Engineering

Overall, the engineering team worked quite well in 2008, and was much more coordinated in its efforts than in previous years. Weekly development meetings allowed the team members to cross-communicate and collaborate more effectively on projects, and for the project managers to track all the projects across all the developers over time on a cross-functional GANTT chart. Think of it as air traffic control, with geeks.

We expanded the role of one of our team members to focus on system requirements gathering, user interface design, and quality assurance. Leveraging that, the team instituted more traditional development methodologies that allowed for more reliable product development and quality assurance. The success of the project registration forms development effort (and their satisfied stakeholders) are testament to the efficacy of these changes. The team had the smoothest year ever of supporting the off- and on-playa database for the DMV, and provided significant technical support to the Rangers. Suffice to say we’re keeping with these methodologies going forward. Specifically, we:

  • implemented more stringent quality assurance (QA) procedures for testing content and functionality updates, including testing support from members of the stakeholder teams;
  • divided changes into front-end and back-end, and assigned them to appropriate resources, allowing for faster implementation and turn-around on content, especially in the week prior to launch;
  • created flowcharts that documented all related and dependent questions, in both the main and sub-questionnaires;
  • created the PDFs for each questionnaire for participants to download, with added page reminding them to fill it out online;
  • created a project mailing list;
  • wrote test plans for forms QA efforts;
  • kept stakeholders abreast of collector tickets, especially closed and deferred bugs; and
  • supported end users using generic response emails and custom-written responses, sent within 24 hours of receipt.

The creation of the Burning Blog, was a large project, and was well worth the effort. The blog looks beautiful, and is turning out to be an excellent communication tool for the Burning Man Project. All the various extant blogs were consolidated into this blog, including Regionals, the Enviroblog, and Building Black Rock City (which was migrated from its original location in Plone). Allowing moderated comments has been a big win as well, as it allows participants to have their voice on our site, while maintaining civility at the same time. Now that the team is over the development hump with the blog, it’s all cruising from here, except for the occasional maintenance upgrade, adding category icons and finalizing the list of tags.

In 2008, the team began working on migrating everything off “Blaze”, one of our older servers hosting both production and development environments. So far the team has migrated Forms (DMV, Art, Placement, Media, Scholarship, etc.), BRAF and BWB websites. We still need to move the Extranet to this server, and hope to get that moved in the first half of 2009. This Extranet move will be coupled with an upgrade to bring it to a more recent version of Plone.

The team has been working on an implementation of LDAP (Lightweight Directory Access Protocol), which will allow for one central repository of user authentication data, so internal and external users of Burning Man’s applications will have just one login and password that gets them into all Burning Man user-facing systems that they are authorized to access (note: this is not single sign-in, which is a different animal altogether). This is a huge lynch pin to our future development efforts, and one that is coming together over time.

The team’s main limitation is the number of (human) resources available to do all the work we want to accomplish. There are a number of important and interesting projects that suffer due to a lack of resources. To counteract this, the team is reaching out more and more to volunteer developers to help with projects, setting up a development/sandbox environment and exploring new technologies that allow for the expansion of our resource pool.

For instance, a small team (of mostly volunteers) is currently prototyping, analyzing and recommending a new development framework. One goal of this evaluation is to select a technology that will be more accessible to a larger number of engineers (e.g. based on more commonly used technology). Plone, one of the key technologies the team has been using for years, is particularly restrictive in that regard. The team is applying this new framework to other technologies, including the What Where When calendar, the People’s Database, project registration questionnaires, and more.

Readying for this, the team implemented and configured a development server to act as a sandbox for our internal and external engineers to work on and test new projects without risking the production servers. Note that our main website has always used a staging server for this purpose, and the team extended this to more substantial engineering efforts.

Burning Man Earth has been cranking along well in 2008, due to the heroic efforts of a solid team of talented volunteers who have been working on user-facing interfaces that will permit a more distributed process of participant-driven content in the system. Work will continue on this in 2009 and beyond. The team worked with the various data stakeholders (placement, art installations, What Where When) to establish processes for getting the appropriate data to the team and published on the live website. Everybody was happy … and we see a lot of potential with this project.

Submitted by,
Will Chase,
Paul Showalter,
Calliope Gazetas,
and Ian Starr