The methodology of scientific method can be sometimes difficult to understand and implement in psychiatry. Different tools exist to help this process by collecting and interpreting data but they can sometimes be quite expensive. Here we will present free and open source softwares that help to communicate but also randomize, collect, interpret and publish data. These tools can be easily reused and shared. That could improve validity and reproducibility of scientific research. In order to be interactive, we will involve the participants in the collection of the data of a prototypical scientific study, an international cake testing database! (http://bit.ly/cakereport) We will see a study protocol and how we can pregister this protocol defining the primary outcome. Here what is the best cookie from 2 different brands? We will discuss the issue of sample selection. Who will taste the cookies? Randomizer.org will help us to attribute a condition (cookie A or B) to each subject. Limesurvey helps to run easy online surveys. What did you think about each characteristic of your cookie? R helps to analyse and plot the data. What the mean score? How to plot the results? Is the difference significant? GitHub helps to publish publicly or not our data and analysis script. ArXiv systems helps to publish early works and manuscripts before publication. Equator-network.org can give guidelines that could help to write scientific articles. We can promote our work then on wikipedia and discuss what are the advantages and issues to promote your own work on wikipedia.