The roads I take...

Here's a short summary of Mozilla-related work I've done in weeks 50/2011 (December 12 - 18, 2011):

• Mozilla work / crash-stats:
  Followed the Mozilla datacenter network problems that kept disturbing not just Bugzilla and other services but also our ADU numbers.
  Got my explosiveness reports to use ADU data for 10 and 11, now that it's available to the scripts.
  Triaged more topcrashes and regression crashes.
  Discussed with Sheila, the CrashKill team and with Scoobidiver in bug comments on where we set topcrash keywords or not.
  Participated in discussions on last-minute Firefox 9.0 crash concerns.
  Followed the Socorro team working on Android Fennec support as a separate product.
  Wrote up a proposal of Q1 priorities the CrashKill team has for Socorro.
  More watching of new/rising crashes, caring that bugs are filed where needed.
• Web Apps:
  Following the web apps preview announcement, I decided to put some free time work into some web app experiments myself. I created Mandelbrot and Lantea Maps app prototypes, based on work I had previously and an idea I had been contemplating/investigating for a while. Both still have a number of rough edges and work to be done, but they show promise. I hope to get more out of them when I have time.
• Various Discussions/Topics:
  Repository access levels, Gecko linking RAM size, my tablet problems, etc.

A dozen years of my contributing to Mozilla just became full this Saturday, and that looked like a good point in time to try another new avenue of contribution.
I have been thinking about a web-based OpenStreetMap-using mapping and tracking application for some time, and even created a concept page on my wiki some time ago when a volunteer came up who wanted to contribute something to B2G - and a maps app is still missing there. This weekend, with the web app developer preview being announced, I finally took the chance to dig into it myself. That said, Lantea Maps had its first field test yesterday and I recorded GPX tracks successfully on my N9 and they look reasonable in an application that can show them against a map. I'm looking forward to improving it - and hopefully even someone making it into an app to run and looks really fitting on B2G/Gaia at some point.

Here's a short summary of Mozilla-related work I've done in weeks 49/2011 (December 5 - 11, 2011):

• Mozilla work / crash-stats:
  Communicated with Daniel from metrics because of missing ADU data for my experimental reports.
  Together with other people, tried to keep crash analysis together while Mozilla servers had problems.
  Followed up with the Socorro team on the plan for Android Fennec as a separate product.
  Triaged topcrash bugs with Sheila.
  More watching of new/rising crashes, filing bugs where needed.
• SeaMonkey:
  Created linux64 update snippets for SeaMonkey 2.5.
• Firefox code:
  I did split up my search UI cleanup patches into logical parts, and put the engine manager cleanup in its own bug, also split apart into smaller patches.
• Various Discussions/Topics:
  Tested OpenGL layers and reported its problems, data stores for "native" Fennec, B2G and crashes, Windows update service, my tablet problems, etc.

This week, I was pretty happy not to be in Mozilla IT - they said what they've seen was more or less a "perfect storm" of different problems that made multiple services unavailable for some time and caused a lot of us quite some trouble in work. But still, I don't want to complain, this can happen as we have quite complex systems, hardware problems played into that, and AFAIK we are scaling up some capacities which can lead to bottlenecks and strange problems. The Mozilla IT team did manage to get things under control in a very organized and good manner, and I'm sure some of them didn't sleep well in those days. They put up a monster job and are probably working on making sure such things will not happen again. Their job is somewhat unthankful usually, as you only see when things go wrong, but usually not when things work perfectly, which is how they do most of the time.
We should thank those people at chance we get, as without the infrastructure and services they are managing, our project could not exist and work as well as it does. They are the spine that keeps Mozilla upright and stable on our way into the future of the web. We should be grateful for that.

Here's a short summary of Mozilla-related work I've done in weeks 48/2011 (November 28 - December 4, 2011):

• Mozilla work / crash-stats:
  Contributed to the discussion of where we were with chromehangs and filed the bug to disable the hang monitor for now.
  Got the ball rolling on and stayed in the middle of getting a long-term solution to having different crash-stats reports between Fennec XUL and "native" Android UIs. We have a solution now and are working on implementing it. Thanks to Laura for her game-changing idea.
  Filed a bug to make the Socorro graphs make more sense with zero values.
  Investigated some followup work needed to do for signature summary.
  Did some crash regression bug triage with Sheila and Marcia.
  Continued to watch new/rising crashes closely, filing bugs where needed.
• Themes:
  Did the rest of the porting of Firefox 9 changes to LCARStrek, so both EarlyBlue and LCARStrek 2.6 are basically done and only need some testing before I can send them to AMO.
• Various Discussions/Topics:
  SeaMonkey machines, data stores for "native" Fennec, MXR future, ESR plans, Windows update service, L10n and string typo fixes, Mac OS X 10.5 support, Mozillians "phonebook", tablet hardware problems again, enjoying my Nokia N9, etc.

Just as the last month of 2011 is starting, some discussion and press seems to go out there how this year has been a bad one for Mozilla, but while the media tries to paint the sky black, it looks quite different here inside the project.
It certainly has been and continues to be a year where we face great challenges, but because of that, we have been seeding a few quite great things - delivering new features, performance and memory saving to Firefox users a lot faster than before, a web app framework getting created (at Mozilla Labs) that is open and able to grow to a cross-vendor standard, a web identity mechanism (currently dubbed "BrowserID") coming together that puts users more in control of their logins than before and makes life easier for web developers at the same time, growing a new version of Firefox for Android that is really fast and works smoothly, working on solutions for large deployments (or "Enterprises") again in an own working group, starting work on an open, flexible and fully web-based mobile stack (currently dubbed "B2G") and new standards needed there in the form of the WebAPI project, and much more.
There's surely fallout and a year where a lot of really new stuff is being created usually produces some kind of bottleneck in finished products to show off, so it's no wonder that this one isn't too different. Also, large products in open projects of course have some fallout in terms of discussions, concerns, unclear communications, and slight course corrections along the way. Still, the seeds are sown and great results will grow from them - I'm sure that our awesome community, including paid staff, can take care of those young sprouts in a way that makes them grow into great products.
And then there is the simple fact that the web is a competitive market again, which to a large part is our achievement, and we should be proud of that, but at the same time work hard to stay relevant in that market so we can influence it enough to keep it open for everyone. If the web stays one of openness, opportunity and innovation, we win. If we can help to make the user be more in control of his/her online life, we win. If everyone can put up his/her contents in the way (s)he likes, we win.
Mozilla is the only non-profit in this market and we have an awesome community, so I know we can win. And with your help, we will.