· 5 min read

Ideas Still Matter

This article was auto-translated from Chinese. Some nuances may be lost in translation.

In tech circles, we often hear that ideas are cheap and execution is what matters most. I generally agree—without taking action, even the best idea is useless.

Recently, however, my perspective has been shifting. While engineers certainly need solid fundamentals, it feels like every day revolves around discussing the internal workings of some framework, how yet another framework just launched, or how someone built a To-Do app with XYZ. On the surface, it seems like a vibrant and diverse technical dialogue, but once you step outside the software development bubble, what are we really changing?

Don’t get me wrong; I’m not saying these things aren’t important or interesting. Frameworks like React truly sparked a front-end revolution, prompting everyone to rethink the essence of front-end engineering. Everyone makes their own career choices: clarifying requirements, writing maintainable and developer-friendly code, and exploring the core of computer science are indeed an engineer’s duty. In my spare time, I also continue to delve into technology as much as I can.

Yet more and more, I get the sense that whenever we encounter even a simple requirement, we feel compelled to over-engineer it. Actual functionality takes a back seat—the front end must start with React, Angular, or Vue; we must set up CI/CD, SSR, DevOps, various third-party integrations, linters, and Webpack before we even begin talking about features.

With the rapid advancement of tools like Heruoku, Netlify, ZEIT, AWS Lambda, and countless other cloud services, you don’t need to spend massive amounts of time to build a decent, usable product anymore. We are no longer in an era where you have to painfully roll everything by hand. Even with CMS tools like WordPress, watching a quick online tutorial and setting it up on a reliable server might result in a website that is faster and more stable than one you spent a month coding from scratch.

Here are a few examples that come to mind:

GitHub

An example I heard a senior developer share before is GitHub.

Remember GitHub’s article on removing jQuery? Starting from the viewpoint of browser advancements, the post emphasized that modern browsers can now handle modern APIs natively, which allowed GitHub to gradually remove jQuery.

Since then, we haven’t heard much about GitHub adopting any trendy new frameworks. Yet if you look closely at GitHub’s UI and user experience, while not necessarily flawless, it is remarkably stable, usable, and relied upon by developers around the globe.

Behind it all runs Ruby on Rails—a technology constantly mocked by engineers. Its front end doesn’t rely on trendy frameworks like React or Vue either; instead, it uses custom elements.

The service and value GitHub provides have arguably become an indispensable part of software engineering.

The Mask Rationing System

Then there’s Taiwan’s mask rationing system. When it first launched, complaints flooded in: the system was clunky, and the UI was ugly.

I think any engineer who has ever inherited a legacy codebase can easily understand why. The objective at that moment was to solve an urgent crisis. If they had insisted on making everything perfect before shipping, it likely would have ended up in limbo.

It serves as a prime example of rapid iteration: put out a viable solution to handle the emergency and meet user needs first, then gradually adapt and refine it over time.

Besides, once the pandemic ends (hopefully soon), far fewer people will be using this system anyway. This case reminds us that there is no such thing as a “perfect system”—systems simply evolve alongside needs.

Real-name mask ordering using National Health Insurance cards and purchasing online were brilliant ideas that genuinely solved critical user needs.

NiceChord’s Subtitling Tool

A few weeks ago, in my article “The Rise of Creators — Coding”, I mentioned NiceChord’s subtitling program as an example.

NiceChord used simple code + p5.js to build a prototype that solved his own subtitling workflow problem. The technology itself wasn’t difficult, but had anyone previously thought of creating subtitles this way and then actually sat down to build it?

Liu Zhiyuan (Huber)

A few weeks ago, Liu Zhiyuan (Huber) also posted a video showing how he used CSS animations combined with Google Sheets to assist his animation workflow. Although it was a simple concept, the visual result looked amazing. Seeing examples like this always makes me marvel: Wow, such a straightforward approach can yield such fantastic results—why didn’t I think of that?


Honestly, there are countless more examples I could list. What I want to highlight here is that ideas still matter. Especially as tools and technology continue to advance and the barrier to bringing ideas to life becomes lower and lower, anyone willing to build things—even without a traditional engineering background—can achieve remarkable results.

Related Posts

Explore Other Topics