Please update your status at: http://benjamin.smedbergs.us/weekly-updates.fcgi/

• Are there any new Rust blockers? https://github.com/mozilla/servo/issues/2853 (larsberg)
• Github just improved comments on rebases - switch back from Critic? (zwarich)
• Linux ref tests travis? (gw)

• larsberg: Please add anything that's blocking you to the issue above so I can bring it up in the weekly meeting.

• zwarich: I'm going to suggest a move back to github for code reviews. While there have been 3 people working on layers, nearly every patch requires a rebase. This causes huge annoyances with Critic. Using the rebase feature doesn't seem to work with the comments unless you close/reopen. However, I just discovered that github now tracks parts that were not changed by a rebase and carry over the comments. It's not clear to me how this is happening. It also doesn't check that you've addressed the comments (similar to critic), but is much more pleasant. It's also a much easier workflow. People have also missed a Critic review / note and merged a PR. The fact that people sometimes don't check Critic is an indicator that it's sort of heavyweight. The advantage of Critic, though, is that it does ensure that people address issues. It also lets you split up a review across multiple reviewers, though I don't know how useful that is (you have to communicate with them out of band to coordinate). So, I'd be eager to move to Github, since it's also likely that github will improve.
• pcwalton: I slightly prefer github's workflow over critic. I find Critic heavyweight and it never logs me in right, causing me to lose stuff.
• brson: I don't have a dog in this fight, but I do thing it would lower the barrier to entry if there isn't a second tool there.
• larsberg: We probably need some more feedback from jack, jdm, ms2ger, etc. before we make any decisions here. I know there are people who like it.

• gw: Where are we at with that? There are a few things slipping through at the moment. What's the status? And anything we can do?
• martin: I was looking at it (Reftest failures) and tracked down one of them. If you're willing to try my branch, we can get some more info on if they're related. I think right now I'm still seeing OSX failures, but it's unclear if that's related to this or just general flakiness that existed before.
• gw: I'm more interested in what's likelihood of doing linux tests on Travis? Is it just time resources
• larsberg: cmr mentioned http://xpra.org/trac/wiki/Xdummy to support our current glfw work. We are working on a port to efl, though, which will support linux ref testing natively. Should be able to create a PR that uses Xdummy directly. [ ed. cmr also mentioned - http://mriksman.blogspot.com/2013/07/our-friendship-is-over-xdummy-replaces.html ]

• larsberg: the echo!