Re: [w3ctag/design-reviews] Partition the HTTP cache (#424)

The resource decides it doesn't want extra traffic by giving cache information, the decision to track/identify the end-used is only a browser issue, and inflicting the burden of extra traffic to solve a browser internal issue is not really fair. 
What would any bandwidth-saving scheme (revalidation only, which is already extra traffic / caching retrieval time and fake delay which doesn't add traffic) leak information back to the server? After all the server already got information that client A visited already, the issue is disclosing information to another server via timing attack on retrieval.
For the deduplication, I agree it is implementation dependent and doesn't need to be in the specification.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/w3ctag/design-reviews/issues/424#issuecomment-544948825