Hey, happy to give some more information on the compatibility risk. As part of the Storage Access Header implementation, we found that, for security purposes, it was necessary to send the `Origin` header in certain cross-site requests that would not previously have had the header. We've seen some breakage occur on servers when they receive an `Origin` header that they weren't expecting, which is the compatibility risk we're anticipating. To mitigate this, we are limiting the inclusion of the "new" Origin headers to requests where we include the `Sec-Fetch-Storage-Access` header with the `inactive` value. Our hope is that since the `inactive` value primarily covers cases where a storage access permission grant exists (with the exception of ABA cases), the impact of this change in `Origin` headers will be limited to "new" servers who are less likely experience breakage when they see the header in requests they weren't previously expecting. This thread provides some more context on the risk and decisions surrounding it: https://github.com/privacycg/storage-access-headers/issues/7#issuecomment-2089764041 .
On Friday, October 4, 2024 at 10:58:12 AM UTC-4 Mike Taylor wrote: > On 10/3/24 4:05 PM, 'Sam LeDoux' via blink-dev wrote: > > Contact emails > > sle...@chromium.org, cfre...@chromium.org, joha...@chromium.org > > Explainer > > https://github.com/cfredric/storage-access-headers > > Specification > > None > > Summary > > Storage Access Headers offer an alternate way for authenticated embeds to > opt in for unpartitioned cookies. These headers indicate whether embedded > resources would like to load with permission they have already been > granted, reducing loads and latency overall for these use cases. > > > Blink component > > Blink>StorageAccessAPI > <https://bugs.chromium.org/p/chromium/issues/list?q=component:Blink%3EStorageAccessAPI> > > TAG review > > https://github.com/w3ctag/design-reviews/issues/982 > > TAG review status > > Completed > > Risks > Interoperability and Compatibility > > There is a small compatibility risk as this feature involves sending the > Origin header in new contexts. We're working to limit the new Origin > headers to be included only on "inactive" requests, in order to minimize > compat impact. > > Hey Sam, could you expand on the compat risk here? What might this break, > and why? > > > WebView application risks > > Not available on WebView > > > Goals for experimentation > > This experiment would allow us to receive and incorporate feedback on the > browser's application of the `Sec-Fetch-Storage-Access` request header, as > well as the browser's handling of the `Activate-Storage-Access` header > before the feature is fully launched. > > Experiment behavior > > This experiment would rely on an Origin Trial (OT) to test the Storage > Access Header feature. Once the OT is active for an origin, requests to the > same context that it was enabled on will include the > `Sec-Fetch-Storage-Access` header, and the browser will handle responses > with the `Activate-Storage-Access` header in accordance with the feature’s > description. > > This experiment relies on the use of a persistent OT > <https://groups.google.com/a/chromium.org/g/blink-api-owners-discuss/c/yzVKv-6Xuts/m/KRRz9RC9DgAJ>. > > Developers opting into the use of Storage Access Headers should not expect > the `Sec-Fetch-Storage-Access` request header to be included on initial > navigations to their origin as the feature will only be active after > receiving > its first OT token. Additionally, all subsequent navigations to an opted-in > origin should include the token, otherwise the browser will take this as a > signal that the origin is no longer participating in the trial. > > Ongoing technical constraints > > None > > Debuggability > > Currently best debugged via chrome://net-export logs, as Chrome DevTools > does not show the full chain of network events. We may add improved > debugging capabilities in the future. > > > Will this feature be supported on all six Blink platforms (Windows, Mac, > Linux, ChromeOS, Android, and Android WebView)? > > No. > > Supported for Mac, Windows, Linux, Chrome OS, and Android. > > Is this feature fully tested by web-platform-tests > <https://chromium.googlesource.com/chromium/src/+/main/docs/testing/web_platform_tests.md> > ? > > No > > Flag name on chrome://flags > > #storage-access-headers > > Requires code in //chrome? > > Yes > > Tracking bug > > https://issues.chromium.org/issues/332335089 > > Launch bug > > https://launch.corp.google.com/launch/4309903 > > Estimated milestones > > Experiment desktop first 130 > > Link to entry on the Chrome Platform Status > > https://chromestatus.com/feature/6146353156849664 > <https://chromestatus.com/feature/6146353156849664?gate=5226578595545088> > > Links to previous Intent discussions > > Intent to Prototype: > https://groups.google.com/a/chromium.org/g/blink-dev/c/yfxV-jLMqNg/m/NJFVBEAyAQAJ > > -- > You received this message because you are subscribed to the Google Groups > "blink-dev" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to blink-dev+...@chromium.org. > To view this discussion on the web visit > https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CABa1CXyYbxwh%3DPdnigTW80d9jez_835R1SV1bQPDjvk1ra5G4g%40mail.gmail.com > > <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CABa1CXyYbxwh%3DPdnigTW80d9jez_835R1SV1bQPDjvk1ra5G4g%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > > -- You received this message because you are subscribed to the Google Groups "blink-dev" group. To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+unsubscr...@chromium.org. To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/574d255b-95c8-418a-bee9-60dce31c89b9n%40chromium.org.
