Deprecation / Removal of Mutation Events

[mfreed7]

Request for Mozilla Position on an Emerging Web Specification

• Specification Title: Mutation Events (deprecate/remove)
• Specification or proposal URL (if available): https://w3c.github.io/uievents/#legacy-event-types
• Explainer URL (if available): https://chromestatus.com/feature/5083947249172480
• Caniuse.com URL (optional): https://caniuse.com/mutation-events
• Bugzilla URL (optional): https://bugzilla.mozilla.org/show_bug.cgi?id=749920
• Mozillians who can provide input (optional): @smaug----, @emilio

Other information

I am going to make a push to remove Mutation Events from Chromium. This will not be easy, and success is definitely not guaranteed. The timeline is >= 1 year, if things go well. It's possible we might need to fall back to "delayed" mutation events that run at microtask or CEReactions timing. It's possible we might not be able to remove all of the events.

However, whatever happens, I'm hoping to get the support of Gecko (and WebKit), which will help convince developers who own existing usage to migrate to Mutation Observer. Some ideas, if you're supportive, might include:

1. Strengthen (as appropriate) the deprecation warnings to Gecko. I've already done this for Chromium, starting in M114.
2. Plan for a joint, coordinated removal from all browsers at some date/milestone in the future.
3. Write a co-authored blog post. I'm happy to drive this, but would love feedback, edits, and a Mozilla-owned by-line.
4. Other suggestions appreciated.

[zcorpan]

This is great, thanks for driving this! I suggest position positive.

Your ideas sound reasonable to me. Maybe a reverse origin trial could make sense here to allow for a longer grace period where needed.

[smaug----]

MutationEvents have been deprecated in Gecko since 2012.
We haven't been able to remove them because the usage has been so high, but if now also other browser vendors try to remove them more actively, hopefully we can finally get rid of them.
And the ideas sound very good to me.

Thanks a ton for driving this work!

And yes, very strong positive position.

[mfreed7]

This is great, thanks for driving this! I suggest position positive.

Thanks!

Your ideas sound reasonable to me. Maybe a reverse origin trial could make sense here to allow for a longer grace period where needed.

Yep, that's definitely on my list of things to add. I did that with Web Components v0, and it was quite helpful.

MutationEvents have been deprecated in Gecko since 2012. We haven't been able to remove them because the usage has been so high, but if now also other browser vendors try to remove them more actively, hopefully we can finally get rid of them. And the ideas sound very good to me.

Great! I agree the usage is (still) quite high, but let's see what we can do about that.

Thanks a ton for driving this work!
And yes, very strong positive position.

Thanks, much appreciated!

@zcorpan or @smaug----, would one of you be willing to co-author a blog post on this topic, to show joint support? I've got one drafted, and I can share the draft with you for edits and input, if you're interested. It would be published on developer.chrome.com, but we'd show your Mozilla affiliation prominently.

[zcorpan]

Sure

[mfreed7]

very strong positive position

Awesome, thanks! I need to figure out the process for this, and see if I can line up a WebKit person also, so I'll be in touch.

[masayuki-nakano]

FYI: Some google apps still use DOM mutation event listeners if running in Firefox. So I'd be really happy if Google would fix them on this occasion.

I guess that they are in Closure:

• https://github.com/google/closure-library/blob/32ac85cb80480a8570fed1847a175d6c46e8df26/closure/goog/editor/field.js#L1051-L1074
• https://github.com/google/closure-library/blob/32ac85cb80480a8570fed1847a175d6c46e8df26/closure/goog/editor/defines.js#L14-L26

[mfreed7]

FYI: Some google apps still use DOM mutation event listeners if running in Firefox. So I'd be really happy if Google would fix them on this occasion.

Thanks for pointing that out! I actually noticed the same thing last week and was talking to someone about that, but I really appreciate the bug link with a lot more context. That'll help me. I'll pick that up and make sure it gets removed ASAP.

[mfreed7]

FYI: Some google apps still use DOM mutation event listeners if running in Firefox. So I'd be really happy if Google would fix them on this occasion.

Thanks for pointing that out! I actually noticed the same thing last week and was talking to someone about that, but I really appreciate the bug link with a lot more context. That'll help me. I'll pick that up and make sure it gets removed ASAP.

It took some time to verify nothing breaks, but we've landed a change (from this to this) to hard-code USE_MUTATION_EVENTS: false there instead of goog.userAgent.GECKO.

If you notice any other similar things in Google code, please do call them out to me and I'll work on getting them removed.