Participants

Will Hawkins, Steven Bougon, Maxime Villancourt, Thomas Kelly, Gilles Dubuc, Ilya Grigorik, Benjamin De Kosnik, Nic Jansma, Phillipe Le Hegaret, Markus Strange, Nicolás Peña, Todd Reifsteck, Yoav Weiss

Date for next design call?

• 11AM PST on May 28th — conflict for Mozilla folks
• Going with: 1PM PST on May 29th

F2F

• June 11th @ Google Mountain View office — full day event
• Looking into adding an additional hackathon day
• Yoav will setup an agenda doc to collect ideas
• Remote? Yes, we’re planning to stream
• Guests and observers are welcome

HR-Time

Phillippe: revisiting the discussion on privacy and high resolution time

Yoav: we added a list of potential mitigations and recommendations in the spec (e.g. jitter)

… the proposal to mandate permission api, did not get traction from any implementers

Philippe: “is it OK to not expose Navigation Timing?”

Yoav: I guess, but that’s orthogonal

Yoav: We have two green implementations, all issues are closed

… can we advance to PR?

Group resolution: advance

Yoav: AI, send ping to the group with resolution (10 working days to comment)

Overall health

Yoav: lots of specs have editors that moved on

AI: Yoav+Ilya+Todd to review list of specs and see where we’re missing active editors

sendBeacon

Yoav: In CR, no open issues, but we don’t have two green implementations

Reporting + NEL

Douglas: fell behind but catching up now and we have another contributor helping, our plan is to triage in the next few weeks

Idea: could/should we have a hackathon on a day after F2F?

• Mozilla: we could attend and probably have two people
• Nicolás, Douglas: ditto

supportedEntryTypes

https://github.com/w3c/performance-timeline/issues/117

npm: open question on registration list vs a hook

… with the list you have all the entry types in performance timeline

… however, with registration hook it’s easier for implementers to see when to add entries

yoav: from a WG perspective, a registration list would imply that each new spec adding an entry would require updating performance timeline

philippe: making performance timeline a living spec is something we could look at next year

… for now, we can move the actual lxst into a note that enumerates the list

todd + philippe: nod

Example:   https://w3c.github.io/trace-context-protocols-registry/

Conclusion (npm): change the implementation, instead of registration hook, move to list. The list itself will be a working group that we’ll link to. In that note we will also include relevant flags for each entry type behavior.  

        Philippe: I’ll draft the note and start a repo.

https://github.com/w3c/server-timing/issues/61

AI: Yoav to ping Charlie about setting up review w/ Sec/Privacy group.

Yoav: two implementations, close to moving forward

… should above review block CR?

Philippe: block CR

Yoav: OK.