It’s been four weeks since Chrome 99 was released, which means it’s time for Chrome 100 to enter the stable channel. Aside from being an important milestone in itself, this is a critical update, as it could break the experience for some sites when parsing the user-agent string. While Google has implemented some safeguards, the technical issues posed by three-digit version numbers still cause concern.
In addition to this, Chrome 100 also includes a ton of other changes, which you can read about below:
First, Chrome 100 is updating the way it parses cookie strings, allowing the domain attribute to be set to an empty string. This change will bring Chrome into line with the standards specification and also improve interoperability with Safari and Firefox, which already handle empty strings correctly.
The multi-screen window placement API is being enhanced to cater to modern use cases, providing more information about the secondary screen than just tied to the primary display.
Google says this will unlock the following scenarios in terms of accurate window placement:
- A slideshow application that presents on a projector while displaying the speaker’s notes on the laptop screen.
- A financial application opens a dashboard of one window on multiple monitors.
- A medical application opens an image (eg, X-ray) on a high-resolution grayscale display.
- A creative app displays secondary windows (such as palettes) on a separate screen.
- Multi-screen layouts in games, signage, art, and other types of applications.
Another interesting feature in Chrome 100 is that websites can now use a new method to voluntarily forget linked human interface devices (HIDs). This means that websites that use network Bluetooth and WebUSB standards to connect peripherals can revoke this permission if they no longer need it.
Chrome 100 also introduced a digital goods API. This will enable web apps in the Play Store to accept digital purchases. This essentially wraps the Android Play Billing API and enables web apps that offer digital purchases to be installed from the Play Store.
Other relatively minor features include capability delegation so that a framework can offload the ability to call restricted APIs to trusted sub-frameworks, enhanced mixed-mode properties, and better handling of errors with AbortSignal objects, by hashing instead of Relies on Public Key Infrastructure (PKI) to authenticate WebTransport servers, and a Web NFC method that enables developers to permanently make NFC tags read-only.
Finally, Chrome also has an integration between AbortSignal and SerialPort objects, minor tweaks to WebSockets, and some compatibility fixes to reduce user-agent strings. Chrome 100 was the last version of the browser to support unreduced user agent strings. Developers can try it out through Origin until April 19, 2022.
Website developers who need more time can enroll their sites in trials from Chrome 100 to Chrome 113 (inclusive). This means they can continue to use legacy user-agent strings until May 2023, before migrating to the user-agent client prompt API. You can find more details here, and also read more about all the new Chrome 100 DevTools here:
https://blog.chromium.org/2021/09/user-agent-reduction-origin-trial-and-dates.html
Chrome 100 will begin rolling out later today. If it doesn’t automatically update to version 100, go to “Help” > “About Google Chrome” to trigger the update when it becomes available. Next up is Chrome 101, which hits the beta channel on March 31st and will hit the stable release on April 26th.