Skip to content

bmatsuo/rex

 
 

Repository files navigation

#Room Experiences (REx) Build Status GoDoc

Demo screenshot

REx is an experiment attempting to advance the design space of applications and games running on TV set-top boxes (Android TV in this case). REx takes advantage of an underutilized application architecture to provide rich experience shared by the entire room.

REx pairs an Android TV application with mobile client applications running on consumer phones (Android and iOS) or custom hardware. The desired effect is similar to Nintendo's goals with the Wii U platform. But the goal for REx is to provide a low barrier to entry by utilizing existing commodity phones and set-top boxes.

This project contains working example applications to demonstrate the core technologies. The core deliverable is the open source messaging framework that can be used to build more advanced games when more resources (i.e. time and money) are available.

For more information about how a REx application works see the design document.

##Features

  • Event/Message passing framework for state synchronization. The framework is pure Go so any desktop or mobile app should be able to make use of it for low-throughput synchronization and message passing over a local network.

  • The framework intended to work across all mobile and TV platforms. Android TV is just the best existing option. With Apple TV now supporting custom apps and as Go support on that device matures look forward to working demos built for that platform as well.

  • Game servers automatically bind the message bus to a port and make themselves discoverable to clients on the local network. The servers specifically target TV set-top boxes (Google Nexus Player, Nvidia Shield, etc). But should work on any Android devices (tested on Nexus Player and Galaxy S6). Untested for iOS.

  • Game clients automatically discover the server and begin receiving their update events. Works on Android devices (tested on Galaxy S6). Untested for iOS.

  • Technical demo (with .apk files) showing discovery and state synchronization.

##Documentation

The docs/ directory contains various documents that describe the components of REx and how they are used. Important sections include the following:

####Examples

Information about the examples provided with REx.

####Development

Useful for people trying to hack on REx and compile binaries and application archives themselves.

##Motivations

###Why write REx?

Honestly, I want to get together and play board games with my friends without having to spend all that money on physical pieces (and the effort/space to store and take care of them). While many board games exist on mobile devices few offer a low friction multiplayer experience and often they lose an essence they had when played physically with friends, sitting at a table. It is exactly this feeling I am trying to replicate on a digital platform. See Prior Art for more background.

###Why write an application for a TV?

See the previous answer. I believe that a large shared place is key to replicating the board game experience in a fully digital application. In this regard a TV functions almost identically to a table. Furthermore the common shared space allows almost all shared information to be removed from the mobile interface. This can greatly help implementing games involving more complex mechanics and player decision points, especially those revolving around secrets and private information (e.g. Poker, Magic: The Gathering, etc).

###Why use mobile "companion" applications?

See the previous answer. Games and applications relying on private information and complex mechanics including asymmetric player objectives/behaviors benefit greatly from a "controller" that has a (i) a display and (ii) programmable logic. The Prior Art can provide more background on this topic.

###Why write REx in Go?

See the previous answer. The REx architecture typically involves two applications being developed and sharing libraries. What's more, there are multiple mobile platforms to target (and potentially multiple TV platforms). Any sane developer putting themselves in these circumstances would want cross-platform development libraries and tools. The truly sane would go on to ask not to use C(++) for all their development. Luckily these objectives are inline with those of the "gomobile" project (blessed by Google). Go's cross-platform toolchain unifies iOS (tvOS?) and Android application development and allows a single team to maintain the entire game platform.

What's more, when an application grows to also require dynamic cloud-based applications Go easily extends it reach there and allows you develop along the virtuous path of Go servers connected with Go glue to mobile frontend apps written in pure Go. Bare metal feels so good.

Hana Kim's talk from GopherCon 2015 provides an excellent background of gomobile, its motivations, and its benefits. The video is highly recommended as reference and background on the history of Go and mobile platforms.

###Why write REx for Gopher Gala 2016?

See the previous answer. I wanted to create a novel and compelling experience. And I wanted to build my first mobile application (and really my first game using GL graphics). After looking over the gomobile project and how the development cycle goes I thought it would be possible to create a cool, shared interactive experience in a short amount of time, even with limited previous experience with mobile/game development.

##Prior Art

###Games

As alluded to in the introduction, the space being developed is not entirely unexplored. Nintendo's Wii U is the most notable example of a game platform expanding into a complex "controller" architecture. The Wii U has an intriguing story for the market. But the economic barrier to entry for the platform is too high for many people. And you can't write a Wii U game in pure Go!

The concept of local discovery is not new in gaming. Games have been doing this going back 20 years (hyperbolic-ish). More recently in the mobile space games like Spaceteam have delivered very fun and compelling experiences using only local network discovery. These games unquestionably demonstrate the ease of use and low barrier to entry this approach provides over other global discovery engines. Users do not need to create an account; they launch right into the fun. And it is a special kind of fun.

The resurgence of local co-op in video games like Spelunky and the Board Game Renaissance in recent years is evidence that local, in-person gaming utilizing a shared space can provide some experiences and feelings that are (as of yet) unattainable through internet based multiplayer experiences which are starkly impersonal (as starkly impersonal as the rest of the internet).

These are long trends of entertainment and the REx project intends on helping provide an evolution in the game space with rich in-person multiplayer experiences built on flexible, cheap, and abundant tools. REx will take games like Spaceteam to a new level.

###Collaboration and Productivity

The most widely known collaboration product to make use of local-only discovery is probably Google's Chromecast (Google Cast technology in general). Using local network discovery devices can send streaming requests to Chromecast devices on the same network. REx is heavily inspired by this architecture and could be used to emulate the Chromecast's functionality.

Apple is another notable user of local discovery, authoring their own protocol called Bonjour. Bonjour is the discovery protocol powering several Apple services including Air Drop (a file transfer service). The REx authors take inspiration from Apple's work developing a clean user experience and workflow for collaborating with colocated peers.

Both examples here are centralized platforms and have limitations either in their application or the client operating system. The REx author believes that even greater productivity and collaboration tools will appear when application developers are given an open, cross-platform framework and the tools for harnessing local discovery-based mechanics in their own systems.

##Authors

Bryan Matsuo bryan.matsuo@gmail.com

##Copyright and License

Copyright 2016 Bryan Matsuo

This project available under the MIT open source license. See the LICENSE file for more information.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Go 99.0%
  • Lua 1.0%