VerticalSeperator

Homepage

SliderTest

SliderItem

Button

MadeInPlasmic

Navigation

MoreInformationBtn

Screen

Useless.media v2

plasmic-embed-css

[PlasmicKit] Made in Plasmic Badge

plasmic-basic-components

plasmic-tabs

react-slick

We started the project from the lowest possible point attempting to compile the existing Firebase c++ SDK into a custom Objective C wrapper, as our bases for this project, while this only solved a handful of the requirements we were able to reverse engineer the rest of the features from the starting point, Ultimately allowing the customer to deploy a uniform single model analytics solution on all platforms Including tvOS.


tvOS Firebase library 

Another big aim of this project was to raise the QoS for the end users by minimising the effect of playback related errors. Our approach to solving this was to utilise our internal player-framework Wacka’s extensive configuration model to create tailored playback configurations for known devices, as well as, differentiating configurations based on active technologies, and even specialised configurations for individual streams.


We also performed extensive analysis on the errors we were unable to eliminate completely through configuration to allow us to build bespoke recovery flow specifically targeting the remaining known issues resulting in a model with 65 to 70% recovery rate with an average playback interruption of 0.5 seconds.

All of this put together has significantly increased the viewing experience of the customers’ Chromecast offering which is also noticeable from the increased  user base with more than 120 million stream starts in the first 12 months of operation this project has been extremely well received by Telia's users and must be regarded as one of the biggest success stories for the DeCaf framework.

Custom DeCaf Client

Playback & Backend Integration SDK

The project was built as a chromecast framework with a receiver component handling all interactions with the Cast Application Framework, overriding Shaka with hls.js and dealing with all playback, lifecycle and error related operations, while a service layer component was created which could be extended to allow individual services to override default configurations, UI, analytics, as well as, alter basic functionality like “end of stream”, “next content” behaviour and playback restrictions.

One of the most challenging problems we solved in the project was allowing the use of hls.js in conjunction with the Cast Application Framework in such a way that the receiver could still pass Google’s certification process which requires 100% compatibility with Google’s standard sender libraries, as well as, other features like voice and cloud commands and connectivity with Google Home Devices.


Chromecast receiver framework