[blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

Contact emails ri...@chromium.org, nidhij...@chromium.org, denom...@chromium.org Explainer https://docs.google.com/document/d/1pvaMg7J5beBXD7trzHJH_MDULc_wRHLx40MFYAmjknE/edit Specification N/A (because there are no web-exposed changes) Summary This limited experiment measures how much "perv

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

On Tue, Apr 26, 2022 at 1:59 PM Daisuke Enomoto wrote: > Explainer > > https://docs.google.com/document/d/1pvaMg7J5beBXD7trzHJH_MDULc_wRHLx40MFYAmjknE/edit This document isn't public. This particular technique has been discussed before, but there's a flaw which wasn't mentioned in this email. Th

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

On 4/26/22 8:14 AM, Anne van Kesteren wrote: On Tue, Apr 26, 2022 at 1:59 PM Daisuke Enomoto wrote: Explainer https://docs.google.com/document/d/1pvaMg7J5beBXD7trzHJH_MDULc_wRHLx40MFYAmjknE/edit This document isn't public. I just checked the settings, and it should be public now. This par

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

The summary says "payload list included below" - I can't find it though... is the list included in one of the links? On Tuesday, April 26, 2022 at 4:03:11 PM UTC+3 mike...@chromium.org wrote: > On 4/26/22 8:14 AM, Anne van Kesteren wrote: > > On Tue, Apr 26, 2022 at 1:59 PM Daisuke Enomoto > w

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

What is an 'off-by-default experiment'? Is that a dev trial flag? Joe Medley | Technical Writer, Chrome DevRel | jmed...@google.com | 816-678-7195 *If an API's not documented it doesn't exist.* On Tue, Apr 26, 2022 at 4:59 AM Daisuke Enomoto wrote: > Contact emails > > ri...@chromium.org, nidh

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

> > This particular technique has been discussed before, but there's a > flaw which wasn't mentioned in this email. The idea assumes that all > end users can access the same websites and also that all end users > visit similar websites. Neither of those is a given and as such end > users that for o

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

Hi Daisuke, Can you clarify the timeline of the experiment? Would it begin in M103? I have concerns about interactions with the current double-key experiment we're running for Network State Partitioning in M101 and M102. O

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

Hi Noam, Apologies for the confusion! The list of the top 100 pervasive payloads for use in the experiment is pending internal reviews and will be shared on this thread upon approval. On Tue, Apr 26, 2022 at 11:24 PM Noam Rosenthal wrote: > The summary says "payload list included below" - I can

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

On Tue, Apr 26, 2022 at 9:22 PM Vivek Sekhar wrote: >> This particular technique has been discussed before, but there's a >> flaw which wasn't mentioned in this email. The idea assumes that all >> end users can access the same websites and also that all end users >> visit similar websites. Neither

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

On Wed, Apr 27, 2022 at 8:50 AM Anne van Kesteren wrote: > On Tue, Apr 26, 2022 at 9:22 PM Vivek Sekhar wrote: > >> This particular technique has been discussed before, but there's a > >> flaw which wasn't mentioned in this email. The idea assumes that all > >> end users can access the same webs

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

On Tue, Apr 26, 2022 at 9:22 PM Vivek Sekhar wrote: >> This particular technique has been discussed before, but there's a >> flaw which wasn't mentioned in this email. The idea assumes that all >> end users can access the same websites and also that all end users >> visit similar websites. Neither

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

> > What is an 'off-by-default experiment'? Is that a dev trial flag? Just an ordinary experiment, behind a flag which is off-by-default. So most users get the default behaviour (no single-keyed cache), except for those in the experimental group. On Wed, 27 Apr 2022 at 00:50, Joe Medley wrote:

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

LGTM to experiment Thanks for working on this! I agree with Anne's concerns about fingerprinting risks with shipping this, as well as +Alex Russell 's concerns about the risk of this devolving into "playing favorites". But these issues would onl

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

Thanks Daisuke - sounds good. I don't think we'll need to extend beyond M102 (but I probably just jinxed it...). On 4/26/22 8:50 PM, Daisuke Enomoto wrote: Hi Mike, Thank you for your question! We're targeting M103 to start the experiment. So, IIUC, it would not interact with the double-key

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

Hi Mike, Thank you for your question! We're targeting M103 to start the experiment. So, IIUC, it would not interact with the double-key experiment running through M102 unless it's extended. On Wed, Apr 27, 2022 at 5:55 AM Mike Taylor wrote: > Hi Daisuke, > > Can you clarify the timeline of the

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

Finch flag? Joe Medley | Technical Writer, Chrome DevRel | jmed...@google.com | 816-678-7195 *If an API's not documented it doesn't exist.* On Wed, Apr 27, 2022 at 2:30 AM Adam Rice wrote: > What is an 'off-by-default experiment'? Is that a dev trial flag? > > > Just an ordinary experiment, be

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

> > Finch flag? CacheTransparency (but the code is not landed yet). On Thu, 28 Apr 2022 at 03:07, Joe Medley wrote: > Finch flag? > Joe Medley | Technical Writer, Chrome DevRel | jmed...@google.com | > 816-678-7195 > *If an API's not documented it doesn't exist.* > > > On Wed, Apr 27, 2022 at

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

What is that? I'm trying to understand what this is because I need or may need to explain it in writing to the external world in a few weeks. I've never heard of a CacheTransparancy flag. Joe Medley | Technical Writer, Chrome DevRel | jmed...@google.com | 816-678-7195 *If an API's not documented

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

Hi Joe, CacheTransparancy is the feature name we use for the feature flag. As Adam mentioned, the feature is off by default. Yes, we use Finch to determine the control/experiment group on each channel (50% on canary & dev, 10% on beta, 1% on stable). On Thu, Apr 28, 2022 at 11:46 PM Joe Medley w

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

If it's a feature flag, please put milestones in the the appropriate dev trial milestone fields. Thanks Joe Medley | Technical Writer, Chrome DevRel | jmed...@google.com | 816-678-7195 *If an API's not documented it doesn't exist.* On Sun, May 1, 2022 at 5:56 PM Daisuke Enomoto wrote: > Hi Jo

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

Hi Joe, Sorry for the confusion. The feature flag is used to control the experiment through Finch. Cache Transparency feature does not expose changes for developers to try. We chose to send the I2E to open a discussion to the blink-dev audience for feedback and iterate through experiments to quant

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

Hello Apologies for the confusion! The list of the top 100 pervasive payloads for > use in the experiment is pending internal reviews and will be shared on > this thread upon approval. The list of the top 100 pervasive payloads can be found here

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

do we have any new update on this. are we going ahead with this experiment? On Wednesday, June 15, 2022 at 12:34:57 PM UTC+5:30 Daisuke Enomoto wrote: > Hello > > > Apologies for the confusion! The list of the top 100 pervasive payloads >> for use in the experiment is pending internal reviews an

Re: [blink-dev] Intent to Experiment: Load common payloads from privacy-preserving single-keyed cache

Hi Saurabh, Apologies for not communicating the latest status in the thread. We conducted experiments on Stable in Q1 and Q2, 2023 to evaluate the impact of the single-keyed cache for common payloads. The experiments did not show promising results with the key metrics, contrary to our expectations