3 ms·

FTR there is a WebP decoder implementation in safe Rust in the image crate: https://github.com/image-rs/image https://github.com/image-rs/image It used to be q

by est31 3y ago

FTR there is a WebP decoder implementation in safe Rust in the image crate: https://github.com/image-rs/image https://github.com/image-rs/image It used to be quite incomplete for a long time, but work last year has implemented many webp features. Chromium now has a policy of allowing the use of Rust dependencies, so maybe Chromium could start adopting it?

insanitybit 3y ago
If Chrome had a flag for "use the safe versions of XYZ libraries, even if they might be slower" I would absolutely flip that switch and take the performance hit.
- Lt_Riza_Hawkeye 3y ago
  Right, but for most users, Chrome's differentiator is that it's fast. I can't imagine them flipping the flag on by default - and if 99.99% of users aren't going to use it, it's probably not worth it to bloat the binary size either. Maybe someone will make a chromium patch with it that people can keep rebasing, though
  - insanitybit 3y ago
    > Chrome's differentiator is that it's fast. I don't think that's the whole story. Chrome aggressively marketed itself around its sandboxing and security at a time when browser drive bys were huge. Anyone I recommended Chrome to, I did so because of security, not performance. > if 99.99% of users aren't going to use it, Ideally the crate would get closer and closer to the performance required to enable it entirely. I just meant as a temporary state. Also, I think more than .01% of users would enable it if there were a checkbox on install like "We can make your browser a tiny bit slower but potentially increase its security". > it's probably not worth it to bloat the binary size either. eh, there's so much in `chrome://flags` I don't think that they mind too much? idk
NelsonMinar 3y ago
It's crazy to me in 2023 we're still writing new C/C++ code for something with as enormous an attack surface as a web browser.

by est31 3y ago

insanitybit 3y ago

If Chrome had a flag for "use the safe versions of XYZ libraries, even if they might be slower" I would absolutely flip that switch and take the performance hit.

Lt_Riza_Hawkeye 3y ago
Right, but for most users, Chrome's differentiator is that it's fast. I can't imagine them flipping the flag on by default - and if 99.99% of users aren't going to use it, it's probably not worth it to bloat the binary size either. Maybe someone will make a chromium patch with it that people can keep rebasing, though
- insanitybit 3y ago
  > Chrome's differentiator is that it's fast. I don't think that's the whole story. Chrome aggressively marketed itself around its sandboxing and security at a time when browser drive bys were huge. Anyone I recommended Chrome to, I did so because of security, not performance. > if 99.99% of users aren't going to use it, Ideally the crate would get closer and closer to the performance required to enable it entirely. I just meant as a temporary state. Also, I think more than .01% of users would enable it if there were a checkbox on install like "We can make your browser a tiny bit slower but potentially increase its security". > it's probably not worth it to bloat the binary size either. eh, there's so much in `chrome://flags` I don't think that they mind too much? idk

NelsonMinar 3y ago

It's crazy to me in 2023 we're still writing new C/C++ code for something with as enormous an attack surface as a web browser.