On March 29th, Plank had its first Hack Day. A hack day for an agency? Normally this idea is reserved for developers and is often the proving ground for the startup process. We liked the idea and decided to try it out with a full production team at our disposal.
Make something cool. Keep it simple. Document it. Learn something. Have fun.
Why would we do this?
We are a curious bunch of folks. We all acknowledge that running and working at a client service focused agency sometimes leaves little time for experimentation and doing our own work.
We have a strong production model that can help get an idea from single thought to working prototype fairly quickly. Pausing and using it for ourselves is a tantalizing idea.
How did we do it?
We created three teams; we wanted each to be multidisciplinary and to play to the strengths that are part of our normal workflow. We often run projects with a core team, including organisational/management, design, html/css, and backend developers - something we're very good at. This seemed like a natural way to approach the challenge.
We chose a single work day. We hedged our bets against an 8-hour block when we knew there was the least chance of anyone or anything pulling us away. (In all honestly, we all accepted that our clients do come first and that any emergency would put a pause on the day, but thankfully our clients were able to fend for themselves for the whole day.) We tried for no email, no phone calls - as few distractions as possible.
We moved our workspaces to get closer to one another and got to it. There was no shortage of tools at our disposal: Basecamp, email, Bug tracking, Skype, Git, Dev servers (local and remote), sketchbooks, super fast internet, and an awesome work space!
We had all contributed to a master list of ideas; some were fairly extensive, others small and precise. Each team had to pick one idea to use as a base to try and create something great around it. The work didn't need to be completed but the goal was to have something figured out and presentable by end of day.
Here's what we told everyone:
"Make something cool. Keep it simple. The ideas might seem bigger than a day - bite off a chunk that represents something possible. Plank is good at making objects. If the whole thing can't be real - fake it to get the idea across. Document it. Learn something."
What was the day like?
We gave everyone a rough schedule:
- 9:00am - 10:00am Setup and day prep
- 10:00am - 11:00am: Group kickoff, teams and ideas (meeting in kitchen)
- 11:00am - 12:00pm Phase 1 work (in groups)
- 12:00pm - 1:00pm Lunch (provided), discussion and previews/feedback (lunch in kitchen)
- 1:00pm - 4:00pm Phase 2 work (in groups)
- 4:00pm - 4:15pm Project status update (quick 'stand up')
- 4:15pm - 5:00pm Final home stretch (in groups)
- 5:00pm - 6:00pm Presentation and wrap-up
We made small commemorative notebooks for everyone to sketch out ideas and jot down notes - we expected the day's activities would likely generate more inspired ideas.
Geoff Skyped in from Toronto. Murphy's Law dictated that Skype would be acting up on that day of all days, so one other thing that came out of this experience was setting up an excellent permanent Skype connection for involving him in meetings - we fixed this a couple of weeks ago.
Amelio's Pizza for lunch, lots of coffee, snacks, and a real battle against the clock - everyone felt that the time flew by way too quickly.
How did we do?
Team 1 - aka "Happy Face"
The Problem: Plank uses many web services to aid in its production process. Most Plankers have their own system for how often they check certain services, and by which manner they do so. Some visit the service's website directly to get their updates, some use various RSS feeds, and so on.
We identified two key issues with this approach. The first is that a wide range of "information consumption" behaviors lead to cases where not everyone is up to date on all matters. The second issue we found is that given the largely decentralized nature of our web-accessible tools, it is difficult to get an easy-to-digest overview of all our systems.
Our team set out to solve this issue.
The Solution: We came up with the not-so-new idea of a dashboard. This dashboard would highlight all the key information across all of our systems in an easy-to-digest manner. Each service could have a corresponding module, plus some for-fun modules to keep things interesting (example: bus schedules, new music releases). Such an interface has an endless number of possibilities, so our Hack Day focus was to make a proof-of-concept to convey the usefulness of such a tool to the whole team.
Our team's primary goal for the dashboard is best explained by the following:
"The idea is not to replace any existing system, but to draw attention to all activity across all systems in one central location for the Plank team."
Team 2 - aka "Crazy Eyes"
The Problem: A visual way to communicate quick stats and easy-to-miss facts around a current, passionate local debate - the Quebec Student Strike.
The Solution: With so much information in the media around the tuition hike, our team wanted to showcase the key points and themes that are currently being discussed in Quebec. We wanted to find a way to communicate these topics at a quick glance and with a bold, eye-catching interface.
Where did we start?
- Big Picture: Interactive infographic? Static infographic? Quebec tuition vs. the World? All of Canada? Our initial ideas were big and, as with any project, you need to be realistic. We couldn't cover it all, so we quickly zoned in on facts that dealt with Quebec-related themes only.
- Content: Research, Research, Research, Writing, Editing, more editing. Once our intern, Samantha Graham - who is fairly active in the movement herself - communicated all of her amazing knowledge, Sarah Bagnall worked with her on editing and framing the facts for our layout.
- Design: As always, content guided design. Once we knew what we were tackling, Gary was able to put a look to the themes and the style of content we had chosen.
- Integration: Geoffrey Weeks, participating via video chat, took the final content and design at the 11th hour and started the build of the "FEE FREE FREEZE" mini-site.
While we didn't have enough time to complete the mini-site in its entirety, the day proved to be a good lesson in evaluating time-restrictions, scaling-down content for design and web, and working as a team in a research- and content-driven project.
Team 3 - aka "Big Mouth"
The Problem: With turntable.fm now limited to US IP holders, we are not able to use their service to develop a Plank-curated streaming station.
The Solution: We would begin the process of building a local version of a streaming player to which anyone could upload music into a queued playlist.
- The specifics of what we attempted:
- Set up a user login system
- Use http://www.uploadify.com/ to manage audio file uploads
- Decide on a user interface and emulate a diner jukebox
- HTML5 interface
- Simplified user playback interface
While much of our time was spent figuring out uploads and queuing, we discussed how we would like it to work given more time to develop it. Would users be able to skip or interrupt songs? How many songs in the queue would be displayed? Would we provide a history? Would users take turns uploading or would it be a free-for-all? It was a good exercise in UI/UX planning as much as in working with open source technology.
We're planning on making this a seasonal thing. Our next Hack Day will be the Summer Edition, to be scheduled for August. The ideas are already brewing - or that might be the coffee. Either way, it's great stuff.