My 2 cents. (Hey, everybody else has got theirs so far..)
UPB is nice, safe conservative and frankly boring. That's not a bad thing though, but its not going to appeal to the folks who buy X10 switches. There is no "zing!" to justify the price premium. Smarthome is positioning insteon (via Icon) to blow X10 right out of the water. If they succeed in getting the $20 icon switches into the entire Home Depot chain after the trial run, that is really going to make waves.
UPB is nice for installers to "sell" a product with decent margins. But if smarthome doesn't screw up and shoot itself in the foot and PCS/SA/HAI don't take enough countermeasures, Insteon could easily outsell UPB by 100 to 1, or even 1000 to 1. Insteon is currently sitting in a very good position to become "the" X10 replacement. X10.com may have to take down the "X10 is THE standard for home automation" off their web site.
UPB is a fully open programming spec and really easy to develop for (on any platform!) but that depends on people finding it interesting to develop for. Insteon's programming specs are trivially easy to get, and almost open. Again, they work on any platform. Add the price difference and the fact that you can get backwards compat X10 for free along with Insteon and it suddenly becomes rather interesting to develop for. (And a pox on Zwave for keeping their specs locked down!)
If smarthome can manage to establish themselves as the ubiquitous standard device, then it will be very hard for UPB/Zwave/etc to compete against the critical mass. It isn't hard to forsee some point when new products are launched with native support for Insteon first, X10 second, and maybe UPB/Zwave/etc at some point later via patches, add-ons and plugins. And that turn of events will make it a tougher sell for Installers to sell their premium or niche devices. Sure, that doesn't negate the reasons to use UPB, but it's going to be tougher to sell when the customers have heard of Insteon as the replacement for X10 but have not heard of UPB orZwave.
All it takes is something like the Windows Media Center to pick up some basic HA support out of the box and use X10 or Insteon and suddenly its on the radar of a group of people many thousands of times larger than the current potential market.
It doesn't matter if UPB is better. What will matter is that UPB switches are noisy compared to silent Insteon switches (at least my SA and HAI ones are distressingly noisy, as is the PIM) and UPB keypads and controllers are seriously ugly. What will matter is that Windows MCE will be able to talk to X10 and or Insteon without having to run an external custom tool like UPStart.
And then there's the nagging gotchas and missing features that turn up with UPB. There is still no active phase coupler/repeater that mortals can buy. And the annoying switch latency in 'snap' mode where it takes about 1/3 of a second to turn the local load on or off. Or that the nice SA switch bodies don't fit in the older style (non metal, non plastic, some sort of composite material) J-boxes with the thick screw posts. Or that SA don't make relay wall switches. Or that running other manufacturers dimmers in the 20%-30% range can utterly destroy your UPB network (I assume the triac firing at around that point on the sine wave causes interference for the UPB signal). Or other triac based variable heat devices. I've yet to find anything that interferes with Insteon, apart from boosterlincs. My washing machine with its electronic controlled variable speed synchronous motor drive system (stepper motor) totally wiped out X10, but doesn't seem to hurt Insteon. My 60KHz aquarium lighting ballasts wipe out X10, but not Insteon.
But Insteon is far from perfect. The ApplianceLincV2 has serious problems controlling fluroescent lights. SwitchlincV2.0's (not v2.1 or v2.2) seem to have a 'sudden death' problem. I've lost count of the revisions of the PLCv2, and I've heard that the LamplincV2's are now all the way up to v2.6 (from v2.0.. if at first it doesn't work, try, try, try, try, try and try again?).
Anyway, that's all pie-in-the-sky stuff. Smarthome could still shoot itself in the foot. The UPB folks might pull a rabbit out of the hat and produce cheap, quiet switches (no buzzing, both relay and dimmer versions) that fit in all J-boxes including the older ones with thick screw posts. Zwave might suddenly see the light and open up the specs. X10 could even shock us all and magically pull X10plus out of the hat and eat everybody's lunch. But the UPB, X10 and Zwave folks can't help but notice those menacing torpedos streaking in their general direction. Smarthome have seriously raised the stakes.
Anyway, back to earth again..
Insteon and RF.. Insteon under-uses RF right now. I think smarthome are too busy to even think about it yet. They've always said it is for things like the X10 style key-fob remotes and wireless devices like motion detectors etc. Among other things. I suspect they're kinda hoping that somebody else will make the devices so they don't have to split their attention even more. They're already obviously struggling with the core infrastructure devices, let alone other fruit.
It isn't hard to imagine a microcontroller and some support glue to embed Insteon in things. They've talked about this sort of thing right from the start and that makes things real interesting. Imagine if garage door openers start embedding Insteon inside (yeah, right! they make too much money replacing lost "proprietary" remote openers!) or if your sprinkler systems get Insteon onboard at some point. Or your automated blinds/drapes come with Insteon-RF onboard. Oh wait, somfy have already announced that they are taking the first steps on this.. Or the application in their original press release announcement.. having a washing machine or dish washer send an insteon event when a cycle is finished so that you HA system can announce it for you.
I for one would die for a simple Insteon-RF controlled "relay". I can imagine no end of applications for a low-voltage Insteon-RF transciever running on 12V or 24V circuits to control relays. Skip the PLC interface and the costs of high voltage interfacing and do a simple low voltage RF transciever that drives a small 1A or 5A relay. Or a cheap contact sensor. You get the idea. (I've switched several of my aquarium systems from UPB Appliance module + 24VAC transformer + relay or solenoid to using a remote
http://ontrak.net I/O module. But that means long wiring runs from central points. I'd love to have a 24V chain of devices with a transciever at each actual node.)
Anyway, UPB doesn't have any apparent RF ambitions. If you want to integrate RF keypads or remotes into an UPB system, you have to use a second technology. You get integrated RF with Insteon. You can stick as many signalinc-RF's around as you want in order to get the best coverage and when RF remotes etc finally appear, you get automatic support for them without a second system.
RFID is cool etc but it is still a second system. And it doesn't have remotes with buttons. I'd find an RFID presense based system a bit creepy. And I still want to be press a remote control button to override the automatic blinds if I want to. (I tracked my bluetooth phone for a while but found I kept leaving it downstairs.. with my car keys. So an RFID tag on my keys wouldn't work either..)
On a complete side track, we generally don't ever lock our house. Actually, I think we once locked our front door when we went on a holiday out of the country. But the back and side doors were open for the neighbors who were feeding my fish while we were away. I know several of our neighbors do the same.
).
