Compatibility/Meetings/2021-12-07

From MozillaWiki
Jump to: navigation, search

Minutes

Scribed by James

QA Labels (Paul)

We were thinking of ways to improve the QA processes and we were looking at how much time we invest in creating the reports we send. In order to reduce the time spent, we were thinking of adding some QA-specific labels in GitHub that would help us identify faster the data for the reports. Of course, if this wouldn't disturb the team in any way. As an example, for triaged issues, we could add something like "QA:triaged", so we can identify the number of issues triaged per week/month/year using queries.

  • Karl: Labels never bother us. We already have 200! What do you want to track; maybe there's a way to track it with the event data we already have? That would avoid needing to assign a label.
  • Paul: Can't filter by > 1 milestone in the GH UI.
  • Karl: GH WebHooks send all the events and we are already listening for those. So we aren't limited by what's possible in the UI.
  • Dennis: In theory we can answer the questions based on the data in the ES instance. We should chat about the exact requirements, but I think this is doable without any labelling.
  • Paul: We will discuss more about what we want and then get in touch.

Firefox Focus labeling (Raul)

With the new version of Focus being out, and with more reports coming from this side, we have difficulties in separating which reports are for Focus without hints or clues from reports or users, and which are for the Release/Beta version, due to the fact that the ML Bot does not label issues specific for Firefox Focus (it does for versions below 94)

  • Raul: After version 94 ML bot stopped labelling issues. Issue for imgur was thought to be a unreproducable Fx issue and closed, but were actually a focus issue. This could be happening on other cases.
  • Karl: Did we change the UA for focus? How do we identify focus issues compared to normal Fx issues.
  • Dennis: Reports via the in-product reporter ought to be labelled correctly. There was an issue that no focus labels were applied, but I fixed that 20 days ago. I see issues with the label added. If there are still some issues that don't have the label, it could be a bug in the reporter. If you see issues that aren't labelled, let Dennis or Ksenia know.
  • Raul: We have about 20 issues in a document.
  • Tom: Smartblock and WebCompat interventions should now work on Focus and should reduce the number of issues.
  • Raul: ML bot not working for 94+, it just puts a label "fenix", "firefox-mobile" or "fixme". Working fine for !focus.
  • Dennis: I will need to check the situation outside the meeting. Recent issues with `browser-focus-geckoview` labels: https://github.com/webcompat/web-bugs/issues?q=is%3Aissue+label%3Abrowser-focus-geckoview

Firefox VR (Raul)

Just to confirm, from now reports regarding FxR received on our webcompat repo, will be moved to the FxR repo on Github, by creating there a new report with the details received on our webcompat repo, and the original report closed with the MOVED milestone and a specific comment? e.g. (https://github.com/webcompat/web-bugs/issues/93510)

  • Dennis: OK to just ask the reporter to create the issue in the FxR repo.

Webcompat work - What can we do better? (karl)

We didn't have any All Hands or meetings helping us to figure out a bit what we would like. It would be cool if we could take a bit of time to think about 2021 Webcompat work and make three lists: * What was bad? * What was amazing? * What would be your dream?

  • Karl: All Hands turns out to be useful. I'm interested to know what's going well and where we need improvement. Where are the oppertunities to do new or better things.
  • Dennis: Hard to talk about this in a live meeting. I've been trying to close webcompat issues more proactively. Sometimes this causes more work for Oana/Raul, but sometimes it's not a good use of time to spend time on low priority things. But it's hard to make the call on priority. This helped me reduce my personal backlog.
  • Karl: Need to avoid creating more triage work.
  • Oana: All issues get a severity label, we should use that.
  • Karl: New auto assignment system. Was that a good idea? It works well for me, but does it work well for others?
  • Dennis: I'm biased, but I also like it. Helpful to keep up within a day or two of the issues being filed.
  • Tom: Can be hard to always do the work on an ongoing basis.
  • Karl: Everyone has become more autonomous, and more proactive at proposing and solving issues. e.g. "slack thing", site interventions tests, ML bot work. Really good to see.
  • Karl: We still find it hard to be visible in the organisation. Still trying to address this e.g. by talking about things in the platform meeting. I also wish we were better at working together
  • James: One of the problems is to make it possible to be less reactive and more proactive for webcompat. More cooperation with interop effort and more efforts with web platform.
  • Tom: More general exposure for standards so that we could be more proactive about identifying issues and knowing what's going on.
  • James: Interop 2022 is an opportunity to work with people at Mozilla and other browsers and define priorities. Having a way to see the compat issue severity/frequency would be useful.
  • Tom: Being part of the standards discussions seems like it would be helpful to us and others.
  • Karl: There is a standards meeting once/month that you're welcome to attend. Two meetings per timezone. More people are welcome to participate. Good to understand the dynamics of things, even if you're just listening. Good to get context on things that are happening, and the story of discussions.
  • Karl: Raul joined the team and has got up to speed really quickly. Really really nice.
  • Karl: Feel free to talk individually if there are sensitive things.

Two Minutes ( 👹 )

The useful, the silly, the things you did since the last meeting. Pro and Perso accepted.

  • Dennis: Working through some of the old needssitepatch issues; landing the new inteventions; the usual fun.
  • James: Interop 2022 discussions. Updating old patches that got review.
  • Karl: Diagnosis. Starting to organize the review/feedback dance. Meeting about UA string. Went to Shichi-go-san for my son. Prepared the Christmas tree.
  • Ksenia: PTO
  • Oana: Triage and moderation and a bit of Enterprise testing. PTO with bad weather, but managed to do 2 bike rides and a bit of hiking with the dog. Team log report: https://docs.google.com/spreadsheets/d/15ONn602MHXzuhCl5c2ODo5xmdDNnYUzJ9ptGsJTd-8E/edit?skip_itp2_check=true&pli=1#gid=876973640
  • Raul: The usual triage and moderation. Looking forward on hitting the slopes this weeked. First snowboard ride of the season.
  • Thomas: Automated test suite! Also tons of goal and other planning, and finishing up dFPI-related tasks. Also snow.. The doggo is happy to see it, then miserable to realize it's actually real. And then it goes away, and comes back, restarting the vicious cycle.
  • Joe:
  • Paul: Getting a hold of all the teams I took over, trying my magic in solving everyone’s issues and working on metrics. Besides these, while kido was asleep, Netflix and chill due to rainy weather.