Wednesday, January 19, 2011

Various home automation

I have been interested in the MiCasaVerde (MCV) Vera unit for a while so when a chance to pick one up cheap rolled around I took advantage of that chance. So a couple of weeks ago I picked up a MiCasaVerde Vera version 1. The Vera unit (both version 1 and 2) can directly support Z-Wave, Insteon and X10 (?, maybe). I'm currently using it with just the Z-Wave devices. It has a web interface and it uses Lua as the user language. Pretty much this is the model for today's HA system. Small, no fan, runs Linux supports Insteon, X10 and Z-Wave. easy to install (wireless, much less fuss), and a user language. Sounds near perfect ...

Now please understand that I run a few open source HA projects of my own and work on MisterHouse too. Also understand that my opinions are biased and I'm almost the equivalent of a competitor to MCV. Hey I'm a programmer and engineer, I like things my way. :-).

Now having said that, the Vera has a proprietary library/extension for UPNP. MCV calls this version of Lua: Luup. Other than the user code and the OpenWRT distribution MCV is not Open Source. MCVs forums are pretty active and the folks there are willing to help you answer questions and solve problems. Some of MCVs support folks roam the forums and will also chime in. One thing I've noticed in the forums is that there are a large number of folks who are very unhappy with MCVs support. Seems that MCV went silent as they prepared for CES in January and the users didn't hear much from them, actually still haven't (though support still handled support questions and issues). The CES announcement was a big one. Seems they had Samsung as a customer. I wonder what this will do to support for the little guys?

I had a bit of a problem when I tried to add a Z-Wave lamp module (that had been matched up with another controller). I contacted the support email and got a prompt response. The information was accurate and my problem was solved in few steps. One of the posters in the forums noted the support seems to only pay attention to half the message. I did note this as I had clearly stated that the unit I had was upgraded to User Interface 4 (UI4) and they said I had UI2 but no harm, no foul. Still I did get the correct answer. The GUI interface is okay (it is pretty) though I don't like the results as it isn't in a useful format and finding things is not straight forward. The Lua language itself is very much like Pascal. So far I haven't attempted any scenes (where you control a group of devices all at once) and I haven't attempted to create any user code. I'm currently waiting to pick up a few more Z-Wave devices that I ordered from Radio Shack (they had huge a sale). So I'll get a chance to do all of that in the near future. In the mean time here's what I think about the Vera:


  • Router/Firewall size device
  • Wireless/Wired device (router, firewall or bridge)
  • Low powered
  • Quiet
  • Easy to install
  • Wireless (Z-Wave and WiFi)
  • User language (Lua)
  • Forums
  • Decent array of Z-Wave products support
  • It's based on OpenWRT/Linux, so it's portable
  • Not too expensive
  • Easily expandable
  • Medium to Advanced DIY friendly


  • Lack of support
  • User Language (Lua - too Pascal like)
  • Administration is ... odd
  • Lack of current documentation
  • Lack of useful, basic examples
  • Lack of up-to-date bug fixes (major problem)
  • Lack of information sharing from the company
  • Lack of Vera reliability/stability
  • Taking on a huge customer (Samsung) while ignoring their current user base.
  • More and more looking like they're moving towards a subscription based services (part of the CES show).
  • 240 MHz processor/32M RAM/16M Flash - too small for more complex HA

My opinion is that MCV has around 6 months before their competitors (HomeSeer and several others) have their products in consumers hands. In that time MCV really needs to step up on the bug fixes (partial X10 and Insteon? come on!), fix the stability, make nice with the current user base and prove they're a better company. The product has promise. I must also say that the Lua language is not one of my favorites. I'm going to play with the OO version of Lua and see if I like that any better (but that doesn't mean I can put it on the Vera). I must say that I like MisterHouse (MH) better. It has the same kind of language problem (Perl vs. Lua though) but the MH Wiki has a lot more examples and I can edit the wiki to fix things. Also MH is infinitely more flexible and supports more protocols than the Vera. And for the amount I need to know to really make use of the Vera I might as well work with MH.

So there you have it, while I'm biased, I think the product has potential but the company has handled a lot the wrong way. They need to do some work and regain the trust of their user base. If they work hard they have a future. If they continue on this path they have none and will be eaten by their competitors. Still in all this the rest of us can sit back and hopefully learn from MCVs mistakes. I'm hoping I will.

Update: I've edited this a bit to be more fair to MiCasaVerde. I am a bit biased and it's better that I state that than leave it off. I have no love of PASCAL, I like my open source projects and Z-Wave is not one of my favorite companies. Also if you search for "micasaverde open source" you'll find that MiCasaVerde has always stated that they're built on the open source Linux kernel/OpenWRT and not stated they are open source, though much of the community seems to think this (myself included).