Scribed by Thomas
Incoming issues numbers are rising (Oana / Sergiu)
We had more than 120 incoming issues over the weekend. On Monday morning, there were 200+ issues that needed triage. The numbers are increasing, according to http://adamstevenson.ca/mozilla/reports/. As we cannot triage this ammount of issues on a daily basis and also test the Enterprise sites, we propose working 4 hours on triage, and 4 hours on Enterprise. We risk creating a large backlog of untriaged issues, but we'll be able to finish the Enterprise testing by the end of the quarter.
- Sergiu: we've come up with a plan to do both triage and enterprise testing, but I'm not sure how that will work on a 4-hours-a-day for each.
- Oana: it won't be enough to burn down the triage queue.
- Mike: how many issues can you typically triage in that time?
- Sergiu: I did about 30 issues.
- Mike: it sounds like we'll have to be "ruthless", and close lower-quality issues.
- Sergiu: yes, I was doing that today, I'd normally do about 30 in 8 hours.
- Oana: the time involved does vary from issue to issue, of course.
- Mike: agreed that it's time-consuming and advanced work especially at that level.
Enterprise testing (Oana / Sergiu)
We're (almost) done with the Enterprise domains we had access to. We'd like to get credentials/access for the remaining ones.
- Mike: I've reached out to the manager for Softvision contracts. My manager Panos is also planning to talk to him. We're hoping to get 1 or 2 more people, one for triage and one for diagnosis if possible.
- Mike: in general, I think splitting "50% triage" like this should probably end up okay. Adam, what do you think?
- Adam: tough question. I don't know; it's difficult to find the right balance. You can't go too fast (and lose things), for instance. But to me, it makes sense in the short term to at least make sure we put in half an hour a day to keep things under control. The quality of bugs *does* seem lower than in the past.
- Mike: I can probably do about 5 hours of triage a week, and I think Adam can pitch in a little time each week as well. If the two of us can handle 8 hours a week, I think it will be fine. We have 100 issues right now, after all. So let's see how this setup works for the next couple of weeks, and see if we need to adjust.
- Mike: a related question on the agenda: do you have enough enterprise sites to test?
- Sergiu: at the moment, we have enough on Sharepoint and Slack. We were going to talk with Adam after that to get accounts for more sites.
- Adam: I spoke recently with someone about getting a Workday test-environment, and I'll let you know how that goes.
- Sergiu: We need a ServiceNow account as well.
- Mike: I know Adam's been doing a lot to get accounts. I'll circle back with Romain to see if he can help us in that regard. Will that help, Adam?
- Adam: I have a feeling they'll just circle back to me, so let's not bother. I'll try to find the time this week to figure out the rest of the accounts.
- Mike: You might want to reach out about Webex soon, since I think they're meeting with Cisco.
- Sergiu: I think this week the workload may be lower for enterprise; if so we'll shift over more time to triage.
- Mike: Sure, and we can always file Reddit and Twitter stuff under enterprise, I suppose :)
- Adam: Sergiu and Oana, did we end up getting a Salesforce test-account? I know they jumped onto a bug somewhere... did they give us access?
- Sergiu: the last note we have here is one about asking (?) for access...
- Adam: right, but I recall there being a bug where they asked us about access.
- Sergiu: ah, yes, I remember, but I ended up being blocked by a security question. (gives bug #)
- Adam: ok, I can follow up on that, thanks.
OKR check-in (miketaylr)
Adding some bots to web-bugs (miketaylr)
- Mike: Comment on this issue if you have opinions about adding bots to the web-bugs repo, to help is with some management tasks.
- Mike: also, once we set a needsinfo status on a triage issue, we can have a bot come back and close it a week later if there's no response, to help with the workload.
Issue with priority-labels (Sergiu)
- Sergiu: I logged an issue against webcompat.com where some site issues weren't being labeled with priorities, like for GMail.
- Mike: I'll ask Karl about that.
- Sergiu: what about domains not in the top-1000? Are they not assigned a priority label at all?
- Mike: they should; the default should be the lowest priority.
- Sergiu: we've been assigned a LOT of them manually.
- Mike: then I may be wrong, but that's what I thought was supposed to be happening.
- Sergiu: it would be nice if it was happening automatically that way, even outside of the top-1000 sites.
- Mike: but GMail seems like a top-1000 site, right? So I'd guess there's some sort of bug.
- Oana: for GMail is has normally always assigned a critical priority, so a bug seems likely.
- Mike: alright, I'll ask Karl to take a look.
Two Minutes ( 🐝 )
This is the summary of what you have done during the week. * Feel free to add your own 2 minutes. * Keep it short. * Feel free to add your name if you think you need to share something.
- Adam: Outreach, fun meetings, Tracking Protection triage process and scripts
- Dennis: More work to get GoFaster landed... :/
- Karl: needsdiagnosis. Worked also on re-triaging the remaining viewport issues Bug 1123938 so they are in their appropriate bugs. A bit of outreachy stuff. Some "O Canada" administration kerfuffles for traveling to Orlando (at risk).
- Mike: Outreachy. PTO. Planning stuff.
- Oana: Triage and Investigate issues and PTO.
- Sergiu: Triage and issues investigation, and some Enterprise
- Thomas: Landed the webcompat reporter webextension, and have been working through the remaining issues with it (also some other XHR/Fetch bugs and stuff)