Organization of the cookbook

[lafita]

The goal of the BioJava Cookbook is to introduce users to the applications and demonstrate the possibilities that the BioJava library offers using a set of examples.

The BioJava Cookbook has to be a collection of self-contaned recipes to common bioinformatics problems. Each recipe has to start with a problem statement and its biological motivation. Then, a solution to the problem will be given using BioJava code in a set of numerated steps with brief explanations.

The recipes should point to other sources of information, such as the BioJava Tutorial or the javadoc, but never other recipes. They should only assume knowledge about Java, Maven and how to install Biojava.

Additionally to the explanation text, a working Java program that actually solves the problem has to be attached to each recipe. This way, the code described in the recipe can be compiled and tested.

Each recipe can use a different BioJava version, since the ultimate goal is to solve the problem at hand. If there are major API changes between versions (or solving a particular problem has become easier) and re-implementing the solution is of interest, different recipes (one for each version) can be created. Recipes can also have a limited amount of other dependencies to help in the solution, if needed.

To implement all of that, each recipe is delivered as a separate folder in this repository with a README file that contains the explanation (in markdown). Java classes are organized inside the folder using Maven. At least a Java class and a test are needed to check that the code compiles and produces the desired output.
Therefore, the whole Cookbook repository will at the same time be a Maven project with each recipe being a submodule.

If anyone agrees with this setup, I will generate a template for the recipes and start migrating them from the old wiki pages.

[josemduarte]

Sounds good to me!