Fingerprinting

Font fingerprinting

Also called font fingerprint, font enumeration, font list fingerprint, font metrics

Definition

Font fingerprinting is working out which fonts are installed on a device, and how it draws them, by measuring text rendered in a long list of font names, because the installed set reflects the operating system, language packs and software on that computer.

Last updated 7 November 2026

How it works

A page can’t simply ask for your font list. Instead, a script writes a test string in a candidate font with a common fallback behind it and measures the result, either as an element on the page or with the canvas measureText() method. If the width or height differs from the fallback’s, the candidate font is installed. Repeat for a few hundred names and you have a list. David Fifield and Serge Egelman showed in 2015 that the exact sizes of rendered glyphs vary enough between systems to identify browsers on their own. Chromium also has a Local Font Access API, but it asks the user for permission first.

Why it matters when you run several accounts

Fonts are tied to the real computer: Windows, macOS and Linux ship different sets, and office suites, design tools and language packs add more. Every profile on one machine shares that set. It also gives away a profile that claims another operating system, because a “Mac” that has Windows’ fonts and draws them the Windows way is not a Mac.

How SoftGlaze Browser handles it

The Fonts setting (Default or Custom) is on a profile’s Fingerprint tab. On the standard engine, a profile also adds tiny seeded noise, about a hundredth of a pixel, to text measured through the canvas, in pages and in workers, so width-comparison probes get slightly different answers per profile while layout looks the same. That does not change which fonts your computer really has. New profiles report your own operating system for that reason, and the editor warns when a profile claims a different one, because host fonts still say what the machine really is.

Sources

  1. Fingerprinting Web Users Through Font Metrics (Fifield and Egelman, 2015)Financial Cryptography and Data Security
  2. CanvasRenderingContext2D: measureText() methodMDN Web Docs
  3. Local Font Access APIMDN Web Docs

Outside sources were checked in November 2026. Browsers, standards and vendors' pages change, so details that depend on them are as of that month.

Read more on this site

See it in SoftGlaze Browser

Separate profiles, each with its own cookies, device values and proxy, on Windows. Try the app's screens in your browser first, or start the free trial; no card needed.