As simply an end user and not a developer I have been pondering some of the “get s@*t done” obstacles of Vera and the current crop of HA solutions in general. While Vera does allow a good deal of control over a limited set of domestic devices such as lights, thermostats and alarm systems there is a large gap in its ability to automate and fully integrate with domestic life. While having the ability to turn lights on and off via a timer, or even more advanced logic thanks in no small part to the community, the buck seems to stop there. I am often given hope for what the future may hold by reading the blog of Ian Mercer(1). He seems to have hit the nail on the head with regards to this topic. Vera generally gives you what he would refer to as a “dumb home”, which simply gives one computer control, possibly with an attractive UI, over things which can easily be done manually. True home automation on the other hand can make predictions, is highly aware of what the occupants of the home are doing, why they are doing it and can explain the decisions it made.
We are steadily moving forward with Vera’s integration with our domestic products thanks to the hard work of our community developers. It seems to me that the missing components that would bridge the gap at this point are a data base and a natural language interface. I imagine that for a developer, creating a DB of Vera’s actions might not be overly difficult. Even if an end user were to try to use Vera’s logs for such a purpose it would be fruitless as it’s not “human readable”. In my opinion one of the most exciting developments recently in the Vera world excluding the Program Logic tools provided by the Venerable RTS is the “Event logging database” by forum member Quinten (2). Unfortunately I don’t believe this goes far enough and we need more actions logged, more sources of input and more accessible Vera integration. Furthermore there is still no way to interface with it in a conversational fashion. For example you wouldn’t be able to ask it a question like “what did I do on the first Monday of last September?” even though Vera has a bare bones rudimentary form of a Google Calendar interface. I personally have been using Vera Alerts as a sort of primitive method of keeping track of some actions and I personally have begun using Google Calendar to keep track of my activities but this is still all manual for the most part and the amount of information I can log and later review is extremely limited due to the hassle of recording it querying it for later inspection.
With the advent of PLEG we can now use more complicated logic without resorting to code. This is probably the single greatest advancement with Vera I have seen since I first obtained the device close to three years ago. Absolutely the only related feature PLEG is lacking in is a natural language interface. I envision an interface where I can say something like “every year for the entire month of October turn the porch light on 30 minutes after sunset and off at 10 PM” or some such. This will be the real heavy development burden. The developer who brings this to fruit would revolutionize home automation and could well end up a very wealthy man.
I am thinking that Vera’s greatest use is to be a controller. I think any attempts to make it do more should stop there. I believe that the next step should be a software package which provides a DB and a natural language interface. As a controller and a physical method of integrating devices I believe that Vera is well suited for the task. What we desire from it very well may be beyond the scope of its keeper’s ability and vision. Not to mention beyond the ability of the hardware.
Thoughts?
(1) The Blog of Ian Mercer http://blog.abodit.com/
(2) Event logging database by forum member Quinten http://forum.micasaverde.com/index.php?topic=15245.0
.