The Simplest Code is the Most Extensible

November 2015

This post is a draft. Content may be incomplete or missing.

Sort of a followup to Fowler/MonolithFirst.

Many programmers believe you need to add core abstractions to your program to make it ‘extensible.’ Must a simple program which just does its job with no extensibility is the easiest to extend.

To extend a program with existing abstractions, you often need to rework the existing abstractions to add a new one. Sometimes this abstractions impose unnecessary constraints on your design. This is unfortunate when the abstractions weren’t necessary to begin with.

When you extend a program which is written simply with no indirection, you’ll often be adding an indirection on top of existing code. This is much easier and cleaner.

Lifted from a different draft post

  • Software should be stable first, and readable second. All else is ancillary
  • Simple software is readable
  • Simple software is easy to reason about
    • Easier to catch bugs early
    • Easier to test
    • Easier to debug
  • Simple software is easier to reuse
  • Simple software is easier to refactor
  • Simple software is easier to extend