And how your favorite programming language may become niche as well

AI content warning: Everything after this paragraph is a copy-pasted response from ChatGPT. Today I do not want to serve you my own thoughts. Instead, I want you to think about what has already happened to many programming languages. If you are too lazy to think, or just curious, ask your favorite language model the following question, replacing XYZ with your language of choice:

Imagine it’s 2050. Based on current trends in software development, explain why XYZ became a niche language.

The language model response to the original question follows below.


COBOL became “niche” gradually rather than collapsing suddenly. For decades it was one of the dominant business programming languages in the world, especially in banking, insurance, and government systems. Several trends pushed it from mainstream to specialized use:

1. COBOL was built for a specific era

COBOL was created in 1959 for business data processing on large mainframe computers. Its strengths were:

  • handling huge volumes of records
  • payroll
  • accounting
  • banking transactions
  • government administration

It was excellent for those tasks, but not designed for:

  • graphical interfaces
  • web applications
  • mobile apps
  • modern distributed systems

As computing evolved, newer languages fit emerging needs better.


2. Universities stopped teaching it

From the 1980s onward, computer science education shifted toward:

  • Object-Oriented Programming
  • algorithms
  • operating systems
  • newer languages like:
    • C++
    • Java
    • Python

Students rarely encountered COBOL, so fewer new developers entered the ecosystem.


3. Mainframes became less fashionable

COBOL remained strongly tied to mainframe computing, especially systems from IBM.

During the 1990s and 2000s, companies moved toward:

  • client-server systems
  • Linux/Unix servers
  • cloud infrastructure
  • web architectures

Languages associated with these environments grew rapidly while COBOL looked “old-fashioned.”


4. Legacy systems created a paradox

A huge amount of critical infrastructure still runs on COBOL:

  • banks
  • airline reservation systems
  • tax systems
  • pension systems

The paradox:

  • these systems work extremely well
  • replacing them is risky and expensive
  • companies keep them running for decades

So COBOL never disappeared — it just became concentrated in maintenance and modernization work rather than new product development.


5. Developer culture shifted

Programming culture increasingly valued:

  • startups
  • open source
  • rapid innovation
  • trendy frameworks

COBOL environments often involved:

  • conservative enterprises
  • strict processes
  • older infrastructure

That made it less attractive to younger developers.


6. It became specialized instead of extinct

Today COBOL is more like a specialized industrial technology:

  • fewer developers
  • high demand in certain sectors
  • rarely chosen for new greenfield projects

A useful analogy is:

  • not “dead”
  • more like an old but critical railway system still carrying enormous traffic behind the scenes

During events like the COVID-19 unemployment-surge in 2020, governments in places like the US even struggled to find enough COBOL programmers because unemployment systems still depended on it.

So COBOL became niche mainly because the software industry expanded into many new domains where other languages fit better, while COBOL stayed focused on large-scale business transaction processing.


About me: Principal software engineer by day. I’m mentoring humans and models at Agorapulse. I have created open source projects like Spreadsheet DSL Builder, Gru (HTTP Testing Framework) and plenty of Micronaut related libraries. Now I build side projects through Telegram messages anytime I have a minute of free time.