Are we making the web too complicated?

posted 21 May 2017

Everyone's favorite sarcastic talking pushpin asked an honest question about the state of the web:

I replied with a tweetstorm. Here it is as a slightly more readable blog post on my ancient, creaky blog.

The replies to Maciej's tweet are interesting to read. They fall roughly into two camps:

  1. Older/not front-end developers: because the web is shit!
  2. Current front-end developers: because shit is hard!

As is often the case, both camps are correct! The web is a shitshow of wheel reinvention and bad APIs. It's also a blizzard of innovation.

Expectations for what a web site should be able to do have evolved enormously. Users expect snappy, desktop-like responsiveness and rich presentation in web apps. They also expect those same web apps to work equally well on mobile devices. And they expect these apps to load basically instantly. As Tom Dale says, that's actually a harder problem than desktop or mobile apps face. Users expect to download and install those types of apps before they will run.

Devs must meet these high expectations, but they have no more time and no more co-workers than they did before, and they still have to ship just as fast. To meet this requirement, they are throwing ever larger combinations of frameworks, boilerplate code, tools and build chains at the problem. The result is a lot of complexity and it's frequently frustrating.

The web as a platform, as ever, lags behind its developers. They want ES6 syntax, they want modular JavaScript, modular CSS and modular HTML. Browsers provide none of these things, so they are hacked together in frameworks.

Devs aren't adopting all this new tech just because it's new and fun (though it's a bad idea to dismiss fun as a valuable quality in a development tool). They are doing it because they have to in order to survive.

Modern web dev has the complexity of, say, the native mobile ecosystem, but no single vendor or consensus build chain or tools. iOS and Android have compilers and long build steps and dozens of competing frameworks, but nobody complains that these things are unnecessary. But they do for the web, because the web didn't need them before. (And if you're okay with throwing together a simple, 2000s-era web page, they're still not. But few users are happy with those any more.)

Are there some people using a huge pile of JavaScript and a monstrous build chain to throw together a single-pager web site with one box that collects an email address? For sure. And that's silly and unnecessary. But so what? The misuse of technology does not invalidate it.

Over the next 5 years I expect there will be a lot of consolidation in technologies and in tools. A lot of the stuff web devs are currently constructing using frameworks will be adopted as first-class citizens, built into browsers. Because the web has a lot of inertia, people will keep using these tool chains longer than they need to, much as web devs still use jQuery even though mosts of its API is part of browsers natively now.

jQuery is an instructive example: jQuery was a reaction to a terrible API and a lack of raw functionality in the web at the time. It changed the web for the better, forever, by showing browser makers what devs needed, and how it should work. It was a huge success, and its ultimate success is that it made itself unnecessary.

Webpack, babel, react, and the cambrian explosion going on in that ecosystem will do the same thing again. All of these frameworks and tools are devs experimenting, seeing where they want the web to go.

jQuery was a performance problem at scale, and sometimes over- or mis-applied, or used superfluously by newbies who didn't know there was a simpler way. Big deal.

Modern front-end devs are repeating the same mistakes: these new frameworks often create shocking performance hits. Sometimes we over-use them. Sometime we mis-use them. But the web is evolving, and evolution is by nature slow and messy. And the results are already amazing.

Is modern web development fearsomely, intimidatingly complicated? Yes, and that's a problem. Will we make it simpler? Definitely, but probably not as soon as you'd like. Is all this new complexity worthwhile? Absolutely.

The web's amazing capacity to reinvent itself, to evolve and adapt to new needs is its strength as a platform. Things that don't adapt die or are replaced. Nobody is talking about the death of the web. Nobody is demanding it be replaced. If anything they're pleading for it to slow down a bit, and let them catch up. That's a sign of a healthy platform, an innovative platform. The web was like that in 1996 and it's still like that, and that's amazing.

Nobody but nobody loves the web more than I do. It's my baby. And like a child, it's frustrating to watch it struggle and make mistakes. But it's amazing to watch it grow up.

0 comment