Kind of echoes what I tried to say. If you don't already have some basic knowledge then the learning curve will be astronomical.
Yes it will be a steep learning curve but not impossible. If you are the sort of person that enjoys the challenge of taking something that doesn't work, figuring out (by research, learning and testing) what is wrong with it and making it work then you can succeed on this learning curve and will get great satisfaction in the process. On the other hand, if you are the sort of person that just wants to blindly follow step-by-step instructions (without understanding) and throw your hands up in the air and quit as soon as something doesn't do what it's supposed to then you will find it very hard.
Seems to me you have excluded most people I know with this division into 2.
What about all the people who want to blindly follow a step-by-step guide in the first place; get a satisfactory result which will boost their confidence and allow them to see that this is not impossible and they can achieve their goal of packaging an application?
With that small success under their belt they can move on to something a little more difficult which will maybe throw up some problems which they will then have to investigate, research and solve .... with help of the experienced packagers ..... and when successful they get the satisfaction of another hurdle overcome.
I believe the majority of interested parties would fall into this category, and be able to try packaging under those circumstances.
Some might then have the chance to continue on the path and become experienced packagers.
It was for this very reason that I suggested something like a 10 step packaging course, starting with something simple and moving up in difficulty as the previous packages were successfully built.
There is one further suggestion I would make ...
there are a number of simple-to-build (in the whole scheme of things) packages/utilities in the repos which could not only be used as introductory examples for new packagers, but their continued maintenance might also be taken up by new packagers.
An example of such might be LiveUSB-Creator, PCLinuxOS-pvr etc etc ..... (hopefully I am not stepping on anyone's toes) ....... which could be put into a pool for new packagers to 'cut their teeth' on.
.... just some thoughts ......
In a curious way, learning packaging by piecing together sparse and obscure information is actually a good introduction to what the actual process of packaging is like a lot of the time. The upstream developers typically do not supply comprehensive documentation with their software and it often relies on "libraries" whose documentation is also sparse, variable quality and scattered all over the internet. I hope I'm not putting people off but I do think we need to be realistic about what is involved.