Web frameworks comparison: A model


I have been reading different web frameworks comparisons. The most accurate I have found is a little bit old, 2008, and it is part of a project by Polytechnic University of Madrid now abandoned. This article was written by José Ignacio Fernández-Villamor, Laura Díaz-Casillas and Carlos Á. Iglesias. It can be found here. It is a comparative study focused on agile web frameworks, high productivity frameworks. It compares Ruby on Rails and its Java partners: Roma, Grails and Trails.

Comparing web frameworks lacks objectivity and it is a hard task. It has been impossible to reach a neutral view until now. Nevertheless, they try to reach a model described by a set of parameters:

  • Domain & Persistence
  • Component orientation
  • Presentation
  • Service orientation
  • Usability
  • Security
  • Adoption
  • Testing

To evaluate these parameters, they ask some questions and give them a grade. For example, for Domain & persistence, one of the questions is: Is domain definition automatically inferred from schema data? and it has a 20% grade. Answers to the questions can be given a percentage too.

Results are shown in a graphic way which leads us to understand best framework does not exists, it is a question of parameters.

Graphic result of the agile web frameworks comparison
Agile web frameworks comparison. Source Fernández-Villamor, Díaz-Casillas, Á. Iglesias 2008.

Nevertheless, I found interesting to know which would be the result if an addiction of all the grades is calculated:

Roma Rails Trails Grails
750 650 600 475

Roma is the winner but I have not heard a lot about this framework recently.

I guess that things have changed a lot in the last 4 years, what do you think? But the model seems valuable to me. I will try to review questions and answer them for analyzed frameworks.

Thanks for reading. Comments, criticism and sharing will be appreciated.