Intent to Implement & Ship: User-Agent Reduction Phase 5 (platform and OsCpu reduction)

[Victor Tan]

Contact emails

mike...@chromium.org, vict...@chromium.org

Explainer

https://github.com/WICG/ua-client-hints#explainer-reducing-user-agent-granularity

Specification

https://www.chromium.org/updates/ua-reduction is the closest thing that specifies Chrome’s UA Reduction plans today. As these changes land in Chromium and ship to 100% of stable, the Compat Standard will be updated in the UA String section, like we did for the Phase 4 changes.

Summary

As previously detailed on the Chromium Blog, we intend to proceed with Phase 5 of the User-Agent Reduction plan. In Phase 5, the User-Agent string changes the platform and oscpu tokens from their platform-defined values to the relevant unifiedPlatform token value. The `navigator.platform`, `navigator.platform`, and `navigator.appVersion` JS APIs will be similarly reduced.

Blink component

Blink>Network>ClientHints

TAG review

https://github.com/w3ctag/design-reviews/issues/640

TAG review status

Closed with concerns.

Risks

Interoperability and Compatibility

Any time you modify the User-Agent string there is a risk of some content somewhere depending on the previous format.

We do not expect interop risks, as each browser sends its own User-Agent string format. But there is a risk, especially on legacy Windows platforms, that content somewhere is relying on User-Agents to parse platform and oscpu information. We believe the risk is somewhat low. But in order to mitigate the risk of this change, we intend to slowly roll it out via Finch creating two sub groups: one group enabling the feature for all platforms except legacy Windows platforms, another group enabling the feature on legacy Windows platforms and observing health metrics and bug reports. This gives us the option to roll this back specifically for legacy Windows clients if needed, but proceed for other platforms.

Displaying a modern OS version for legacy clients will not create a problem syntactically on legacy Windows platforms. But the web can get pretty weird in ways we don't anticipate, hence the slow roll-out and incremental path towards User-Agent Reduction.

Here is our proposed rollout plan, with the understanding that if we discover concerning breakage or regressions via health metrics or bug reports we will pause the rollout or roll back the feature entirely (and update this thread if so):

Stage	Time	Date
Stable 1% (M107+) Canary/Dev/Beta 100%	M107 stable release is shipping to 100% (a best guess)	Nov 1, 2022
Stable 10% (M107/M108/M109)	~10 weeks after previous stage	Jan 9, 2022
Stable 50% (M107/M108/M109)	~2 weeks	Jan 23, 2022
TOT Default (M111)	~2 weeks after previous stage	Feb 7, 2022
Stable 100% (M107=>M111)	~ Same business day as previous stage	Feb 7, 2022

Gecko: Shipped/Shipping. Firefox has frozen (or capped) much of their UA string already.

WebKit: Shipped/Shipping. Safari has already frozen everything in their desktop UA string except for Safari and WebKit versions.

Web developers: Mixed signals. Reactions have ranged from positive to indifferent to negative, from various channels.

Debuggability

No special DevTools support needed.

Will this feature be supported on all six Blink platforms (Windows, Mac, Linux, Chrome OS, Android, and Android WebView)?

No (Only for desktop platforms: Windows, Mac, Linux, Chrome OS)

Is this feature fully tested by web-platform-tests?

No

Flag name

#reduce-user-agent-platform-oscpu

Tracking bug

https://bugs.chromium.org/p/chromium/issues/detail?id=1339834 

Launch bug

https://bugs.chromium.org/p/chromium/issues/detail?id=1346071  

Link to entry on the Chrome Platform Status

https://chromestatus.com/feature/5098783126323200

[Chris Harrelson]

LGTM1. If any issues come up during this rollout that affect the plan, please bring them back to this thread for our consideration.

--
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/CAJh4P7FysJS2D2i4G3YNkf6X8N4_eyUcP%2BpoFrUB_ikdiATwuA%40mail.gmail.com.

[Daniel Bratell]

LGTM2

/Daniel

To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOMQ%2Bw9fGf_85Y4oFArSbBHmFYFEnZ%2BozUaRUvJqFJork%2BadcQ%40mail.gmail.com.

[Yoav Weiss]

LGTM3

To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/3b26fcb5-2bc9-56d9-8fa2-6918d703bb7e%40gmail.com.

[Victor Tan]

Thanks all. I just realized there are some date typos for the roll out plan. Update as follows: 

Stage	Time	Date
Stable 1% (M107+) Canary/Dev/Beta 100%	M107 stable release is shipping to 100% (a best guess)	Nov 1, 2022

Stable 10% (M107/M108/M109)

~10 weeks after previous stage

Jan 9, 2023

Stable 50% (M107/M108/M109)

~2 weeks

Jan 23, 2023

TOT Default (M111)

~2 weeks after previous stage

Feb 7, 2023

Stable 100% (M107=>M111)

~ Same business day as previous stage

Feb 7, 2023

Bests,

Victor

[Mike Taylor]

Howdy blink-dev,

Phase 5 of UA Reduction is currently ramping up to 1% of the stable release population. No changes will occur between now and January 9th, pending site compatibility feedback.

thanks,
Mike

To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAJh4P7EsOyYro%2BQ0a2VvWwv4SDDB4CCeaCrW8wfrCP4oMiVhgw%40mail.gmail.com.

[Ziling Zhao]

Hello,

YouTube -- and various other Google properties -- have been analyzing the results of this 1% experiment. We're concerned that the 1% may not provide enough data especially due to the amount of slicing (modern chrome, non windows vs. legacy windows, etc.). On YT, we are seeing significant metrics impact on, and given the holidays, there's a short amount of time available for us to iterate and debug issues before we rollout to 10%.

We'd like to propose:

• An intermediate state of 5% on January 9th rather than jumping directly to 10%.
• The 5% should hold for at least a month, enough for us to do a few rounds of experimental analysis and debugging.

Thanks.

[Mike Taylor]

Hi Ziling,

(forgive the delay in replying, I took some time off and managed to not check my email)

Based on the other feedback we've received both internally and externally (and the lack of any negative impact thus far), I don't think we want to add an additional month to our rollout schedule before hitting 100%. Another possible option would be to move to 5% sooner, at least 1 month ahead of Jan 9th, ideally with enough time to observe metrics ahead of the Finch freeze on Dec 16th, in case something scary happens that would warrant a rollback.

Would that work for YouTube?

thanks,
Mike

[Ziling Zhao]

Hey Mike,

I think moving to 5% early sounds good for YouTube. The 1% doesn't seem to be giving us consistent data and we would want to get more traffic as soon as possible. If we were to move forward, would this re-roll the experiment or just expand? How soon would we be able to roll this out?

Thanks!

--

~ Ziling Zhao

[Mike Taylor]

Thanks Ziling. We're unlikely to make the change just ahead of the weekend, but will ping this thread early next week once the move to 5% is submitted.

[Victor Tan]

Hi blink-dev,

UA Reduction Phase 5  is currently ramping up to 5% of the stable release population. We will monitor the change has any impacts. 

Thanks.

Victor

[Victor Tan]

Hi blink-dev,

UA Reduction Phase 5  is currently ramping up to 10% of the stable release population. We will monitor the change has any impacts. 

Thanks.

Victor

[Victor Tan]

Hi blink-dev,

UA Reduction Phase 5  is currently ramping up to 50% of the stable release population. 

Thanks.

Victor

[Victor Tan]

Hi blink-dev,

UA Reduction Phase 5  is currently ramping up to 100% of the stable release population. 

Thanks.

Victor