Consolidate paintTime and presentationTime in performance APIs

[noamr]

Request for Mozilla Position on an Emerging Web Specification

• Specification title: Consolidate paintTime and presentationTime in performance APIs
• Specification or proposal URL (if available): https://w3c.github.io/paint-timing
• Explainer URL (if available): https://github.com/w3c/paint-timing/blob/main/presentation-timestamps.md
• Proposal author(s) (@-mention GitHub accounts): @noamr @mmocny
• Caniuse.com URL (optional):
• Bugzilla URL (optional):
• Mozillians who can provide input (optional): @BasMoz @sefeng211
• WebKit standards-position:

Other information

This has been discussed at the WebPerfWG, see w3c/paint-timing#62

[sefeng211]

Yeah, we talked this in WebPerf and we are positive about this.

Basically the changes on Firefox side will be

1. We need to implement paintTime, and this is the timestamp after Update the rendering
2. Need to change FCP's startTime to this paintTime, probably worth a developer notification because currently it's the timestamp before we do all the rAFs etc.
3. We can decide how we want to expose presentationTime, it's optional though.
4. I think LCP's renderTime will also be this paintTime?

[noamr]

Yeah, we talked this in WebPerf and we are positive about this.

Basically the changes on Firefox side will be

1. We need to implement paintTime, and this is the timestamp after Update the rendering
2. Need to change FCP's startTime to this paintTime, probably worth a developer notification because currently it's the timestamp before we do all the rAFs etc.
3. We can decide how we want to expose presentationTime, it's optional though.
4. I think LCP's renderTime will also be this paintTime?

LCP's renderTime would be the same as FCP startTime, so it would be presentationTime if it's implemented, and paintTime otherwise.