🧩

Why not X?

We're avid users of browser extensions, bookmarklets, and userscripts. They'll always each have their place. We're building PixieBrix to bring their power to a broader audience

ℹ️

Also see our blog post for A Brief History of Browser Extensibility

Why not Browser Extensions?

  • Development Learning Curve: creating a browser extension to extend a SaaS application has a steep learning curve: JS build systems, the extension API, messaging between contexts, SaaS page lifecycle. With PixieBrix, you just need to know programming basics, e.g., variables, CSS selectors, and regular expressions. Our goal is go even further to allow anyone to extend their applications.
  • Publishing Speed: extension stores have slow and unpredictable review times (between 1-7 days). With PixieBrix, you can publish and deploy updates immediately.
  • Security: most browser extensions are closed-source. Despite web store security reviews, there have been high profile cases of bad actors buying popular extensions and repurposing the extension to steal private data. With PixieBrix, the extension and all the bricks you install are open-source. Additionally, because PixieBrix doesn't support arbitrary code, it makes automatic security checks more feasible

Why not Userscripts?

  • Development Learning Curve: while better than browser extensions, developing userscripts that integrate with modern SaaS applications involves a steep learning curve.
  • Security: userscripts are just Javascript, and therefore can perform arbitrary behaviors including stealing your private data. PixieBrix's block and service model makes it easy to understand and control how your data is used.
  • Availability: Chrome is moving to eliminate remote code in browser extensions. This rule will prohibit userscript managers such as Tampermonkey from the Chrome Web Store.

Why not Bookmarklets?

  • Action-only: you have to click on a bookmarklet to trigger it.
  • Context: bookmarklets all live in the bookmark toolbar, so you have to hunt for the bookmarklet you want. With PixieBrix, you can add buttons and menu items to the user interface of the site, so they're always right there when you need them.
  • Limited Capabilities: bookmarklets have length limits (Firefox limits bookmarklets to 64KB) and content security policy (CSP) incompatibility. To create complex behavior with bookmarklets, you end up re-directing a user to a different service, or injecting untrusted remote code into the source page.

Why not SaaS App Marketplaces?

Many SaaS applications have introduced their own App frameworks/marketplaces. Why not use those?

  • Extensibility: integrating N services with each other requires N^2 integrations. With PixieBrix, a brick can be re-used to integrate with any number of applications.
  • Cost: with marketplaces, you often have to pay extra for additional integrations.
  • Long-Tail SaaS: indie and niche SaaS apps can't all afford to create their own app frameworks. With PixieBrix, you can add missing functionality to you favorite SaaS application.

Why not RPA?

Robot Process Automation (RPA) tools simulate user actions (typing, clicking, etc.) to navigate, fill-out forms, and scrape data. They're good at performing high-volume, low-intention, activities at scale.

  • Enhancement: RPA tools interact with the page as-is. They provide no way to enhance the functionality on a page
  • Human Judgement: RPA tools don't handle workflows where part of the task require significant human judgement. (However, many vendors are baking in BPM, form builder, and attended automation features)