Here's a short summary of SeaMonkey/Mozilla-related work I've done in week 35/2011 (August 28 - September 4, 2011):
- Mozilla work / crash-stats:
Some more wrestling with "featured" versions on the Socorro front pages and along with that, verifying some fixes that landed in a Socorro update.
Ran my custom reports directly on the Mozilla server - at the end of the week even in a completely automated fashion.
Created a first version of a per-component report but this will take some more work until it's really useful.
Took part in some planning of crash-related sessions for the upcoming Mozilla all-hands meeting.
Of course, continued to investigate "explosive"/rising crashes. - Themes:
I worked on some first steps of updating EarlyBlue and LCARStrek for 2.4 versions, but there's more to be done there. - Various Discussions/Topics:
DigiNotar incident, upcoming all-hands, German community meetup, mobile performance, "Highlighter" tool name, Web APIs, more Mozilla website merge fallout, maemo and Nightly, etc.
The security discussions around the DigiNotar stuff took a lot of time to read this week, but they taught two lessons, I think: 1) The CA model has serious problems in its design as it needs us to trust that CAs do the right thing and are not broken into, and 2) as long as we need that CA model, any CA that wants to keep their business running needs to immediately and fully disclose to the major browser vendors (i.e. the keepers of the CA root stores) when something goes wrong - if they fail to do that, consequences quickly need to run toward their complete exclusion, like it was in this case.
