Originally published on josnun.com
Larger groups gaining access to tools and capabilities that were previously exclusive to the elite is no new phenomenon. Books and literacy, photography, manufacturing (eg, 3D printing), and computing are all examples of capabilities that used to be out of reach of the average person, but are now commonplace. With the advent of AI, the next capability to become democratized is seemingly software.
With each of these shifts, there have been some naysayers declaring that giving "commonfolk" these tools will destroy the craft, wipe out an entire category of jobs, spoil the capabilities, or otherwise lead to the destruction of society as we know it. To some extent, this holds true, but broadly speaking, the democratization has enabled new opportunities for the common man, while still making space for those who excel at their craft. Just because everyone has a phone with a camera doesn't mean there isn't a place for photographers. Their understanding of composition and what makes a good photo is what sets them apart. That's only learned through effort, practice, study, and more practice. One doesn't become an expert photographer just because they have a camera.
The same will hold true for software. AI makes building software seem easy. And that's great. I'm all for a whole army of people building software that meets their needs, and frees them up to do the work that sets them apart; whether that's getting better at their job, hobbies, or spending time with their families and communities. Quality software is hard to build, though. There's a difference between something that "works" and something that works well. The differentiator is quality and attention to detail. Good software is more than just some code slung on a page. There's design, UX, discovering what doesn't work, iteration. Using it, and discovering that what you thought you wanted isn't what you actually wanted; what doesn't "feel" right. On some level, there's an intuition that you build after doing the reps of building software. It's learning when to trust your intuition, and when you need to prove it.
Everyone having the capability to build software is a good thing. There are whole categories of software that would be nice to have, but aren't financially viable to build a business around. Let the people build it themselves. There may be some people out there whose expertise in other disciplines have made them masters of building good software, but the high barrier to entry has kept them away. People have ideas. Maybe their ability to execute on those ideas will unlock whole new categories of tools and software. I expect that, just like with books, software that has had heart and soul poured into it and is a cut above the rest will rise to the top, raising software quality for everyone.
When everyone can produce, quality becomes the differentiator. Great software makes you feel things; it's art. That's why I'm hopeful. There's a certain joy that I've been able to unlock by building what lives in my imagination, and I'm excited to see that joy come to others. Will this transition be smooth? I doubt it. I'm sure there will be rough spots and setbacks, but the promise of a new class of software and the potential for refined craft still leaves me optimistic.
