Civic technologies such as participatory budgeting and crowdsourced policymaking increasingly provide spaces for crowdsourced deliberation, an asynchronous, distributed, and self-selected form of deliberation. To understand the dynamics of crowdsourced deliberation, we examined deliberative quality, dis/agreement, and elaboration (rationale-sharing), and their association with idea generation in crowdsourced deliberation within a crowdsourced policymaking process led by a national government. The deliberative quality of the crowdsourced deliberation was high and positively associated with idea generation. Elaboration of perspectives was a key feature contributing to idea generation. We present considerations for the design of processes and platforms used in civic technologies to foster high deliberative quality, elaboration, and a balance of disagreement and agreement, which in turn can support idea generation in crowdsourced deliberation.

• Fair exposure ≠ fair outcomes. Visibility in rankings does not guarantee equitable shortlisting in online recruitment. • Human and task factors matter. Recruiter behavior, task design, and candidate cues influence fairness beyond algorithms, as we show in user studies. • Design implications. Our findings translate into practical implications for model evaluations, interfaces, and recruiter practices to support equity. Personnel recruitment is increasingly mediated by Applicant Tracking Systems (ATS), which rank candidates for job positions, making them a central decision-support tool in modern Human Resources (HR) processes. Often framed as an information retrieval (IR) problem, the ranking of candidates in ATS is typically driven by relevance to the job position, with algorithms sorting applicants according to a set of predefined criteria. In recent years, fairness-aware ranking methods have emerged to mitigate the risk of indirect discrimination, where the ordering of candidates may inadvertently favor one demographic group over another. These approaches are inspired by browsing models developed for web search and aim to balance candidate exposure based on protected characteristics. However, ATS in recruitment introduce unique challenges due to their high-stakes nature and the decision-making context in which they operate. In this paper, we present a series of user studies that explore the disconnect between fair exposure and fair outcomes in candidate shortlisting. We focus on how factors such as task design (e.g., how recruiters interact with candidate lists), individual representations of candidates (e.g., national origin cues), and ranking order influence both position bias and demographic balance. Our findings show that while demographic balance may be achieved in terms of ranking visibility, this does not necessarily translate to fair outcomes in terms of who gets shortlisted. Through a crowdsourced experiment and in-depth interviews with recruiters, we identify key task-level, individual, and ranking factors that mediate these effects. We conclude that fairness in ATS rankings is contingent not only on algorithmic design but also on the shortlisting tasks they support, as well as the interfaces, strategies, and assumptions that recruiters use when interacting with candidate lists. Based on these insights, we provide implications for the design of algorithms, interfaces, and recruitment processes that support fairer and more equitable recruitment outcomes.

Employers are adopting algorithmic hiring technology throughout the recruitment pipeline. Algorithmic fairness is especially applicable in this domain due to its high stakes and structural inequalities. Unfortunately, most work in this space provides partial treatment, often constrained by two competing narratives, optimistically focused on replacing biased recruiter decisions or pessimistically pointing to the automation of discrimination. Whether, and more importantly what types of , algorithmic hiring can be less biased and more beneficial to society than low-tech alternatives currently remains unanswered, to the detriment of trustworthiness. This multidisciplinary survey caters to practitioners and researchers with a balanced and integrated coverage of systems, biases, measures, mitigation strategies, datasets, and legal aspects of algorithmic hiring and fairness. Our work supports a contextualized understanding and governance of this technology by highlighting current opportunities and limitations, providing recommendations for future work to ensure shared benefits for all stakeholders.

The advent of blockchain technology has paved the way for numerous innovations in online governance, with decentralized autonomous organizations (DAOs) emerging as a prominent development, often referred to as “digital jurisdictions.” Despite experiencing remarkable growth, currently boasting nearly 7M users and $18 billion in assets, DAOs remain relatively underexplored in the existing literature, particularly from an empirical perspective. This study presents a comprehensive framework comprising 15 dimensions to categorize DAOs based on their operational domain, purpose, scope, voting process, and utilization of crypto-tokens. By applying this categorization schema to 40 DAO communities hosted on the Aragon platform, encompassing over 423 000 participants and managing treasuries worth $960M, we shed light on the prevailing characteristics of these DAOs. Contrary to assertions made by blockchain enthusiasts, our analysis reveals that DAOs predominantly operate in financial and technological domains, primarily offering blockchain-based services. Additionally, our investigation into their governance structure exposes limitations in terms of democratic participation, as decision-making power typically correlates with the number of tokens owned by the voter, resembling plutocracies rather than true democracies. We believe these findings will facilitate researchers’ comprehension of this innovative form of governance and aid practitioners in designing future DAOs with greater effectiveness. Furthermore, our analysis can be replicated on other platforms or at different time periods to validate and contrast our conclusions.

Today, digital platforms are increasingly mediating our day-to-day work and crowdsourced forms of labour are progressively gaining importance (e.g. Amazon Mechanical Turk, Universal Human Relevance System, TaskRabbit). In many popular cases of crowdsourcing, a volatile, diverse, and globally distributed crowd of workers compete among themselves to find their next paid task. The logic behind the allocation of these tasks typically operates on a “First-Come, First-Served” basis. This logic generates a competitive dynamic in which workers are constantly forced to check for new tasks.

Crowdsourced civic engagement is a novel form of democratic participation that allows citizens to share their ideas and deliberate in a multitude of diverse participatory processes that are emerging all over the world, influencing, often with binding power, urban plans, city budgets, and even legislation, among many other forms of public policy decisions. As a result, hundreds of thousands of civic contributions are produced as ideas, comments, and proposals circulate among citizens and between them and government officials, generating an avalanche of mostly unstructured data, which decision-makers have difficulty to manage. Sentiment analysis techniques have the potential to process and classify these contributions in ways that can make it easier to make sense of them. In this paper, we present the design and implementation of a rule based sentiment analyzer that integrates a lexicon, optimized for the Spanish language and the application domain of civic contributions. We present the results of our first evaluation and discuss the aspects of the proposal that have room for improvement in future work.

This paper examines the sense of presence, attitude change, perspective-taking, and usability of a split-sphere, first-person perspective 360 degree video about gender inequality, in which people can choose to watch the narrative from the male or female character's perspective. Sixty-seven participants were randomly assigned to watch (1) the video in 360 degree split-view in a head-mounted display, (2) the same film as 180 degree in a HMD, or (3) a flat control version of the video on a laptop. The 360 degree split-sphere increased the viewers' feeling of personal responsibility for resolving gender inequality, desire to rewatch the video, fear of missing out, and feeling of missing the full story. The 180 degree video created the strongest sense of presence, embodiment, and understanding of the character. However, people with greater egocentric projection onto the male character felt less responsible for resolving gender inequality, particularly in the 360 degree split-view.

In this paper, we examine the changes in motivation factors in crowdsourced policymaking. By drawing on longitudinal data from a crowdsourced law reform, we show that people participated because they wanted to improve the law, learn, and solve problems. When crowdsourcing reached a saturation point, the motivation factors weakened and the crowd disengaged. Learning was the only factor that did not weaken. The participants learned while interacting with others, and the more actively the participants commented, the more likely they stayed engaged. Crowdsourced policymaking should thus be designed to support both epistemic and interactive aspects. While the crowd's motives were rooted in self-interest, their knowledge perspective showed common-good orientation, implying that rather than being dichotomous, motivation factors move on a continuum. The design of crowdsourced policymaking should support the dynamic nature of the process and the motivation factors driving it.