NLP: Build A Better Solution

I have been busy doing therapy lately, and that tend to focus me on NLP. So: More NLP. Below comes an analysis pattern I use for finding new, better solutions from old ones. Running the pattern is quite a bit of work. I have only used it for areas where I am already an expert, on research directions where all proposals have some drawbacks. On these problems, I have gotten world class solutions (new research directions). I would be interested in hearing about results from trying to use the pattern on simpler issues.

The basic idea is to use existing solution proposals to elicit goals, and then use these goals to create new solution proposals that again can be used to elicit more goals. As I work through the technique, I try to make the goals quite concrete. My benchmark is that I should be able to rank each solution "Very good/Good/Neutral/Bad/Very bad" on each goal.

1. Write down the basic issue you are investigating. (E.g, "Solving my present depression problem.")
   
2. Write down all proposed solutions. Include "keeping the status quo" or "do nothing" as one of the solutions.
   
3. Compare each proposed solution to each other proposed solution. Write down the comparative drawbacks and strengths.
4. Collect all strengths, from everything, into a single list of goals.
   
5. Take all the weaknesses and turn on their heads, and add to the goal (strength) list. For instance, if the drawback of one proposed solution is "It is a lot of work", then "Small amount of effort" could be added to the goal list.
6. Go through each of the proposed solutions, and compare that solution to every goal in the goal list. Write down the results.
   
7. Go through each of the proposed solutions, and see how you can change that solution to be better for each goal. Often, you can mix aspects of two proposed solutions to come up with a new and better one.
8. Repeat the process until you have a "Good enough" proposal.

I'll update this post with an example later.