Throughout my time in the misfit capstone group, my group and I were faced with a variety of challenges. For example, while the standard capstone class is given either a specific technology, problem or client, our group was left to define our own challenge and a technology to solve said challenge. Right off the bat, hours were spent research for new technology and their applications. After some time, we narrowed down said technology to a projector that transformed any surface into a touch screen, a mixed reality driving accessory, and the Amazon Echo Spot. Based on our defined devices, we then began to address the openness of each devices developer platform. Within minutes, we immediately realized that Amazon's open sourced technology actively invited developers to create unique products. After settling on the Amazon Echo Spot, we then began to research the target audiences the Spot was able to reach. While older demographics do heavily use the Spot, we found that children are also active users. That said, the number of child related skills on the Spot was small in comparison to other categories. Noting this lack of children skills, we wanted to design a skill that was different, one that hadn't yet been attempted. Keeping that in mind, we wanted to also target parents, the ones who would be downloading the skill. Although we teased the idea of a choose your own adventure bedtime storytelling skill, we found these type of skills to be oversaturated in the amazon marketplace. After some deliberation, we decided to address the number one problem facing parents, motivating kids to get ready quickly. To solve this problem, we wanted to create skill that doubled as a game, coupled with a reward system. It's from this idea that Ready Freddy was born.
Once the idea was solidified, we split into teams. While my background is in public relations, the recent skills I've gained through the New Media Institute having made me adept at coding. As a result, I became the lead coder and skill designer. That said, Amazon uses a proprietary SDK, with node.JS at the primary coding language. Seeing as the NMI certificate program does not teach this language to students, it was up to me to learn it myself. Luckily, Amazon's partnership with Codcademy gave me foundational tutorial to build upon. After about two weeks, I felt confident enough to begin laying the framework for Ready Freddy. Although my team came up with elaborate goals for our skill, I eventually began to realize the limitations of, not myself, but the Alexa software. In particular, I ran into quite a few problems because of Amazon's stringent privacy restrictions. For example, skills cannot be open for more than eight seconds without interaction, due to consumer concerns about Alexa devices spying. To solve this problem, I found that the skill could stay open as long as audio was playing. Nevertheless, Amazon restricts audio files to 90 seconds. Additionally, while we originally planned for users to make their own routines, this would limit our own ability to design the skill. Instead, I decided to make three different routine sets, with tasks ordered in different variations.
In these routines, the child can navigate through the skill by just the click of a button. For reference, when the skill launches, I set it up to ask the user which routine they would like to load. If the user doesn't know the routines, it reprompts the user to ask what the routines are. Therefore, the user can either immediately say a numerical value to load a routine, or the can load a tactile list with descriptions of each routine. Once the routine is loaded, the child is left to click a button on each task slide, with each clicking moving them closer to the end result of a gold star and animated video.
After a month, a rough draft of the skill was complete and presented to Megan, the advisor of our capstone project. In this update, I went in depth, explaining the ins and outs of the skill, our limitations, and the future direction of the project. Based on her input, changes were made. Once these changes were finalized, I helped conduct a usability test of the skill with one of our target audiences, children. From this experience, we learned that the custom voice we gave the skill was "creepy", the buttons were too small, and some popular routine tasks were missing. Luckily, Amazon is actively updating the Spot, and while developing, Amazon gave developers the ability to make images into buttons, instead of relying on hyperlinked text. Once the voice, buttons and routines were updated, I began to add some additional features. For example, I found research that showed that children respond better to personalized experiences. Therefore, I created a database structure in the skill, able to collect the user's name, or names, and use said names in the dialogue of the skill. Additionally, in order to create a more fun experience, I designed said dialogue to rhyme, keeping things light and friendly. While I continued to push my coding skillset, I ran into problems when attempting to include weather API data into the skill. Unfortunately, not only did the break the code, it bricked the entire device. After tirelessly working to resolve the issue, I decided to simply let the user determine what they though the weather was, and based on the button they click, the user will be prompted with different instructions.
Although I was responsible for coding the skill, I also had to make sure it stayed true to brand. Therefore, throughout the coding process, I communicated closely with the design team. Additionally, I assisted the website design team, providing support for any jQuery, HTML, or CSS coding issues. Finally, my last role involved editing our writing and presentations deliverables throughout the length of the course.