The only company with a stable history is Insteon.
You left out Vera, Universal Devices, Indigo, Homeseer, Control 4, and Crestron; albeit the latter two are very high end. But all of the above do what they were designed to do very reliably.
As someone who has been around for 2 years with SmartThings, both as a user and developer, what really concerns me is not that things are still broken, but that they haven't been able to follow through on virtually
anything. Seriously, just about every aspect of their business (except the marketing and getting acquired) seems paralyzed. Here is a short list I could add many more to:
1.) Mobile apps are in a constant state of unfinish, with admitted placeholders for future features that are never implemented; not to mention are a train wreck in general.
2.) The inability to publish a decent changeLog.
3.) Official integrations have been at a virtual standstill for years, sad given the relative ease in which they're created.
4.) IDE Simulator essentially broken.
5.) Scheduled events being lost, and then AWOL going forward.
6.) Polling has never worked reliably.
7.) No migration path, officially announced it would be delayed to end of year (end of last year).
8.) V2 hardware released unfinished over 6 months ago, promise of local "appengine" still unfulfilled.
9.) Community based app submissions not being reviewed (WTF).
10.) Premature, broken UK rollout.
11.) Lousy documentation.
12.) Inadequate support.
13.) Inadequate QA.
14.) Developer calls are a joke (go view some on YouTube; do these guys seem lost or what?).
I could go on and on but surely you get the point. While I still use SmartThings in a limited fashion, I have stopped developing and building on it until they can demonstrate some level of progress/competence, which has been missing since the get go. Here's hoping that will happen before someone else comes along and eats their lunch.