I'm really having a hard time envisioning a functional improvement on the standard Kidde smoke detector. There are two states: "it's cool", and "hey guys. There might be a fire".
I really can't imagine much improvement on that. If the level of sensitivity that induces a couple false alarms a year gives me an extra 10s of notice in the event of an actual fire, I'll be fine without a configurable thermostat.
In my uninformed opinion (as, frankly, is everyone else's before there's actually an announcement), there would need to be some non fire-related benefit to make it worthwhile.
You're describing a binary fire/else system. I would consider that a minimum viable product, and this is a very mature market. I'm expecting more than that, and I'm willing to pay for it.
(You not being able to think of any improvements doesn't mean they don't exist.)
A better fire alarm could connect to my Nest via wifi and add a remote temperature sensor to my existing thermostat system, allowing it to increase overall accuracy/efficiency. It could talk to other detectors in the same manner allowing for a whole-house fire alarm system that's doesn't require running wires. It could text me when it's battery needs replacing instead of emitting a high-pitched chirp I have to hunt down at 2am. It could do the same thing via notification on my existing Nest app. It could detect humidity levels in my house and increase it's sensitivity if appropriate. It could notify me of alarm (once again via text or through the app) while I'm away from my property. It could snap picture with a low-res 360 pano camera and send them to my phone so I have a record of how the fire started; if the latency was low enough it could be used to help fight the fire.
I'm not an expert in this field. I'm an interaction designer. It took ten minutes to think of those improvements. (I'm just brainstorming, but surely one or two is feasible.) I'd bet a year's salary that the team at Nest labs is better at this than I am. They've got something up their sleeve or they wouldn't bring it to market.
Anyway, I have no affiliation with company, but I really like their thermostat and I'm expecting good things from them as they expand their product line. Particularly, the kinds of things that will get to me to pay a premium over the existing stuff at my nearest big box store. I paid similar premium for my Nest.
I really can't imagine much improvement on that. If the level of sensitivity that induces a couple false alarms a year gives me an extra 10s of notice in the event of an actual fire, I'll be fine without a configurable thermostat.
In my uninformed opinion (as, frankly, is everyone else's before there's actually an announcement), there would need to be some non fire-related benefit to make it worthwhile.