To build or not to build...

sedwo

New Member
Hello all,

I run a little software company that develops for the Pocket PC platform; and have recently become very interested in home automation, with hopes to incorporate it into my business.

I have been in the robotics and automation industry for a while now and have always loved little devices controlling things. And so, my initial thought is to create a software package for home automation that runs entirely on a tiny Pocket PC device, controlling serially an Insteon network.

But as I research the various other software packages out there, it has occured to me that while a Pocket PC PDA makes an excellent client device; it is too limiting to be a full controlling server for a home. Yet I'm curious to hear your feedback on such an idea. Would you buy such a setup?

This then also shifts my thinking into becoming more of an integrator (COTS, etc.), than a software product developer with yet another home automation suite to push (but on the PDA platform).

I'm gathering my market research here and greatly appreciate any input you kindly have.

Thank you. ;)

Sebastian
www.appliedpda.com
 
Sedwo,
The networked PocketPC is pretty good as a user interface. It is not good as a HA controller. Not hardware extensible enough or have the CPU to do HA logic. There would have to be tons of drivers written for PocketPC, and very few companies would be interested to do that work for the very limited market of a PocketPC controller.

A UMPC is debatedly a better UI for HA than a PocketPC as it has IMHO the right balance of CPU / screen real estate, local processing ability, wireless options, etc. But, at twice the price, with less battery life, and somewhat less portability.
 
Personally I see the PDA as a complementary or additional method of control. I don't think I would ever see the PDA as just a fancy standalone remote and don't think it posseses the power to do more of the needed functions to be really useful. So, say I use CQC and have my touch screen(s), etc as my primary control methods, I do see it being really nice to be able to do some of this control from a PDA as well, but not instead of.
 
I personally would love to see something like a custom linux distro for Linksys based router, which supports HA etc. Look at what they can do with the modified images for the current Linksys wireless routers. If you somehow could manage to add a few serial ports to that router, you would have a winner, especially if you will allow other people to make their own changes to the Linux os.

As for a PDA backend, I can't imagine it being fast enough, even with the newer and faster cpu's, it just isn't built to be doing JUST home automation, so there could be a lot of wasted resources there.
 
Not that I think it's the best choice, but I don't see why (for processing power alone) a PDA is any worse than an Elk M1, for instance. Remember, there are different definitions of HA.
 
The Elk M1 doesn't have to worry about power management, high res displays, touch screen processing etc, it was built to do these few things, while a PDA is meant to do many things (imo).
 
electron said:
The Elk M1 doesn't have to worry about power management, high res displays, touch screen processing etc, it was built to do these few things, while a PDA is meant to do many things (imo).
But a program on the PDA does not HAVE to do those things.
 
I wrote my own PDA user interface application using Embedded VB 3.0 (because it was free) and it does a nice job of establishing a Wifi connection to a main system. I find a PDA too slow for web based user interfaces, which seems to be the current trend in an attempt to support as many different client machines as possible.
 
A hand held cannot ever really be the whole automation system. At best it would have to be combined with some sort of network capable hardware like a GC-100 or something, else there's little it could do on it's own. And it wouldn't make for a very powerful controller either. I would argue that such devices should always be combined with a more powerful back end system for which it just provides a front end. If you had one that spit out IR, you could of course use it as a glorified IR remote in home theater. But that wouldn't be all that much of a benefit over a smart learning remote.
 
Another thing to consider is you are competing with applications that already do what you are after. I know most of the home automation packages available have PPC interfaces that connect with the server doing the work. In some cases these are stripped down web interfaces that run very quickly. So if you are doing it for a profit this is something to keep in mind. If it's a learning effort then go for it.
 
I have a PPC version of my Now Playing (Netremote) skin and it works great. I control my lights as well as my 5 zone whole house audio system. Its a great little GUI front end device. Biggest problem is always going into sleep mode and waking up time to reestabolish the network connection.

John
 
Rupp said:
Another thing to consider is you are competing with applications that already do what you are after.
Or put another way, do many other things that people that would like to use that functionality desire as well. If it is Insteon by itself, then it is limited.

That being said, it seems difficult for many companies to add insteon support, so if you created something that they could use and repackage, you might find some takers on the business side. That is another ball game and probably a tougher nut to crack as well.

You should look at CQC, Cinemar, HomeSeer and Powerhome for some better ideas at the way a more comprehensive offering is typically deployed.
 
az1324 said:
electron said:
I personally would love to see something like a custom linux distro for Linksys based router, which supports HA etc.
Shhh don't reveal my secret project.
Not to take this off topic but please tell us more in another thread... Also will it work with other routers? ex motorola WR850G which I beleive will take one of the other files out there
 
Wow! You people are great. Thanks for the feedback.

You have confirmed my suspicion in not using PDA's as a stand alone HA server. While I do believe it can be done, the feasability seems that it won't be worth it, and scaling it will be a problem. For the price of a decent Windows Mobile device, you can get a small PC which can then run any of the commercial software HA applications.

Controlling basic lights and appliances might be fine, but audio/video become challenging for a PDA unit. Leaving it therefore to be a better thin-client controller.

Thanks people! :)


Sebastian
www.appliedpda.com
 
Back
Top