Reflections on Project A3.

Prototyping Learning Environments


1—What did you learn?

This project was instructive in a variety of ways, below, I will talk about a few of the the lessons that stood out.
• We prototyped our 'mat-based' interactions with projected light, this meant we lost a great deal of resolution over what we could see on our Retina Display MacBook displays. This meant we constantly needed to preview our designs via the projector to make sure we were designing 'viewable' interactions and flows. Next time, I'm going to look into some 'down-resing' applications for my computer display to speed this up somewhat.
• Again due to prototyping with projected light, we didn't account well for the fact that the light would fall on top of people's hands while they're interacting with the system. This meant the interfaces available on the mat would sometimes become distracting when a raised hand would intercept the light before it could hit the mat. In the future, I would like to explore ways of projecting light up onto the mat from below—i.e. some kind of projection table.
• While prototyping our Digital ToolBox concept to allow for continuous interaction between digital and physical components, we didn't consider how this would change the types of interaction patterns that would be possible. For example, it's difficult to pinch-to-zoom on a physical object. This rethinking of interaction validity was a huge learning area, I would honestly like to explore it a lot more going forward.
• We used Keynote and a projector to prototype our system, this meant while one person was roleplaying through the design using the physical components, another other was triggering certain interactions by advancing the slides manually. Though this was the quickest way to prototype our idea, it did present a number of problems regarding choreography. In future, prototyping a system with a simple advance trigger built into the mat may be able to help with this issue of multi-human coordination.
• Because we were trying to prototype for a future 'imagined' material, we did not have the luxury of actually being able to have multiple prototyping surfaces available. In essence, this means we end up designing with a piece of 'installation projection art' and it makes it very hard for everyone in the group to learn the material to the same level when only one person can play with the imagined material at any given time.

2—What were you challenged by?

I was fundamentally challenged by two main things during project three: (1) scooping into the right part of our idea, and, (2) getting caught up in setting up a system for working *better* instead of actually working.

With regard to our scope, our conceptual foundation was clearly derived from Bret Victor's Seeing Spaces and also generally inspired by his theories of representation and communication. Victor's ideas underly one (that is SeekingKit) of the three main pillars of our system, the other two are Digital ToolBox and Guided Learning. Digital ToolBox is largely based on thinking about systems as hybrid digital/physical environments, and is in a way owed to Peter and Austin's teaching and examples in this mini. The final pillar, Guided Learning, I believe is the least well foundation-ed and lacks the conceptual strength of the other two, despite this, it utilized a large amount of our time during the design phase. This meandering around with all of the possible directions our 'design could go' was a distinct mistake that we did not notice until very late in the design process.

From the start, I was adamant that we design within the constraints of the material wherein we were proposing a solution. What I mean by this is simple, we were not going to design a hybrid environment on a screen or in physical form either. We were going to prototype and design directly on our physical mat, this worked fine for early prototypes, but as we advanced into the territory of more nuanced interaction design, we struggled excessively with trying to engineer a better system for working with projected light. To do this, we spent almost two full days trying to setup a system for Keynote to have continuity while being edited or 'played' back to enable a more seamless working system. We messed around with projection software, applescripts for full-screen state-switching, and more, but ultimately couldn't get our improved system to work. The kicker was that without fixing some of the small issues, we were still able to design fine and would have been able to achieve a lot more had we not gotten distracted on engineering the system.

3—What did you do really well?

To start, our idea was heavily inspired by Bret Victor's prior art (as mentioned above). I believe we effectively did two things really well: (1) moved towards his theory of seeing spaces by prototyping a possible 'look & feel' of such an environment, and, (2), prototyped a version of real-time continuity between digital and physical realities with our Digital ToolBox concept.

4—What would you do differently?

Were we to do this again, I would try to not get caught on either of the two primary errors discussed in the 'challenges' section and I would have completed a more thorough literature review before beginning the interaction design phase.