I keep hearing about ...(redacted)... and wonder if manufacturers are building drop-in devices for those technologies as available modules to solve some issues like this that z-wave is just not broad enough for yet.
So right, I'm not holding my breath anymore, either. Your approach with a gc-100 seems very
...(redacted)... but I can't help but think that if a device contains it's own logic to get a signal from a sensor/switch/meter/lightcondition/etc that the 'work' would then be in getting LMCE positioned to listen in respond to whatever these sub networks are sending around ethernet, wifi, rf, etc.
I use my MCV Vera in a similar way to the smart controller idea you're describing...
The Vera can trigger events that LinuxMCE can "see" happen (status of the lights), but doesn't "own." The Vera is easier to configure for "if A, then B," scenarios that are triggered by a Z-wave device (especially security protocol devices, like locks). LinuxMCE is better/required for control of events that are caused by anything else (media, communications, etc.). I haven't made any attempt to get the Vera and LinuxMCE to "talk" across the network itself, as I understand they would just conflict with each other (related to MCV "lifting" the Vera's DCE logic from LinuxMCE/Pluto, if I understand correctly). LinuxMCE can (with ?!limitations?!)
trigger a scenario based on an observable Z-wave event:http://wiki.linuxmce.org/index.php/Advanced_Lighting_Settings
- The z-wave window sensor wiring (this would be a good place for a device as the gate is between the house and garage where we have z-wave plans that will need devices mesh networking to reach)
There are two approaches in the links I posted (I think
both are mentioned, or can be found there) ,...
(1) Is to use the magnetic contact sensors to be triggered by the magnets in the doorbell ringers.
(2) Is to wire the sensors directly to the contact switch of the doorbell button, itself.
Both have issues, as discussed in those threads (voltage, in direct wiring or duration of sensor event when rigging the sensor with another magnetic source). The Z-wave radios in the sensors might not be strong enough to bridge long distances (So, other stuff needed). The radios in keychain remotes, compatible with a WDHA-12R (see below), might have better range. Plus they're relatively simple, and potentially easier to physically hack.
- Z-wave in-wall scene controllers. ...
I have a WDHA-12R, which is similar, in some ways,.. to the in-wall controllers, you are talking about, in how they are set up, if I'm not mistaken. I've never taken the time to get mine working the way I want.
Issues I've researched/experience with these gadgets;
(1) P.I.T.A. in terms of configuring it ("dumping" Z-wave net to them as secondary controllers)... EZ to screw that up.
(2) I don't even know if LMCE even responds directly to Z-wave events from these gadgets (don't think so, maybe custom?!)...
(3) Expensive (most of them)...
(4) To make it more complicated there are scene controllers and zone controllers,... and they work differently ... &
(5) There's a Z-wave "zone protocol" to consider, which I don't know if LinuxMCE supports (or whether that even matters to LinuxMCE, or is completely superfluous)...