Performance

What forces layout/reflow: a comprehensive list by Paul Irish

All of the below properties or methods, when requested/called in JavaScript, will trigger the browser to synchronously calculate the style and layout*. This is also called reflow or layout thrashing, and is common performance bottleneck.

Generally, all APIs that synchronously provide layout metrics will trigger forced reflow / layout. Read on for additional cases and details.

The self-fulfilling prophecy of poor mobile performance

There are plenty of stories floating around about how some organization improved performance and suddenly saw an influx of traffic from places they hadn’t expected. This is why. We build an experience that is completely [unusable] for them, and is completely invisible to our data. We create, what Kat Holmes calls, a “mismatch”. So we look at the data and think, “Well, we don’t get any of those low-end Android devices so I guess we don’t have to worry about that.” A self-fulfilling prophecy.

I’m a big advocate for ensuring you have robust performance monitoring in place. But just as important as analyzing what’s in the data, is considering what’s not in the data, and why that might be.

Brutalist Web Design

The term brutalism is often associated with Brutalist Architecture, however it can apply to other forms of construction, such as web design.

[…]

The term brutalism is derived from the French béton brut, meaning “raw concrete”. Although most brutalist buildings are made from concrete, we’re more interested in the term raw. Concrete brutalist buildings often reflect back the forms used to make them, and their overall design tends to adhere to the concept of truth to materials.

A website’s materials aren’t HTML tags, CSS, or JavaScript code. Rather, they are its content and the context in which it’s consumed. A website is for a visitor, using a browser, running on a computer [or mobile device] to read, watch, listen, or perhaps to interact. A website that embraces Brutalist Web Design is raw in its focus on content, and prioritization of the website visitor.

Brutalist Web Design is honest about what a website is and what it isn’t. A website is not a magazine, though it might have magazine-like articles. A website is not an application, although you might use it to purchase products or interact with other people. A website is not a database, although it might be driven by one.

They list the following principles:

Measure Performance with the RAIL Model

RAIL is a user-centric performance model that breaks down the user’s experience into key actions. RAIL’s goals and guidelines aim to help developers and designers ensure a good user experience for each of these actions. By laying out a structure for thinking about performance, RAIL enables designers and developers to reliably target the work that has the highest impact on user experience.

Every web app has four distinct aspects to its life cycle, and performance fits into them in different ways:

The 4 parts of the RAIL performance model: Response, Animation, Idle, and Load.

Summary

RAIL is a lens for looking at a website’s user experience as a journey composed of distinct interactions. Understand how users perceive your site in order to set performance goals with the greatest impact on user experience.

  • Focus on the user.
  • Respond to user input in under 100ms.
  • Produce a frame in under 10ms when animating or scrolling.
  • Maximize main thread idle time.
  • Load interactive content in under 5000ms.

Document.createDocumentFragment()

Creates a new empty DocumentFragment into which DOM nodes can be added to build an offscreen DOM tree.

[…]

DocumentFragments are DOM Node objects which are never part of the main DOM tree. The usual use case is to create the document fragment, append elements to the document fragment and then append the document fragment to the DOM tree. In the DOM tree, the document fragment is replaced by all its children.

Since the document fragment is in memory and not part of the main DOM tree, appending children to it does not cause page reflow (computation of element’s position and geometry). Historically, using document fragments could result in better performance.

Code language: JavaScript

var element  = document.getElementById('ul'); // assuming ul exists
var fragment = document.createDocumentFragment();
var browsers = ['Firefox', 'Chrome', 'Opera', 
    'Safari', 'Internet Explorer'];
 
browsers.forEach(function(browser) {
    var li = document.createElement('li');
    li.textContent = browser;
    fragment.appendChild(li);
});
 
element.appendChild(fragment);

Using requestIdleCallback

Many sites and apps have a lot of scripts to execute. Your JavaScript often needs to be run as soon as possible, but at the same time you don’t want it to get in the user’s way. If you send analytics data when the user is scrolling the page, or you append elements to the DOM while they happen to be tapping on the button, your web app can become unresponsive, resulting in a poor user experience.

[…]

The good news is that there’s now an API that can help: requestIdleCallback. In the same way that adopting requestAnimationFrame allowed us to schedule animations properly and maximize our chances of hitting 60fps, requestIdleCallback will schedule work when there is free time at the end of a frame, or when the user is inactive. This means that there’s an opportunity to do your work without getting in the user’s way.

[…]

Why should I use requestIdleCallback?

Scheduling non-essential work yourself is very difficult to do. It’s impossible to figure out exactly how much frame time remains because after requestAnimationFrame callbacks execute there are style calculations, layout, paint, and other browser internals that need to run. A home-rolled solution can’t account for any of those. In order to be sure that a user isn’t interacting in some way you would also need to attach listeners to every kind of interaction event (scroll, touch, click), even if you don’t need them for functionality, just so that you can be absolutely sure that the user isn’t interacting. The browser, on the other hand, knows exactly how much time is available at the end of the frame, and if the user is interacting, and so through requestIdleCallback we gain an API that allows us to make use of any spare time in the most efficient way possible.