I look at the complexity of the Red Lion device the other way around; they have a product that can be programmed through a complex menu-driven method, or through a proprietary serial protocol. They adapted that product to a couple of open fieldbus protocols and the complexity of their functionality still shows through.
On the DeviceNet side, they're really as simple as pie; polled I/O with one byte for action, one byte for selection, and four bytes for data.
I am a DeviceNet expert and a Rockwell employee, so I am naturally predisposed to DeviceNet. I use it for everything; sensors, I/O blocks, drives, overloads, MMI.
I am routinely surprised with how little regard third parties give to the PLC's they'll be working with. Any ODVA member in my region has met with me and left with a stack of configuration example documents using Rockwell DNet scanners, which are arguably the most common in the industry.
But I can see their perspective, too; it's not feasible for them to be Omron, A-B, G-E, Emerson, and Entivity programming and configuration experts, and then to do it all over again and become Siemens experts when they want to do Profibus.
I have a hard-hat that I take with me to startups and field engineering projects. There's a big blue sticker on the back that reads "I LOVE DEVICENET".
Underneath it is a smaller sticker reading "soon you will too".