I am tired of thinking about how to evaluate or select a proper solution. What I need is a methodology. Below is my idea, and I’ve applied it several times, it’s not bad.

Methodology:

1. define features, such as
   • functionality
   • performance
   • ease of use
   • ease of operation
   • scalability
   • relation with current product/platform
   • potential (popularity + community + …)
2. give a weight between [1-10] for each feature i, mark as w(i)
   • the weight value should be adjusted based on the real environment
3. give a score between [1-10] for each feature in each solution, mark as s(ij)
   • I don’t have a standard for score assignment
4. calculate sum(s(ij) * w(i)) for each solution and compare