Re: [blink-dev] Intent to Experiment: Cache sharing for extremely-pervasive scripts

Sorry if the doc didn't make it clear - the "attesting" that the doc envisions isn't expected to be an API or anything in code - it's expected to be something along the lines of a legal agreement or at least manual verification and confirmation by the owners (process - whatever legal and privacy th

Re: [blink-dev] Intent to Experiment: Cache sharing for extremely-pervasive scripts

No developer-facing changes implies no approvals[1], but you will need a mechanical API OWNER approval in the chromestatus entry to ship this change (and run the experiment). AIUI, the design doc mentions sites attesting to benefit from this feature, so there will be a new API involved at some

Re: [blink-dev] Intent to Experiment: Cache sharing for extremely-pervasive scripts

Experiment candidates in Design docs only includes - google-analytics.com - googlesyndication.com - doubleclick.net - googletagmanager.com - googleapis.com - youtube.com - etc Even if this list were based on a neutral data analysis, it can't help thinking that Chrome is inte

Re: [blink-dev] Intent to Experiment: Cache sharing for extremely-pervasive scripts

Is there a reason to change the chromestatus type? Presumably we'd want to be able to experiment with things that are internal and not developer-facing (and developers won't directly see any change from this). A lot of the Networking internals tend to fall into that bucket and it'd be nice (at leas

Re: [blink-dev] Intent to Experiment: Cache sharing for extremely-pervasive scripts

Yeah, probably. Jason, are you able to do some backend magic to convert the chromestatus type? On 4/30/25 10:33 AM, Patrick Meenan wrote: I'm hoping for 138 (code is in-flight now). I added 138 to the "dev trials section" but there doesn't appear to be any way to "experiment". The only stages

Re: [blink-dev] Intent to Experiment: Cache sharing for extremely-pervasive scripts

As Mike mentioned, the design doc links to the spreadsheet with the full list of candidate URLs t

Re: [blink-dev] Intent to Experiment: Cache sharing for extremely-pervasive scripts

I'm hoping for 138 (code is in-flight now). I added 138 to the "dev trials section" but there doesn't appear to be any way to "experiment". The only stages that come after (or can be added) are shipping. Maybe because it's "No developer-facing change" for the feature type? On Wed, Apr 30, 2025 a

Re: [blink-dev] Intent to Experiment: Cache sharing for extremely-pervasive scripts

The design doc also links to https://docs.google.com/spreadsheets/d/1hW-xXj9zEdkTJMVE4b4qcflB00WxsOG72Y0LVtSjcWI/edit?gid=1602510613#gid=1602510613, and an HTTP archive query to produce it. On 4/30/25 5:07 AM, Jxck wrote: Experiment candidates in Design docs only includes * google-analytics

Re: [blink-dev] Intent to Experiment: Cache sharing for extremely-pervasive scripts

Thanks Patrick. It will be interesting to learn how the experiment goes, and any possible improvements, especially if resources are changing hourly. I look forward to learning more about how sites will attest to always return the same payload, if the experiment proves successful. One last que

Re: [blink-dev] Intent to Experiment: Cache sharing for extremely-pervasive scripts

Personally I'm not a fan of this idea of giving special performance boosts to big properties like YouTube or Google Ads. That's creating an unfair advantage over smaller players in the market. And I have to point out that Google is currently under antitrust scrutiny for allegedly doing such. On Tu

Re: [blink-dev] Intent to Experiment: Cache sharing for extremely-pervasive scripts

The previous experiment required an exact match of a full URL and payload hash (on a path to automate allow-listing responses). Unfortunately, most of the interesting third-party resources are updated quite regularly (usually at least daily if not every few hours) so the URL list and payload hashes

Re: [blink-dev] Intent to Experiment: Cache sharing for extremely-pervasive scripts

Hey Patrick, This sounds related to a previous experiment (https://groups.google.com/a/chromium.org/g/blink-dev/c/9xWJK3IgJb4/m/SYLfUccOBgAJ) - could you describe how this experiment is similar or different (or perhaps builds upon the results from the previous experiment)? thanks, Mike On 4