July 2019

Browser extensions are underrated: the promise of hackable software

Lego bricks
Photo by Rick Mason on Unsplash

Recent conversations about web browser extensions have focused on controversy: malicious browser extensions capturing web history, and Google limiting the capabilities used by ad blockers. These are important discussions, but we shouldn’t lose sight of the big picture: browser extensions are a special ecosystem worth celebrating.

Among major software platforms today, browser extensions are the rare exception that allow and encourage users to modify the apps that we use, in creative ways not intended by their original developers. On smartphone and desktop platforms, this sort of behavior ranges from unusual to impossible, but in the browser it’s an everyday activity.

Browser extensions remind us what it’s like to have deep control over how we use our computers.

Assembling our own software

Once a software platform reaches a certain level of openness, it can fundamentally change the way that normal users relate to their software. By installing four different Gmail extensions that modify everything from the visual design to the core functionality, in some sense, I’ve put together my own email client. Instead of being a passive user of pre-built applications, I can start assembling my own personalized way of using my computer.

The popularity of browser extensions proves that many people are interested in customizing their software, and it’s not just a hobby for power users. There are over 180,000 extensions on the Chrome store, and nearly half of all Chrome users have browser extensions installed.1 When people have an easy way to extend their software with useful functionality, they apparently take advantage.

Hackable platforms, not custom APIs

Browser extensions have remarkably broad use cases. I personally use Chrome extensions that fill in my passwords, help me read Japanese kanji, simplify the visual design of Gmail, let me highlight and annotate articles, save articles for later reading, play videos at 2x speed, and of course, block ads.

The key to this breadth is that most extensions modify applications in ways that the original developers didn’t specifically plan for. When Japanese newspapers publish articles, they’re not thinking about compatibility with the kanji reading extension. Extension authors gain creative freedom because they don’t need to use application-specific APIs that reflect the original developers’ view of how people might want to extend their application.

The web platform has a few qualities that enable this sort of unplanned extensibility. The foundational one is that the classic web deployment style is to ship all the client code to the browser in human-readable form. (Source maps are a key to preserving this advantage as we ship more code that’s minified or compiled from other languages.) The web’s layout model also promotes extensibility by encouraging standardized semantic markup—my password manager extension works because web pages reliably use form tags for password submissions instead of building their own version.

Even with these advantages, it can still require clever tricks to modify a site in ways that it wasn’t built for. But it’s often a reasonable amount of work, not a years-long reverse engineering effort. The sheer variety of extensions available shows that extension authors are willing to jump through a few hoops to create useful software.

Occasionally there are tensions between website developers and extension authors, but it seems far more common that developers are fine with their sites being extended in creative ways, as long as they don’t have to do any extra work. Extensions can even make life easier for application developers: if there’s a niche request that a small minority of users want, a motivated community member can just build an extension to support it. By building on a hackable platform, developers allow their users to get even more value out of their applications.

Small tools, not big apps

Many browser extensions are generic tools designed to enhance my use of all websites. I can use my annotation extension on every website everywhere, instead of needing a different highlighting tool for each article I read. Just like using a physical highlighter with paper articles, I can master the tool once, and get a lot of leverage by applying it in different contexts.

In many software platforms, we think of the operating system as providing the cross-cutting tools, and third parties as providing standalone “apps” that are used in isolation. With browser extensions, third parties are also adding tools; a single piece of software has the leverage to change my experience across all the apps I use.

When software is built in small units, it also changes the economics. Most extensions I use are free, and are perhaps too small in their feature set to support a full-blown business. And yet, people still choose to make them, and I benefit immensely from these little bits of software. Browsing the extension store feels more like going to a local flea market than going to a supermarket. Massive software built by huge companies isn’t the only way.

The origins of openness

It’s not an accident that this openness emerged on the web platform.

Since the beginning of personal computing, there’s been a philosophical tradition that encourages using computers as an interactive medium where people contribute their own ideas and build their own tools—authorship over consumption. This idea is reflected in systems like Smalltalk, Hypercard, and more recently, Dynamicland.

When Tim Berners-Lee created the World Wide Web, he imagined it fitting into this tradition. “My vision was a system in which sharing what you knew or thought should be as easy as learning what someone else knew.”2 There were some hiccups along the way3, but eventually that vision largely won out, and the Web became a place where anyone can publish their opinions or photos through social media platforms.

Still, there’s a catch. When you’re using Facebook, you’re operating within a confined experience. You’re forced to publish in a certain format, and to use their app in a certain way (that includes, of course, seeing all the ads). There’s more room for authorship than just browsing a news website, but only within the strict lines the app has painted for you.

Browser extensions offer a deeper type of control. Instead of merely typing into the provided text box, we can color outside the lines and deeply modify the way we use any application on the web. Browser extensions offer a kind of decentralization: large companies building major websites don’t get to dictate all the details of our experience.

Improving on extensions

We clearly need to work on protecting people from malicious extensions that invade their privacy. But beyond that, here are some bigger picture opportunities I see for improving on extensions:

Accessibility: Today, it requires a big jump to go from using browser extensions to creating them: you need to learn a fair amount of web development to get started, and you can’t easily develop extensions in the browser itself. What if there were a quick way to get started developing and sharing extensions in the browser? You could imagine smoothly transitioning from editing a website in the developer tools to publishing a small extension.

Compatibility: Because extensions hook into websites in unsupported ways, updates to websites often result in extensions temporarily breaking, and extension authors scrambling to fix them. Can we make it easier for website developers and extension authors to form stable connections between their software, without necessarily resorting to using explicit extension APIs?

There are existing practices that fit into this category already—for example, using clean semantic markup, human-readable CSS, and source maps makes it easier to develop an extension.

A simple change that would allow for more stable extensions would be to give users more control over when they upgrade to new versions of cloud software. If I have a 3 month window to continue using an old version after the new one is released, that would give extension authors more time to upgrade their software for the new version.

Power: Web extensions are limited in their power by the typical architecture of web applications: they have broad rights to modify the browser client, but the server is off limits. For example, if my social media app’s server only provides an endpoint for querying my posts in chronological order, no browser extension can ever search through all my posts by keyword. How could we rethink the client-server boundary to enable extensions to make even deeper modifications?

This raises tough questions around security and privacy. The modern browser extension API has done a good job balancing extensibility with security, and yet we’re still grappling with the consequences of browser extensions invading people’s privacy. Giving extensions more power would raise the stakes further. Still, we shouldn’t give up in the name of security—we should fight for extensibility as a value and find ways to balance these interests.

The next platform

I’m intrigued by a couple projects that are rethinking the web in ways that might make it more extensible:

The Beaker Browser and the decentralized web community are exploring how the web works without centralized servers. It seems like their proposed architecture would give users fuller control over modifying the “server” side of web applications.

Tim Berners-Lee is working on a new project called SOLID. I don’t yet understand precisely what they’re up to, but given Tim’s involvement I figure it’s worth paying attention. A key principle is giving users more ownership over their data, which would enable people to use extensions and other software to manipulate their data in flexible ways beyond what application server APIs allow.

Computing is still young, and platforms are changing quickly. Modern browser extensions and smartphone platforms have only been around for about a decade. These platforms will evolve, and there will be new platforms after them, and we will get to collectively decide how open they will be.

Browser extensions give us one example to strive for: a place where we routinely hack the software we use and make it our own.

2024 Updates

I originally wrote this post in 2019. Coming back to this five years later, here are a few related projects which you might find interesting:

  • I built a popular browser extension for Twitter, and wrote some reflections on using extensions as a way to fix problems in the software you use every day. I also touch on some of the downsides of the extensions platform in that post, if this one was too optimistic for your taste 😉
  • During my PhD at MIT with Daniel Jackson, I developed a tool called Wildcard that enables non-programmers to build a browser extension in a spreadsheet.
  • My friend Glen has also been working on a neat platform called ExtensionPay for monetizing browser extensions, for anyone interested in making an extension into a sustainable project.

More recently, I’m working towards malleable software powered by AI at the research lab Ink & Switch. If you’d like to follow that work you can sign up for my email newsletter.

Discuss on Hacker News (2019)

Discuss on Hacker News (2024)

  1. https://www.blog.google/technology/safety-security/update-project-strobe-new-policies-chrome-and-drive/ 

  2. Weaving the Web, by Tim Berners-Lee (p33) 

  3. Tim thought web browsers should also be website editors, and was disappointed when the Mosaic browser took off in popularity without including that feature.