The social protocols organization has the goal of creating algorithms to enhance collective intelligence. We consider a social protocol to be a set of rules and mechanisms that govern the interaction of users on social computing platforms. We started out by analyzing and reverse engineering the Hacker News algorithm since it is very simple by design and has been working pretty well for many years. We scraped the frontpage every minute for several months to obtain a time series data set (available here) where we could exactly recreate a story's entire lifecycle. We then developed a new metric to measure attention for stories and estimated how much a story over- or underperformed and built a hacker news client where we display this metric for users to play around with it. Currently, we are applying our learnings from these projects on a new social networking site that we hope will enable large-scale constructive discussion and collective intelligence. All our work is open source and available on GitHub.