2 years ago, I bought a controller to automate my reef keeping task. The moment I received my controller I was disappointed. No ethernet? Easy fix. Emailed the company with my solution and some suggestions.
Basically the reply I got was, thank you we are always looking to improve our products.
Now 2 years later they have a new controller with some features I asked for, but at the price of a new controller which is 3X more than the controller I originally bought.
Now here is my biggest rant at this point. Ethernet from my suggestion could've been sort of easily implemented. You just get a serial to ethernet device, such as the telnet Site Player... but for it to be actually useful other than just getting current status, commands had to be made public. Their Web Software that runs on the PC uses these command to send and recieve not only status but for control. <-- This is just my opinion and my speculation.
Of course, I started work on WRRC just a few weeks before I found out the new controller was coming out. I was designing it around my requirements. A real operating system, just like using a PC, but I wanted an embedded which are more reliable for 24/7 operation.
When I found out the then Belkin router had a serial port and it was free after rebate, my work began. A week later, I fried my Belkin and replaced it with a $39 Linksys. More flash meant at least I can do a proof of concept, then go back and try to make it fit a Belkin later.
At this point, I've managed to prove my concept. It is possible to build a controller with all the features I want, easily upgradeable and very flexible. The road ahead is to integrate all the great programs and provide Web User Interface.
Note, I am not a programmer nor an engineer. Everything I've thrown together is all because of public information and Open Source software. I'm just here to prove that it is possible.