[quota-api] how should file de-duplication be represented? (#1)

Moved from https://github.com/kinu/quota-api/issues/11:

by @wanderview
"I understand blink plans to add file de-duplication for their Cache API implementation. My impression is they are planning to de-duplicate across both Cache API origins and the http cache, but I'm not sure.

How should the quota system account for de-duplicated files?

What happens if an origin thinks they are within their quota, but then the http cache ages out a de-duped file causing it to be solely owned by that origin? Does its usage suddenly go up and trigger further quota eviction?"

---
Reply to this email directly or view it on GitHub:
https://github.com/w3c/quota-api/issues/1