GitHub Copilot - the shape of things to come.
Why Developer need to stop fearing AI and learn to love it.
Software Development took a big step today towards abstracting away huge swatchs of tedium and errors with GitHub’s annnouncement of GitHub Copilot.
Developers have been fighting a losing battle as programming has become more complex. When I started out a long time ago, all I had to learn was a single language running on a single platform. Every couple of years the foremost author would release an update of his “Big Book” that I would then read, reread and markup.
Those days are long gone. To do my job, I need to know literally dozens of languages and tools, and keep up with them. That single language has evolved into a long toolchain of languages, libraries, standards, that is constantly changing.
Off the top of my head, I need to keep know well and keep track of Ruby on Rails (5, 6 and soon 7), CSS (TailwindCSS), React, React Native, StimulusJS, Hotwire, and VS Code. Let me pick on just one [favorite] tool in my toolbox, TailwindCSS for a second: it’s now 2.2.4; that’s 13 successive releases in less than two years.
The promise of GitHub Copilot is, “Trained on billions of lines of public code, GitHub Copilot puts the knowledge you need at your fingertips, saving you time and helping you stay focused.”
It [supposedly] abstracts the day-to-day tedium of programming away, letting developers craft code at a higher level, and hopefully have more time to think about the code and functionality they are building.
GitHub Copilot and other AI-based tools aren’t going to make our profession go away. But it will give us a much-needed assist so we can write better code, make better decisions, and for the right reasons.
I’m not sure if GitHub Copilot will live up to its hype. But I’m eager to start using it.