We are joining the Z-Wave Alliance Board

@balloob Do you only plan on making a USB-connected Z-Wave Controller? Wondering if Nabu Casa has also considered making an optional stand-alone network-attached Z-Wave Gateway appliance?

Asking since unlike Zigbee, the Z-Wave specification has an official Z-Wave over IP Gateway solution called “Z/IP Gateway” (ZIP Gateway), so do not have to hacky and unreliable serial-over-IP solution.

There is an old libzwaveip open-sources library (which looks to have PyZIP reference client), see:

At one point developers from Telldus even took inspiration from zigpy and wrote an initial open-source “pyzwave” implementation that supports connecting to Z/IP Gateway hardware with compatible firmware.

Z/IP gateway is a UDP/IP wrapper for Z-Wave command classes, which implements a set of Z/IP management command classes as defined in the network command class specification. Z/IP Gateway can communicate with the Z-Wave SerialAPI through USB to a UZB-7. Z/IP Gateway is delivered as source in Simplicity Studio.

A Z-Wave controller or gateway lets you add/remove/control all the devices running on the Z-Wave network via the Internet by bridging the Z-Wave PAN to UDP/IP. The components of the gateway supplied by Silicon Labs are shown in the diagram below.

The main benefit of offering a stand-alone network-attached Z-Wave Gateway appliance is that it could be placed anywhere in your home as long as it is connected to your local LAN network (via Wi-FI or Ethernet).

Using the Z-Wave over IP Gateway (Z/IP Gateway) as the Host bridges the Z-Wave PAN to UDP/IP and allows the communication between Internet Protocol (IP) and Z-Wave technology. Z/IP gateway client application (Z/IP Client) like Z-Ware and PyZip, can be used as a Web interface to the Z/IP Gateway and allow contact and control nodes in a Z-Wave network.

One word!

Nice!

It’s great to hear about this level of engagement in a key IOT hardware protocol. It shows that you believe in the future of Z-Wave and want to contribute to it’s evolution.
One tenet that I’d suggest is that you bring is the concept of “design for diagnosis” into the product design from the ground-up as many closed-source products often suffer from a lack of accessibility at that level. A loose example of this is an un-named TRV manufacturer who’s devices go offline but don’t report the state change, nor can it be detected without looking at the physical device. A reasonable attempt at a design for diagnosis would likely have caught that at the development stage and prevented significant user frustration.