“Let’s do a Doodle poll!” This is a phrase you’re likely to hear whenever a group needs to reach an agreement on a date for an activity. The Doodle scheduling tool has been extremely popular for many years, helping us to bring people together efficiently. This example shows that online group decisions have long since become an integral part of our everyday lives.

“Doodle is probably the best-known example of an online group decision process, but it’s by no means the most exciting one,” says Jan Felix Maly from the Institute for Data, Process and Knowledge Management at WU Vienna. Maly is a computer scientist who investigates the mechanisms that underly such group decisions – his goal is to gain a mathematical understanding of how we can optimize the fairness of these decisions. For this research project, Maly has been awarded a netidee SCIENCE grant, Austria’s most substantial privately funded research grant for excellent online research.

Jan Maly’s research interests extend far beyond the prosaic realm of the Doodle poll. “We want to investigate how fairness can be defined, and how it can be put into practice in an online context,” he says. This ranges from small team planning activities and social media voting behavior all the way to political decisions involving thousands of people.

Can fairness be calculated?

The question that lies at the heart of the FairOGD project is what fairness means in the context of digital decision-making processes. Jan Maly has spent years doing research in an area where such questions are particularly important: participatory budgeting. The idea behind this concept is simple: when a city has money to allocate, it asks its citizens to decide where the funds should go. “The standard approach is that the projects with the most votes get funded,” explains Maly. “That’s also how it usually works on social media: whatever gets the most likes, wins.”

In recent years, mathematicians and computer scientists such as Jan Maly have developed ideas for applying the concept of proportional representation to complex decision-making processes. “Based on game theory models, we look at how decisions can be made in a way that is proportional and fair,” Maly explains. The aim is to develop voting rules and algorithms for decision-making processes that reflect these principles. The expectation is that concepts for fairer budget allocation decisions could also help to ensure greater fairness in other areas.

The range of potential applications is enormous. “We’re talking about a spectrum that ranges from small, everyday decisions among groups of friends to landmark political decisions," says Maly. “In online forums and social networks, for example, the majority often decides which posts remain visible and which do not. These decision-making processes affect many people and should be fair,” he points out.

Basic research with a practical angle

Even though the project primarily focuses on basic research, its long-term goals are of a practical nature. “We want to create the basis for developing technical solutions – for example in social networks, in collaborative decision-making at companies, or in politics,” Maly says.

In addition to the technical aspects, fairness also has a philosophical and ethical dimension that should not be ignored. “It’s not just about developing algorithms but also about understanding the basic principles that constitute fairness in these contexts,” Maly explains. “What’s perceived as fair can vary greatly depending on the situation and cultural background. This is exactly where our research comes in,” he says.

“The grant awarded to this project is an important step towards fair and equitable online decision-making processes,” says WU Rector Rupert Sausgruber. “Innovative approaches like this one provide important contributions for making our societies more democratic and more inclusive,” he says. The FairOGD research project was chosen for a netIdee grant by the Austrian Science Fund (FWF) and will receive €400,000 worth of funding from the Internet Foundation. Apart from WU, the other project partners involved are TU Wien, Université Paris Dauphine, the University of Amsterdam, and the University of Oxford. The project is scheduled to run for three years, so the results are expected to come in no later than 2027.