Intent to Extend Experiment: LazyEmbeds

202 kali dilihat
Langsung ke pesan pertama yang belum dibaca

Minoru Chikamune

belum dibaca,
1 Agu 2022, 06.17.0301/08/22
kepadablink-dev

Contact emails

chik...@chromium.org, sisid...@chromium.org


Explainer


Specification: https://github.com/nyaxt/lazyembeds


Summary: LazyEmbeds aims to improve LCP by delaying the iframe loads outside the viewport when all of the following conditions are met.


  • The loading=eager or loading=lazy attribute values are not specified in the iframe tag.

  • The iframe's src URL matches a pre-curated list of cross-origin embeds (for the sake of quick experimentation).

  • The page is visible.

  • The iframe's src URL must be cross origin.

  • The main frame is not loaded in the following ways. In the following operations, the user probably wants to avoid any breakage caused by LazyEmbeds.

    • The main frame is reloaded.

    • The main frame is a result after the user (re)submits a form.


LazyEmbeds has a timeout mechanism that loads all the iframes that are eligible for LazyEmbeds after a few seconds have elapsed even when the viewport doesn't come near the iframe. This timeout mechanism is the main difference between  LazyEmbeds and <iframe loading=lazy>.


Site authors can specify loading=eager on frames to opt-out of LazyEmbeds.


The current LazyEmbeds implementations target a 1% stable channel for data gathering purposes. We don't have any plan to release LazyEmbeds in its current form. This experiment is to help us assess the potential impact, in order to motivate a proper, launchable, design.


Blink component: Blink>Loader


TAG review

None


TAG review status

Not applicable


Risks

Interoperability and Compatibility

The goal of the experiment is to test if the idea actually improves page load UX. Once that is proven, we would like to start pitching in the standardization forum on having the new behavior part of the spec.


We have a rough sketch of how it would look like in terms of specification changes at https://github.com/nyaxt/lazyembeds. LazyEmbeds applies a different loading schedule to offscreen cross-origin <iframe>s. Site authors who believe offscreen content is a critical part of the user-experience may find this breaks their expectations. To restore the previous behavior, authors can specify loading=eager on those frames.


Ergonomics

Websites don’t have to do anything.


Activation

LazyEmbeds works without any developer activation.


Security and privacy

LazyEmbeds doesn't have any security and privacy concerns.


Goals for experimentation

We would like to judge if this is a good idea or not before we would like to validate our hypothesis using the performance data (e.g. Core Web Vitals) collected through the experiment before we proceed to the next steps (open-ended discussion with other vendors, involved 3rd-parties).


Reason this experiment is being extended

The previous timeline was to start an experimental 1% stable rollout in M104. But while running experiments in M104 beta, we have noticed several problems. So we would like to change the schedule. We want to restart the experiment in M105 beta and 1% stable rollout in M105.


The following are the notable differences between M104 and M105.

  • The main frame is not loaded in the following ways. In the following operations, the user probably wants to avoid any breakage caused by LazyEmbeds.

    • The main frame is reloaded.

    • The main frame is a result after the user (re)submits a form.

  • Added a timeout mechanism that loads all the iframes that are eligible for LazyEmbeds after a few seconds have elapsed even when the viewport doesn't come near the iframe. This timeout mechanism is intended to minimize the risk of breakage caused by LazyEmbeds.

  • Expanded the pre-curated list of cross-origin embeds.


Any risks when the experiment finishes?

No.


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

Yes.


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

No, through the experiment phase, but if the idea proves to be useful through the experiment and makes it to the spec, we will add WPTs to cover them.


Tracking bug

https://crbug.com/1247131


Launch bug

https://crbug.com/1247130


Links to previous Intent discussions

N/A


Link to entry on the feature dashboard

Not yet.


Links to previous Intent discussions

Intent to Experiment


Yoav Weiss

belum dibaca,
1 Agu 2022, 10.31.4301/08/22
kepadaMinoru Chikamune, blink-dev
A change of schedule doesn't really qualify as an experiment extension. My LGTM from the previous intent still stands.

--
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/CAEy%2BVDL14VfnmGO2Nv_MmpAY064CiP7jAGhJxnCs6ERVbH%3DhSA%40mail.gmail.com.

Joe Medley

belum dibaca,
1 Agu 2022, 13.50.2901/08/22
kepadaYoav Weiss, Minoru Chikamune, blink-dev
Where's the status entry for this?
Joe Medley | Technical Writer, Chrome DevRel | jme...@google.com | 816-678-7195
If an API's not documented it doesn't exist.


Minoru Chikamune

belum dibaca,
1 Agu 2022, 21.07.3301/08/22
kepadaJoe Medley, Yoav Weiss, blink-dev
> A change of schedule doesn't really qualify as an experiment extension. My LGTM from the previous intent still stands.

Oh, I see.

> Where's the status entry for this?

I'll create a chrome status entry. Thank you for pointing this out.

Minoru Chikamune

belum dibaca,
2 Agu 2022, 01.06.4102/08/22
kepadaJoe Medley, Yoav Weiss, blink-dev
>> Where's the status entry for this?
> I'll create a chrome status entry. Thank you for pointing this out.

I've created a chrome status entry.

Joe Medley

belum dibaca,
2 Agu 2022, 10.26.5202/08/22
kepadaMinoru Chikamune, Yoav Weiss, blink-dev
Thanks! 

Joe Medley | Technical Writer, Chrome DevRel | jme...@google.com | 816-678-7195
If an API's not documented it doesn't exist.

Balas ke semua
Balas ke penulis
Teruskan
0 pesan baru