Weekly Status Report, W34/2010

Here's a summary of SeaMonkey/Mozilla-related work I've done in week 34/2010 (August 23 - 29, 2010):

• Releases:
  Finally released SeaMonkey 2.1 Alpha 3 this week, after spending a few hours on readying relnotes and the announcement.
  Shortly after that, I started off the release process for the 2.0.7 security update, which is now available in candidate builds and on the "beta" update channel.
• Build Machines:
  An issue with Windows L10n repacks turned out to be a conflict with pre-existing trees on the machines, I clobbered all boxes to fix it.
  I also investigated some missing L10n nightlies and should have a fix very soon now.
• Build System:
  The small packaging fixes reported last week could land, just like those for jsperf (covered by a test).
  Later in the week, I worked on and fixed the reporter removal and the new Addon Repository prefs.
  I also did some more investigation of building with libxul and filed bugs on S/MIME bustage and import linking failure with this enabled.
• Automated tests:
  The test I created based on some WIP code from Frank for the FAYT / Findbar focus issue was reviewed and I could land it so we should catch possible regressions there.
  I couldn't take much of a look into out test oranges, but some other people thankfully did and some are being fixed - but it looks like others are being added as well.
• Places:
  The rest of the first batches of Firefox backports of Neil's review comments have been reviewed, approved and landed, so we're one bug down on those. The bug for cleaning up the places library is probably up for a longer story, as the scope bar part might not live up to theming expectations yet.
  The fast bookmarking button went through some review and also is up for some backporting, I updated the SeaMonkey patch and hope to get this integrated soon. Building on that, I'd like to convert the editing panel in the browser to an arrow panel later on to make it feel more logical and fluid in usage.
  And on Mac, the places bookmarks stuff doesn't work when no windows open, which needed some fixes on the Mac-specific "hidden windows" that serves their window-independent menu in that case.
• Smaller Work:
  I don't feel for doing an own bullet point for each of them, so here's a few smaller items:
  I updated the doorhanger notifications patch once again for some review comments, I hope it finally gets somewhere.
  On the ongoing user agent discussions, I finally abandoned the utopia of getting clean as-they-should-be UA strings when Minefield decided to switch to a "Firefox" UA, taking away the majority of technically potent people testing the non-Firefox-branded Gecko web. Still, as I'm for doing things fully or not at all, I concluded no Gecko should be delivered without "Firefox" in the UA and filed a bug for an automation of that. This was WONTFIXED and I guess we'll stay where we are in the end, so we'll probably solve this on SeaMonkey with the UI pref and a default to "on" for now.
• Data Manager:
  Once again, I updated the Data Manager patch for recent review comments, of course only to trigger more of those never-ending review comments and a negative super-review based on a crash I can't investigate, reproduce or debug and which very probably isn't my fault or in my code. In the end, that might just be what will stall this feature indefinitely. I still filed a bug on correctly listing IDN.
• Themes:
  I updated my LCARStrek and EarlyBlue themes for SeaMonkey 2.1 Alpha 3 on Sunday to not leave users of this milestone empty-handed when it comes to theming - though they now have personas as well, of course.
• German L10n:
  I created and updated a beta channel version of the German language pack for SeaMonkey 2.1 Alpha 3, so that people can test this release in German language as well if they like or need to.
• Various Discussions:
  SeaMonkey Development Meeting, SeaMonkey build machines, work on final look for new addons manager, visit to Bay Area and Mozilla HQ, probable PPC end-of-line, status bar, etc.

Our statistics on active daily SeaMonkey 2.x installations have been stagnating since the start of the holiday period in July, but this last week have been returned to increasing significantly, and on Saturday, we had our over-a-week average going over 90,000 active daily installations, just to have the single-day value break the 100k barrier the first time ever two days later, i.e. yesterday.
This is an important milestone to hit and tells us we're going strong, but what I'd really like is to have ten times that number, i.e. a million people starting SeaMonkey 2.x every day. Can we hit that? Can you talk 9 more people into using SeaMonkey 2.x? If each of our users can, we should be able to make it - in theory. Will you take the challenge?