 1970s that it was installed by default on many of the earliest home computers.
With BASIC came a shift to “syntactical” commands consisting of common words, instead of numerical sequences, that could be reused over and over. The language “C” was developed in 1972 and used simple keywords to tell a computer what to do. In 1987, Larry Wall, a systems administrator at NASA’s Jet Propulsion Laboratory, created Perl, a language designed for programmers with any amount of coding experience. That same year, Apple released HyperCard, which enabled users to build apps using a graphical interface instead of if-then logic statements.
With the rise of the internet in the 1990s, new languages appeared that streamlined the creation of websites and web applications. Van Rossum released the earliest version of Python in 1991. Java, released in 1995, underpins Android development at Google and desktop computing at most of the world’s Fortune 500 companies. In a 2021 Workflow article, ServiceNow CIO Chris Bedi wrote: “Each new language reduced the complexity of software development, making the field accessible to more people.”
Low code to no code
Two factors are driving low-code demand. First, the shortage of skilled programmers: there were 1.2 million job openings for software developers in September 2021. Second, the pressure companies have been under during the pandemic to accelerate the creation and launch of apps to keep millions of remote employees and ecommerce customers happy. According to an IDC January 2022 survey of 380 enterprise leaders, nearly half said they had begun investing in low-code or no-code platforms.
“Front-line business people aren’t trying to write code,” says Matt Hubbard, head of operational excellence at TrackVia, a low-code app-building platform. “They’re trying to solve problems with apps.”
In the future, machine learning and other advanced technologies promise to make low-code/no-code app design even easier. A developer might start with a basic template, then refine it by inserting snippets of code written by trained AI models, in much the same way that Gmail suggests quick-reply responses. It’s already happening. OpenAI’s Codex project takes simple language instructions such as “Create a web page” and translates them into working code using natural-language processing.
To be sure, highly skilled professional programmers aren’t becoming an endangered species anytime soon. As van Rossum sees it, low-code specialists (and AI tools) will take over the more rudimentary aspects of programming in the years to come, freeing up pros to focus on more complex, mission-critical tasks and projects.
“Writing the more interesting and creative possibilities of code will always be fundamental,” van Rossum says. “Humans will still be very much in need.”