For what it is worth, here are my two cents…
There are 3 fundamental aspects to home automation - 1) protocol 2) devices and 3) controllers.
1) Protocol (network fabric) - The front runners are Z-Wave and Zigbee with BLE 5.1 (Mesh) holding a lot of promise but lacks broad adoption because it is early in it’s lifecycle. WiFi is not an HA protocol although some would like to make it that; WiFi just isn’t designed for efficient low-power device operation. I personally think it is a huge mistake to populate your WiFi network with WiFi/IP devices from China because they have access to your main network and run behind your firewall (the ideal trojan horse scenario). For example, there have been countless security exploits with WiFi cameras. Said another way, using a completely separate purpose-built and secure HA network fabric like Z-Wave greatly reduces the surface area for an exploit of your home WiFi system that hosts your most sensitive information and infrastructure.
2) Devices (sensors and actuators) - Depending upon your needs/use cases, you may find devices for Z-Wave, Zigbee, BLE and WiFi that suit your purpose. From my own personal experience, I have yet to find a device for an alternate protocol that wasn’t available on Z-Wave. With Goggle’s pushing Thread and Samsung’s SmartThings, the Zigbee gap has closed but there are still more interoperability issues on Zigbee than the longer-standing Z-wave ecosystem because the Zigbee standards are looser.
3) Controllers (orchestration/automation) - Most HA protocols/standards include simple automation through mechanisms like Z-Wave’s direct device-associations so you can do things like turn a light on when a sensor is tripped, etc. However, a robust and capable HA system requires broader and more sophisticated orchestration that is not possible without a controller that serves as the brain for you HA. The ideal controller will be extensible through plugins to encourage developers and enable users to solve their specific problems easily. Furthermore, the controller should be able to integrate with multiple protocols and devices to address the broadest set of end user needs and maintain cadence with current trends such as voice-control (Alexa, Siri, Google Assistant, Cortana). And of course, ideally the HA controller platform is well-supported and has a large developer community.
Perspective (the grass isn’t always greener) - Right now, Z-Wave is broadly adopted, supported, secure and more standardized than Zigbee with a myriad of devices being available to suit most HA use cases. I personally standardized on Z-Wave for my network fabric years ago (circa Vera 3) but devices that “speak other protocols” are components of my total solution (I am one of the original MySensors developers) however, I choose to use a single HA controller for automation orchestration. It is important to not confuse an HA Controller (i.e. Vera, Habitat, SmartThings, OpenHab, etc.) with the HA Protocol/network fabric (Z-Wave, Zigbee, etc).
I’ve been a long time Vera user and I’ve taken a cursory look at every new thing that comes along but I still run my HA systems (my primary home, a vacation home and my 85-year old mother’s home) on Vera Plus’ right now. The Vera is still a front-runner with it’s Lua/luup extensibility and broad developer/community support that allows end users to bridge the Vera to other modalities such as Homekit for Siri support (which I do).
As my HA demands have increased, I have never found a use case that I couldn’t solve with the Vera and community extensions. Yes, I have certainly experienced Vera’s mysterious and frustrating failing under load but that is a solvable problem and often can be mitigated by examining the logs, identifying the root cause/pattern and adjusting the configuration to avoid the failure scenario (I’m in the midst of trying to figure one out). The Vera Customer Support Team is very quick to help users when they take the time to ask and have spent countless hours with me to fix a number of issues with my systems.
Furthermore, if you visit the forums for other HA controllers, you will find that it isn’t a bed of roses either and that they suffer from their own set of problems - SmartThings cloud-responsiveness, Hubitat’s maturity (e.g. lack of udp support), Homeseer’s pricing model, OpenHab’s complexity, etc. At this point there still isn’t a perfect controller but remarkably, even though the Vera development cadence has waned due to lack of resources, the fundamental architectural/design choices have withstood the test of time and for now, Vera remains my controller of choice until some TBD player really upsets the HA market by doing something uniquely different and better than what currently exists - that may never happen.