Ambient OS and Natural Language as the Universal Home Device Protocol

A few days ago, during an interview at the Code Conference and a profile in Wired Magazine, Android creator Andy Rubin disclosed his plans to release a new operating system optimized for smart home devices. Code-named Ambient OS, the new platform provides abstractions and services that enable the implementation of applications that execute in smart home environments.

The conceptual similarities between Ambient OS and Android are impossible to ignore which makes Rubin the ideal candidate to tackle this problem. Based on the little information available about Ambient OS, it seems that natural language will play a foundational role in the new platform. Beyond the initial hype surrounding Ambient OS, the announcement is an important example to highlight how natural language is becoming a key element of smart home devices. In fact, I believe natural language could soon become the universal protocol for devices interacting in smart homes.

Natural Language as the Universal Home Device Protocol

Interoperability between devices is one of the biggest challenges of home automation solutions. While there are many standards like MQTT and others, they still require two devices to be painfully aware of each other’s programmable interfaces in order to interoperate. This level of knowledge-coupling is even more painful is we consider that the programmable interfaces for home devices are rather simple from the protocol standpoint. They are mostly constrained to the execution of commands and the on-demand query of data.

Now, let’s imagine a scenario in which gome devices could simply interact with each other or though s smart hub device using natural language sentences. The universality of natural language makes it an ideal vehicle to address many of the communication and interoperability challenges between smart home devices. Leveraging natural language protocols could bring some very unique dynamics for the communication between devices in a smart home. Here are a few ideas:

1 — Pervasive Introduction

Imagine a scenario in which, upon installing a new home device, this one will broadcast an introductory message to other devices in the network: “Hello (gotta be polite ;) ), I am device X and these are my skills: <list of skills>”. Natural language will allow other home devices to immediately interact with the new device without needing too much understanding of its skills. Devices will issue requests in natural language and process the responses.

2 — Richer Contextual Interactions

Let’s take a scenario in which we have a smart device that is able to open the garage’s doors based on a simple voice command. A few weeks later, we install a new sensor in the house that is able to track and report weather parameters such as rain, temperature, humidity, etc. Now, we could direct the game controller device to open the garage doors only if is not raining. Natural language should allow the gate controller device (or a smart hub such as Essential’s) to automatically interact with the new weather sensor to gather the rain parameters.

3 — New Skills

Natural language will allow home devices to incorporate new skills a-la Alexa without needing to modify the front-end interfaces. This model will allow smart home environments to evolve organically without major disruptions.

CEO of IntoTheBlock, Chief Scientist at Invector Labs, I write The Sequence Newsletter, Guest lecturer at Columbia University, Angel Investor, Author, Speaker.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store