Write less code

The most important metric you’re not paying attention to

Rich Harris Apr 20 2019

All code is buggy. It stands to reason, therefore, that the more code you have to write the buggier your apps will be.

Writing more code also takes more time, leaving less time for other things like optimisation, nice-to-have features, or being outdoors instead of hunched over a laptop.

In fact it’s widely acknowledged that project development time and bug count grow quadratically, not linearly, with the size of a codebase. That tracks with our intuitions: a ten-line pull request will get a level of scrutiny rarely applied to a 100-line one. And once a given module becomes too big to fit on a single screen, the cognitive effort required to understand it increases significantly. We compensate by refactoring and adding comments — activities that almost always result in more code. It’s a vicious cycle.

Yet while we obsess — rightly! — over performance numbers, bundle size and anything else we can measure, we rarely pay attention to the amount of code we’re writing.

Readability is important

I’m certainly not claiming that we should use clever tricks to scrunch our code into the most compact form possible at the expense of readability. Nor am I claiming that reducing lines of code is necessarily a worthwhile goal, since it encourages turning readable code like this...

for (let let i: numberi = 0; let i: numberi <= 100; let i: numberi += 1) {
	if (let i: numberi % 2 === 0) {
		var console: ConsoleThe console module provides a simple debugging console that is similar to the
JavaScript console mechanism provided by web browsers.
The module exports two specific components:

A Console class with methods such as console.log(), console.error() and console.warn() that can be used to write to any Node.js stream.
A global console instance configured to write to process.stdout and
process.stderr. The global console can be used without calling require('console').

Warning: The global console object’s methods are neither consistently
synchronous like the browser APIs they resemble, nor are they consistently
asynchronous like all other Node.js streams. See the note on process I/O for
more information.
Example using the global console:
console.log('hello world');
// Prints: hello world, to stdout
console.log('hello %s', 'world');
// Prints: hello world, to stdout
console.error(new Error('Whoops, something bad happened'));
// Prints error message and stack trace to stderr:
//   Error: Whoops, something bad happened
//     at [eval]:5:15
//     at Script.runInThisContext (node:vm:132:18)
//     at Object.runInThisContext (node:vm:309:38)
//     at node:internal/process/execution:77:19
//     at [eval]-wrapper:6:22
//     at evalScript (node:internal/process/execution:76:60)
//     at node:internal/main/eval_string:23:3

const name = 'Will Robinson';
console.warn(`Danger ${name}! Danger!`);
// Prints: Danger Will Robinson! Danger!, to stderr
Example using the Console class:
const out = getStreamSomehow();
const err = getStreamSomehow();
const myConsole = new console.Console(out, err);

myConsole.log('hello world');
// Prints: hello world, to out
myConsole.log('hello %s', 'world');
// Prints: hello world, to out
myConsole.error(new Error('Whoops, something bad happened'));
// Prints: [Error: Whoops, something bad happened], to err

const name = 'Will Robinson';
myConsole.warn(`Danger ${name}! Danger!`);
// Prints: Danger Will Robinson! Danger!, to err
@seesourceconsole.Console.log(message?: any, ...optionalParams: any[]): void (+1 overload)Prints to stdout with newline. Multiple arguments can be passed, with the
first used as the primary message and all additional used as substitution
values similar to printf(3)
(the arguments are all passed to util.format()).
const count = 5;
console.log('count: %d', count);
// Prints: count: 5, to stdout
console.log('count:', count);
// Prints: count: 5, to stdout
See util.format() for more information.
@sincev0.1.100log(`${let i: numberi} is even`);
	}
}

...into something much harder to parse:

for (let let i: numberi = 0; let i: numberi <= 100; let i: numberi += 1) if (let i: numberi % 2 === 0) var console: ConsoleThe console module provides a simple debugging console that is similar to the
JavaScript console mechanism provided by web browsers.
The module exports two specific components:

A Console class with methods such as console.log(), console.error() and console.warn() that can be used to write to any Node.js stream.
A global console instance configured to write to process.stdout and
process.stderr. The global console can be used without calling require('console').

Warning: The global console object’s methods are neither consistently
synchronous like the browser APIs they resemble, nor are they consistently
asynchronous like all other Node.js streams. See the note on process I/O for
more information.
Example using the global console:
console.log('hello world');
// Prints: hello world, to stdout
console.log('hello %s', 'world');
// Prints: hello world, to stdout
console.error(new Error('Whoops, something bad happened'));
// Prints error message and stack trace to stderr:
//   Error: Whoops, something bad happened
//     at [eval]:5:15
//     at Script.runInThisContext (node:vm:132:18)
//     at Object.runInThisContext (node:vm:309:38)
//     at node:internal/process/execution:77:19
//     at [eval]-wrapper:6:22
//     at evalScript (node:internal/process/execution:76:60)
//     at node:internal/main/eval_string:23:3

const name = 'Will Robinson';
console.warn(`Danger ${name}! Danger!`);
// Prints: Danger Will Robinson! Danger!, to stderr
Example using the Console class:
const out = getStreamSomehow();
const err = getStreamSomehow();
const myConsole = new console.Console(out, err);

myConsole.log('hello world');
// Prints: hello world, to out
myConsole.log('hello %s', 'world');
// Prints: hello world, to out
myConsole.error(new Error('Whoops, something bad happened'));
// Prints: [Error: Whoops, something bad happened], to err

const name = 'Will Robinson';
myConsole.warn(`Danger ${name}! Danger!`);
// Prints: Danger Will Robinson! Danger!, to err
@seesourceconsole.Console.log(message?: any, ...optionalParams: any[]): void (+1 overload)Prints to stdout with newline. Multiple arguments can be passed, with the
first used as the primary message and all additional used as substitution
values similar to printf(3)
(the arguments are all passed to util.format()).
const count = 5;
console.log('count: %d', count);
// Prints: count: 5, to stdout
console.log('count:', count);
// Prints: count: 5, to stdout
See util.format() for more information.
@sincev0.1.100log(`${let i: numberi} is even`);

Instead, I’m claiming that we should favour languages and patterns that allow us to naturally write less code.

Yes, I’m talking about Svelte

Reducing the amount of code you have to write is an explicit goal of Svelte. To illustrate, let’s look at a very simple component implemented in React, Vue and Svelte. First, the Svelte version:

How would we build this in React? It would probably look something like this:

import import ReactReact, { import useStateuseState } from 'react';

export default () => {
	const [const a: anya, const setA: anysetA] = import useStateuseState(1);
	const [const b: anyb, const setB: anysetB] = import useStateuseState(2);

	function function (local function) handleChangeA(event: any): voidhandleChangeA(event: anyevent) {
		const setA: anysetA(+event: anyevent.target.value);
	}

	function function (local function) handleChangeB(event: any): voidhandleChangeB(event: anyevent) {
		const setB: anysetB(+event: anyevent.target.value);
	}

	return (
		<type div = /*unresolved*/ anydiv>
			<type input = /*unresolved*/ anyinput type="number" value={a: anya} onChange={handleChangeA: (event: any) => voidhandleChangeA} />
			<type input = /*unresolved*/ anyinput type="number" value={b: anyb} onChange={handleChangeB: (event: any) => voidhandleChangeB} />

			<type p = /*unresolved*/ anyp>
				{a: anya} + {b: anyb} = {const a: anya + const b: anyb}
			</p>
		</div>
	);
};

Here’s an equivalent component in Vue:

<template>
	<div>
		<input type="number" v-model.number="a">
		<input type="number" v-model.number="b">

		<p>{{a}} + {{b}} = {{a + b}}</p>
	</div>
</template>

<script>
	export default {
		data: function() {
			return {
				a: 1,
				b: 2
			};
		}
	};
</script>

In other words, it takes 442 characters in React, and 263 characters in Vue, to achieve something that takes 145 characters in Svelte. The React version is literally three times larger!

It’s unusual for the difference to be quite so obvious — in my experience, a React component is typically around 40% larger than its Svelte equivalent. Let’s look at the features of Svelte’s design that enable you to express ideas more concisely:

Top-level elements

In Svelte, a component can have as many top-level elements as you like. In React and Vue, a component must have a single top-level element — in React’s case, trying to return two top-level elements from a component function would result in syntactically invalid code. (You can use a fragment — <> — instead of a <div>, but it’s the same basic idea, and still results in an extra level of indentation).

In Vue, your markup must be wrapped in a <template> element, which I’d argue is redundant.

Bindings

In React, we have to respond to input events ourselves:

function function handleChangeA(event: any): voidhandleChangeA(event: anyevent) {
	setA(+event: anyevent.target.value);
}

This isn’t just boring plumbing that takes up extra space on the screen, it’s also extra surface area for bugs. Conceptually, the value of the input is bound to the value of a and vice versa, but that relationship isn’t cleanly expressed — instead we have two tightly-coupled but physically separate chunks of code (the event handler and the value={a} prop). Not only that, but we have to remember to coerce the string value with the + operator, otherwise 2 + 2 will equal 22 instead of 4.

Like Svelte, Vue does have a way of expressing the binding — the v-model attribute, though again we have to be careful to use v-model.number even though it’s a numeric input.

State

In Svelte, you update local component state with an assignment operator:

let let count: numbercount = 0;

function function increment(): voidincrement() {
	let count: numbercount += 1;
}

In React, we use the useState hook:

const [const count: anycount, const setCount: anysetCount] = useState(0);

function function increment(): voidincrement() {
	const setCount: anysetCount(const count: anycount + 1);
}

This is much noisier — it expresses the exact same concept but with over 60% more characters. As you’re reading the code, you have to do that much more work to understand the author’s intent.

In Vue, meanwhile, we have a default export with a data function that returns an object literal with properties corresponding to our local state. Things like helper functions and child components can’t simply be imported and used in the template, but must instead be ‘registered’ by attaching them to the correct part of the default export.

Death to boilerplate

These are just some of the ways that Svelte helps you build user interfaces with a minimum of fuss. There are plenty of others — for example, reactive declarations ($: statements) essentially do the work of React’s useMemo, useCallback and useEffect without the boilerplate (or indeed the garbage collection overhead of creating inline functions and arrays on each state change).

How? By choosing a different set of constraints. Because Svelte is a compiler, we’re not bound to the peculiarities of JavaScript: we can design a component authoring experience, rather than having to fit it around the semantics of the language. Paradoxically, this results in more idiomatic code — for example using variables naturally rather than via proxies or hooks — while delivering significantly more performant apps.