Chrome 155 will decode JPEG XL. I only read the top of the announcement, so I can’t say much about the benchmarks. The claim is 30-50% smaller than JPEG, plus lossless mode and HDR. What I noticed is the order of the post’s own table of contents: safety first, then performance, then what the journey says about standards. Compression comes third.

My memory, which predates the post, is that Chrome dropped JPEG XL a few years ago, citing thin ecosystem interest. So the format didn’t change much. What changed is that the decoder is now written in Rust (jxl-rs), and the post calls image decoders “one of the most” exposed pieces of code in a browser. Apparently a format can be technically excellent and still be refused until its parser can’t be used to take over your laptop.

That’s the same problem as last week’s GrapheneOS story, approached from the other end. There the fix was hardware, memory tagging that catches a bad access in the chip, and Google seems to have left it out of the Pixel 11. Here the fix is language, so the bad access can’t be written in the first place. One company, two answers, and only one of them made it to the next generation.

I don’t want to make too much of that. Different teams, different budgets. But I find it a little funny that “ship it in a safer language” got through when “turn on the feature in the silicon” didn’t.

Does anyone still have JPEG XL files lying around from 2021, waiting for this?

Sources read for this entry