The Secret Skill of Winning XR Companies: Data Integration
When those outside of the industry are asked “What are the foundational skills required to build and deploy extended reality experiences and applications?” you often get answers like:
3D modeling
Animation
Game Design
Programming
Companies building these experiences know (or experience a steep learning curve) that one of the most underrated foundational skills is data and sensor integration. Yes, you need a solid understanding of 3D modeling and programming to bring a virtual environment to life, but when it comes to B2G and B2B applications you must then populate that environment with useful data to deliver experiences that create value and aren’t just flashy demos that never get used.
Below we explore the top principles we employ to integrate data that populates our applications and deliver useful software that is more than a pretty demo.
#1. Use Open Source (when possible/appropriate). Use open source resources as much as possible to keep your solution scalable and future-proof. Open source resources primarily include the utilization of data types such as CSV, NetCDF, and geotiff to allow for integration with a wide variety of sensor, robotics, machine, and geospatial data. The XR industry is and will closely be coupled with the web development environment for the foreseeable future. Other open source data types to consider are Open Neural Network Exchange (ONNX) to ingest machine learning models for AI-enabled features and Universal Scene Description (USD) to depict three-dimensional XR data using a common format.
#2. Build for Cross Platform Compatibility. The web, mobile and XR markets are nascent and rapidly changing. To ensure that your work does not become deprecated before it comes to market (sorry if you were an early XR adopter and built for Magic Leap’s custom OS…), use architecture designed with an agnostic paradigm and industry standard frameworks. Even your third party integrations should target integration feasibility. Your software should be able to communicate to system-level functions or hardware without the need to build customized solutions tailor-made for each device. This means you can deliver software that is compatible with any hardware, even devices that do not exist yet that also adhere to these industry standards.
#3. Use API driven development whenever possible. When developed in a manner that is well documented, with error handling, APIs offer the most scalable and flexible way to query data from a wide variety of sources. Many data sources that we first approached do not have APIs. Rather than pursuing a custom integration to make the data available and XR ready we chose to take the time to deliver APIs whenever possible to take advantage of the longer term reduction in maintenance costs and improved future compatibility with various systems.
#4. Design modular components and/or with microservices architecture. Break the software into smaller, independent modules or services, ensuring that changes in one area don’t have a significant impact on others. This enhances maintainability and reduces risk of outages by allowing updates to be made in isolation. This is standard practice in web and most enterprise applications, but can be overlooked for XR applications.
A data-first approach to development will deliver seamless data exchange and functionality across different platforms to your customers. In a world where XR is a part of an ecosystem of software that includes mobile and web applications, data and sensor integration competency is a necessary skill to develop truly game-changing software.
