Cart 0

Google Brillo and Weave: An Explainer and Analysis

Mike Wolf Android Brillo Google HomeKit Weave

The good news: The rumors about Google creating a new IoT/smart home-centric OS are true.

The better news: Project Brillo is more than just Brillo the OS, but also Weave, a 'communication layer' for IoT.

The bad news: details are still sparse on both.

But don't worry, we'll parse through what we have. First we'll try to explain each and then analyze the implications.

Brillo The OS

Brillo is an stripped down version of Android. It takes the kernel and hardware abstraction component, but reduces the total hardware requirements so that the OS can run on low-power, low-memory devices.

How low memory? At this point it's hard to tell as there's not a lot of specifics. In the keynote, Google talked about light bulbs and locks, but running anything resembling Android - or even a slimmed down version of Android - would seem potentially problematic. Typical on-board memory for a device like a smart lock is less than a one MB. This has meant operating systems for these types of devices have been extremely slim. Lockitron's original Wi-Fi smart lock, for example, ran eCos, an embedded RTOS that can run on a few hundred kilobytes of RAM.

But the bigger problem for Brillo may be the fact it would require developers of IoT and smart home devices to switch to an entirely new SDK and set of development tools. This might be welcome if you're an existing Android developer looking to get into IoT development, but it's a big deal if you're already have an existing family of devices or are accustomed to working with an existing software development environment for your products.

I still have some suspicions that Brillo may still have its most success in more resource-rich device categories like home routers.

Weave

So what is Weave? It appears to be more akin to HomeKit, a 'communication layer', or protocol that consists of a standard set of schema that will try to bring some order to what some consider chaos in the current IoT environment.

Other important things to know about Weave:

  • It's OS and PHY/MAC Layer agnostic. It will work with Brillo, but it will also work with other operating systems. It will also work on top of a variety of radio technologies such as Thread, Bluetooth LE and Wi-FI. 
  • Weave "exposes APIs in a cross-platform way" according to Google head of Android, Sundar Pichai. This is interesting and could create a standardized way for app makers and hardware makers to connect to each other.  The example given in the presentation was a recipe app connecting to a smart oven. It also will make adding features like voice recognition easier through exposure of voice APIs.
  • Weave will have an associated certification program, which tells me Google is going to run it as a de facto "standard" competitive with other protocols like Allseen, UPnP and other "standards" that are also trying to codify the IoT into a common set of descriptions and schema. What Weave isn't is a radio standard competitive to Zigbee, Z-Wave or Thread. It sits on top of existing radio standards 
  • Because Weave will be built into core Android, it will mean any Weave enabled device will be recognizable from your Android phone or tablet. This makes me wonder if Google will create a "Home" app, similar to what iOS is rumored to have for HomeKit device management.

Analysis

While all the rumors were initially about Brillo the OS, Weave is the blockbuster announcement. In Weave, we have what is Google's attempt to create a standard, create an answer to Apple HomeKit, and something that will work with any existing operating system. 

Brillo, on the other hand, appears to have more of an uphill battle. It's entering a market where there's already some significant entrenchment for small footprint OS's and their associated development environments. Further, it's counterpart in Weave almost seems to make Brillo somewhat unnecessary, since Weave is trying to solve the biggest problem today with smart home and IoT, which is fragmentation and a lack of interoperability across different platforms. 

In other words, I get what Google wants to do with Brillo in trying to Android-ize the world of IoT devices, but I don't think we really need that. What we need is less fragmentation, and THAT is what Weave is trying to solve.

Together Brillo/Weave show us a much more fuller and complete strategy for Google beyond its fairly limited efforts so far with Nest. Nest had always seemed a somewhat limited vision that depended on consumers buying Google's own hardware, but now with Brillo and Weave, the company appears to be getting back to its software platform roots.  

Lastly, if you're a smart home product manufacturer, now you have another choice, one that  likely will be an important one. The good news if you're choosing between HomeKit or Weave is you don't have to choose one or the other - conceivably you could create a product that works with both- the only cost is the additional investment in resources - human and otherwise - need to bring both to market. 

That's it for now - We'll have more thoughts soon as more information becomes available. 

Want to receive marketing intelligence like this in your inbox every week? Subscribe to the Smart Home Weekly. Click here



Older Post Newer Post


  • Hirsch on

    Google’s all over the map. Weave does make sense, Brillo not so much. And Thread – that’s kinda the forgotten man, isn’t it?

  • Finnsky on

    Great explanation. I think Google probably could have done a better job explaining what each Brillo and Weave does compared to alternatives in the marketplace. But since they didn’t, I’m glad you did!


Leave a comment

Please note, comments must be approved before they are published