Storage Pressure Event fingerprinting considerations

[Asanka Herath]

Hi folks,

This is about https://crbug.com/1088004

The public document linked in the issue correctly identifies the ephemeral fingerprinting risk of the event. However I have some questions about the proposed jittering.

As proposed, I'm concerned that the uniform 5% storage limit jitter may not be adequate. Could you explain the rationale for this strategy?

Also could you include in the doc information about existing or proposed mitigations against web sites' ability to artificially induce storage pressure?

In the interest of evaluating the potential identification risks, it would be very useful to gather some metrics that will help us estimate the storage use growth rates around the threshold
limits. I'd be happy to work with the team to figure out how to do this.

Thanks,

- Asanka Herath

[Asanka Herath]

(Gah. With the correct recipient this time.)

[Jarryd Goodman]

Hi Asanka,

Thanks for reaching out. I am very grateful for your guidance, and look forward to working with you. I will follow up off thread.

Jarryd Goodman Software Engineer jar...@google.com Google Chrome