Thanks again for replying - hopefully you understood my frustration.
However, I still don't believe you have begun to refute my mechanism.
One problem is that many of your arguments are based on gross morphology of organisms (dogs growing feathers), when the point at hand is the addition of new genetic information:
A canine kind will not change into a new kind of animal, this would be a increase in information.
Again - this statement has nothing to do with one gene becoming two useful genes with divergent function.
Another problem is your terminology:
When we talk about a mutation in a dog, we mean a change in structure.
No 'we' don't, a mutation is simply a change in the sequence of DNA, and says nothing about the impact on gross morphology. You are arguing against my DNA-based mechanism with an absurd organism morphology argument.
It is a mutation, which is downhill.
Again, incorrect. A mutation is neither 'uphill' or 'downhill' - in any case, I've provided a mechanism by which one gene can diverge into many, and others have provided cases of genes becoming more efficient or gaining new properties (sounds both 'uphill' and like additional genetic information).
Evolution must be observed all around us. We must be able to point to millions of instances were a dog grew feathers, then many mutations later into a bird.
Evolution does NOT predict that we would regularly see such gross morphological changes and speciation within our lifetimes. Also, you reveal your simplistic view of evolution here, that one mutation can cause feathers to sprout, and a few more turn a dog into a bird.
However it seems that a mutation has remixed the genes etc causing more information. Which is not case as it cannot add new genetic data.
This statement is particularly disappointing to me. You ask for a mechanism for the 'addition of genetic information,' so I give you one. You say it seems plausible, but it has to be wrong. Why? Because you claim "it cannot add new genetic data". What does this mean? It sounds like you are simply being stubborn in the face of a very plausible mechanism.
In any case, you still have NOT refuted the validity of the mechanism: duplication and/or rearrangement, followed by mutation, produces more genes with divergent function.