Compatibility/Meetings/2016-06-work-week: Difference between revisions

From MozillaWiki
Jump to navigation Jump to search
(→‎Topics Bank: adding topics marker, moving one section)
(→‎Classes of WebCompat Issues: adding agenda for community)
Line 72: Line 72:
* A brainstorming in our team would be to define what we consider a Web Compat issue and that should help rationalize our answers to the rest of Mozilla and Public.  
* A brainstorming in our team would be to define what we consider a Web Compat issue and that should help rationalize our answers to the rest of Mozilla and Public.  
* We should also probably survey people at Mozilla.  
* We should also probably survey people at Mozilla.  
* Also wondering if a usability study could be done in cooperation with Mozilla UX team  
* Also wondering if a usability study could be done in cooperation with Mozilla UX team
 
==== Measuring compatibility (and progress) ====
@@mike@@ to explain
==== Structured Triage (bugzilla and webcompat.com) ====
@@mike@@ to explain
==== * Training contributors to triage and do outreach ====
@@mike@@ to explain


=== 💡 Automation ===
=== 💡 Automation ===

Revision as of 05:45, 12 April 2016

This is the agenda and logistics page for Web Compatibility Team Work Week meeting in London (UK) from June 13 to June 17, 2015.

Logistics

  • Location: London, UK (Tea and Biscuits)

Attendees:

  • Karl Dubost
  Arrival: Monday, June 13 (probably Boat from France + Train to London)
  Departure: Saturday, June 18 (Train + Boat)
  • Mike Taylor
  Arrival: 
  Departure: 
  • Hallvord Steen
  Arrival: Monday, June 13, 10:20 (LHR)
  Departure: Thursday, June 16th, 11:20 (LHR)
  • Adam Stevenson
  Arrival: 
  Departure: 
  • First_Name Last_Name
  Arrival: 
  Departure: 

Regrets

Agenda

Mon, June 13, 2016

Welcome everyone.

Tue, June 14, 2016

Wed, June 15, 2016

Thu, June 16, 2016

Fri, June 17, 2016

Sat, June 18, 2016

Topics Bank

Use the following template in one of the section below:

 ==== Topic ====
 Description with [http://example.com links] when necesssary

💡 WebCompat.com Development

Usability testing for webcompat.com

Usability testing of our various "report a site problem" features (aiming to streamline as much as possible). I suggest we list or set up couple of pages that have known bugs/brokenness, and challenge random all-hands participants to load URLs and report the problems, then observe them using either extensions, the firefox:input feature or the webcompat.com form.

Better filtering UI for webcompat.com

Better filtering UI on webcompat.com is very important. I'll keep nagging about https://github.com/webcompat/webcompat.com/issues/795 until we get it right.

Deployment scripts for webcompat.com

(when miketaylr is on holidays or sick) See Also

Merging PR for webcompat.com

(when miketaylr is on holidays or sick) How do we manage the PR merging and its conflicts for deploying.

💡 WebCompat.com Community

Expectations for NeedsDiagnosis

As a group we do not share the same expectations for what should be in needsdiagnosis state, and what it should not. It would be good to discuss it so we have a kind of agreememnt about it. For example, this was switched to needscontact without where in the scripts it is bogus. It's just a general description of the issue.

  • What are the entrance criteria for needsdiagnosis flag?
  • What are the entrance criteria for needscontact flag?

Classes of WebCompat Issues

A discussion has been started on the list about classifying the issues for WebCompat. There are probably multiple answers.

  • A brainstorming in our team would be to define what we consider a Web Compat issue and that should help rationalize our answers to the rest of Mozilla and Public.
  • We should also probably survey people at Mozilla.
  • Also wondering if a usability study could be done in cooperation with Mozilla UX team

Measuring compatibility (and progress)

@@mike@@ to explain

Structured Triage (bugzilla and webcompat.com)

@@mike@@ to explain

* Training contributors to triage and do outreach

@@mike@@ to explain

💡 Automation

  • Compatipede-2 plug-ins: the ones we have (demo), how to write them (workshop)
  • Looking at the DOM, HTTP, CSS and JS: what are interesting data points to record for comparison and tracking over time? (Discussion - good to do before that workshop - we'd write plug-ins to track new data points).
  • Matching/diffing error messages from Gecko and WebKit - can we write some code that can give us a meaningful diff and tell us what errors happened in engine A but not in B? (new Compatipede analysis module?)
  • What should our approach be to automating *interactivity* for automated tests, such as automated log-in? (discussion, hack session)
  • We can create some spin-off statistics for purposes that are not strictly compat but "nearby" like memory usage. How, and what's important? Memory usage? Can we measure jank? Scroll performance?
  • Explore working on the proposed WebExtensions-based Firefox/Chrome add-on to control the browser on behalf of Compatipede, to move off SlimerJS and Phantom (there's some initial code somewhere on GitHub - a Boar branch I think)
  • Should we aim for testing on real devices? If yes, how do we bridge the gap between Compatipede and the Firefox instance on the device? Marionette (whenever it comes to Fennec..)? WebExtensions (whenever it comes to Fennec..)? Nodejs-Marionette (testing on FxOS devices)? A custom-written devtools protocol client? Something else?

💡 Tech Evangelism Bugzilla

💡 Platform

💡 DevTools

  • Can we contribute more to DevTools QA?

💡 MDN

💡 Misc

WebCompat work sharing what we do

A distributed team is working well when we are all sharing what we learn about the project and our actions. Hallvord and Karl are sharing through minutes and blogs otsukare whatcouldbewrong. How do we discover about the work of Adam (meetings with desktop, access to physicial meeting in Toronto) and Mike (commits on Gecko/Platform). Some of these are sometimes popping up through discussions but were not necessary shared in advance. So basically how do we better work together?

IRC bot and IRC Web Archive

Some contributor expressed the desire to be able to read the logs of our IRC channel.

  • Archives of logs == Good 1. history of our discussions we can point to with proper URI. 2. Could minute the webcompat meeting directly on IRC (W3C style). Bad 1. soft opacity has a benefit for certain jokes and discussions to not be broadcast and shared with only the people in the room. 2. Might create expectations of work stable information on the channel instead of bug trackers, mails, etc. so create another source where to look for information.
  • Bot == 1. tools giving information for leaving messages to people in the future (answering machine). 2. Giving information about bugs when typing number (Github and Bugzilla) 3. Meeting logs for minutes.