I vote for Dick & Carey! (I am only half-joking)

What I find interesting is how people get attached to a specific process model and they don't let it go. Where I went to school, it was Dick and Carey, and ADDIE is seen as too simplistic, and a model that does not work as well. We've also covered ASSURE.

Honestly, It's just a model. It's not like we're debating cognitive theory that describes and defines everything about creating instruction. ADDIE does have a future not because it's a common language that people know, but because it's a model that can help you start thinking about how to implement instruction. You may start with ADDIE, but your workflow might lead you to use HomeGrownModel more than ADDIE, ASSURE, Dick & the gang.

I think the problem here is that people tend to confuse a theory, with a model. ADDIE is a model for getting things done, it's a model for a process. It's not a cognitive theory that can be proven or disproven. ADDIE is recursive. At the evaluation stage of things you can go back to analysis if the results are not what you intended them to be.

10 different IDs following ADDIE might give you 10 different results because ADDIE is a model, not a theory. It's like saying that you know how to drive - driving might get you down town, or across the border into Canada. Two different results, but the application of SGBL (steering,Gas, breaks lights) got you there. If you get to your location and you see on your GPS that you are not where you need to be, you can go back to steering and throttling to where you need to be. This analogy has had its time I think ;-)