Once the prototype was complete I transitioned onto the main Disney+ Xbox client team. The motion designs and animation specs we defined during the prototype were now the guide for each client team, and I worked with our design team to publish generalized specifications for all animations in the app.

Between April and November 2019 we pushed hard to get the client apps ready for prime time. I produced pages and controls and provided finishing touches on pages and controles produced by others. I did final optimization passes to ensure controls were loading properly, that data operations were not blocking the UI, and that animations were running smoothly.

One major project I completed before launch was building a sprite sheet player to power the animated brand tiles (the Disney/Marvel/Star Wars tiles that transform into short looping movies on focus). The UWP video player control had a bug that caused it to pause very briefly before restarting looped video.

Other platforms with lower performance capabilities implemented the brand tile animations using sprite sheets (image files containing every frame of a short video in a single large file). UWP did not have a control capable of playing sprite sheets, so I undertook the task of building one. Ultimately my player provided high-performance looping animations with low memory usage which powered the brand tiles as well as the animated splash logo on startup.

Another major project I undertook was refactoring all design artifacts into a standardized module. Working with a cross-functional team of engineers and designers we produced a workflow that translated design artifacts such as colors and fonts into code repositories published in every language used by Disney+ client apps. I managed the integration of the UWP repository into our codebase and produced a reference page in our in-app debug panel showing all color references used in the app and comparing them to the Figma color references to ensure alignment.

My manager pulled me into the Disney+ prototype early on, before it was called Disney+. In those days it was called Project Dominguez, named after the Dominguez Tree, the oldest living thing in Disneyland. We were soon joined by a motion designer and future creative director. Our mission was to built out the Disney+ UI in an interactive form, so the design team could experience their work "in motion", so executives had something to play with and show off, and eventually so that user testing could be accomplished.

We had static page designs to start from, so we quickly built out the basic structure of the UI: Horizontally scrolling rows of tiles, five static brand tiles, and a looping auto-scrolling carousel at the top. Brian, our motion designer, and I worked together to develop the transitions and animations that would bring the interface to life. The prototype became a blueprint for all client teams to follow when implementing animation and transitions.

Two interesting engineering challenges stand out to me from this time. First is the infinitely scrolling carousel at the top of the screen. Building a control that scrolls left and right is easy, however looping it proved very challenging. Many different methods existed for manipulating the visual tree or the underlying data structures to move elements around within the list, but each of them caused different and unpredictable glitches, especially when scrolling quickly. This prototype was being used by executives to demonstrate to investors, which meant any build we sent out had to be bullet-proof. Eventually I figured out a model that combined under the hood data structure manipulation with advanced XAML visual tree manipulation which resulted in a near-seamless experience.

The other challenge involved the behavior of content tiles at the edge of the screen. Disney+ content rows show five tiles at any one time, with the sixth tile half-visible at the edge of the screen. Once scrolled, a half-tile is visible at both ends of the screen. The designs called for these edge tiles to be semi-faded. We tried various overlays but there was always a visible line where the tile could be seen going under the overlay.

I developed a solution using a decoupled event/receiver system that could be implemented by the list view and by the tile controls. This system allowed each tile control to know it's relative position within the list and when it's position was about to change. The tile could then manage its own visual state and run animations locally when appropriate. Because the event/receiver system did not depend on tight coupling of lists and their element views the solution scaled to any list view or tile set anywhere in the app.

The culmination of the prototype project was the demonstration given by Disney CEO Bog Iger announcing Disney+ to the world at Disney Investor Day, April 11, 2019. Everything shown that day ran on our prototype.

NHL.TV was my first real assignment when I arrived at MLB Advanced Media, what would eventually become Disney Streaming. This was on Xbox One which used the UWP framework. I was already familiar with UWP, XAML, and C# from my time at Microsoft so I was feeling confident.

My first task on NHL.TV was updating their scoreboard UI, which included a significant animation component. I had never done XAML animation before. By the time I finished rebuilding the scoreboard I had learned a little about hockey and a lot about XAML animation.

My next task for NHL.TV was a home page redesign and rebuild. This was the most advanced design I had ever implemented. I learned about page loading, XAML performance analisys, lazy loading, and XAML optimization.