Agenda for Nov 9, 2023

[nairnandu]

Proposed agenda for the Interop team meeting on Nov. 9, 2023

• Review and finalize Interop 2024 proposal selection (round 3 process)
• Review and land #102

[foolip]

I'd also like to discuss https://github.com/web-platform-tests/interop/pull/102, call for final review so we can merge it.

[nairnandu]

Attendees: @bkardell, @gsnedders, @meyerweb, @fantasai, @nairnandu, @boazsender, @chrishtr, @foolip, @jensimmons, @jgraham, @dandclark, @zcorpan, @rnyman, @tantek

Notes

• nairnandu: Ranking of proposals is the next step
  • jensimmons: there are a couple of issues in the current process. Would like to discuss the scope of the effort. We have a desire to make the scope go back to earlier year (smaller scope). The impact is that high priority engineering features get knocked off because of Interop work. Are we all in agreement that we want to reduce the scope?
  • foolip: try not to use the same number twice in the ranking. I dont think we can totally agree on a fixed size or scope for every year. Its too early to arrive at a cutoff line since we have not ranked the proposals yet and hence we do not know the effort involved.
  • jgraham: the process requires us to have consensus. Similar feedback in Mozilla as well that the size of Interop 2022 was more manageable. We need to be mindful that every organization’s experience would be different based on the scope each year.
  • bkardell: in the abstract it seems like we could roughly say more or less than last year though? Or about the same? Doesn't it?
  • jensimmons: would like to lock down the scope today since the process would be very different based on that
  • foolip: without seeing the outcome of the ranking, we cannot really determine effort
  • chrishtr: this effort is very high value to developers. The more we can do, the better
  • jgraham: to some extent, agree that it doesn't make sense to set a firm scope today. With that being said, we should pick the highest value work for next year. That should be done along with the carryover proposals
  • bkardell: lets aim for a static number as a start to see where we get
  • jensimmons: sounds like Mozilla and Apple are in agreement that the scope should be lesser than last year (Interop 2023) and Google would like for it to be the same. If we are only picking the things that have some sense of urgency, that should be okay and its a good thing. The project being large might be detrimental.
  • bkardell: It seems to me (speaking as an individual, on behalf of no one else) like we should work on initial 'rough agreement' and then take it from there about less or more is practical or an aim
  • elika: I think there is a large amount of time spent reviewing proposals. I suspect the time spent in vetoing a large amount of proposals is not ideal. Agree with Brian’s comment earlier to pick a static number as the first step
  • jensimmons: suggested an alternative idea using points for ranking.
  • jgraham: the ranking is informative, not normative. Not changing the process on the fly is not ideal.
  • gsnedders: as a matter of process, our approach has been to have consensus for decision making. There should be some flexibility in decision making. Should we have an intermediate step to record objections?
  • jgraham: vetos can technically happen at any stage. Pushing that right to the end however makes sense based on the rankings and effort involved.
  • elika: people want a chance to see the rankings for vetos? Can we release the rankings earlier?
  • foolip: open to that, yes
  • jensimmons: do think that we should do rankings live at the Nov. 30th meeting. Still would like to have fewer proposals to evaluate internally. There are other amazing work we could be doing instead of discussing the ranking process
  • jgraham: should stick to the current timeline
  • elika: could we submit the rankings ahead of the Nov 30th meeting?
  • foolip: agree with that
  • jgraham: no objections to that
  • elika: in order to avoid ranking all proposals, we should also allow for a ranking of 0
  • bkardell: you can already do that by not scoring
  • jgraham: agree that we dont need to score everything. Lets discuss the static number (scope) in the next meeting since we dont have time to decide on a specific number now.
  • Consensus: use the Nov 30th meeting to input rankings.
• Interop charter #102
  • gsnedders: we should be explicit whether the decision making, for a new member joining the team, would be public or not
  • Next step - foolip will update the PR. Review and add any final comments for review in the next meeting