Workshop: Call for Participation

Application deadline: September 10, 2019  EXTENDED: September 25, 2019
Acceptance notifications: September 24, 2019 EXTENDED: October 2, 2019
Workshop date: November 9, 2019

Email your submissions (3-5 page papers) to:
workplacecscw2019@gmail.com

Despite advancement and investment in ML/AI in business applications, understanding workers in these ML/AI work contexts has received little attention. As this category experiences dramatic growth, it’s important to better understand the role that individuals play, both individually and collaboratively, in a workplace where the output of prediction and machinelearning is becoming pervasive. There is a closing window of opportunity to investigate this topic as it proceeds toward ubiquity. CSCW and HCI offer concepts, tools and methodologies to better understand and build for this future.

We invite practitioners and academics from a broad spectrum of perspectives and backgrounds to this halfday work shop to bring early contributions toward: the role of the worker in these settings, establishing trust with users of the systems, design best practices, and methodology from an HCI point of view.

We encourage position papers of 3-5 pages in the ACM Extended Abstracts Format describing original research and an articulation of a contributors interest and experience in workshop topics. Submissions will be reviewed by the organizers based on both relevance and originality.

Papers selected for the workshop will be made available on the workshop website beforehand. Contributions must be submitted via email (workplacecscw2019@gmail.com) by September 25th, 2019 (extended deadline). At least one author of each accepted paper must attend the workshop, all participants must register for both the workshop and for at least one conference day.

Workshop activities will include a guest speaker, several rounds of presentation, Q&A and discussion. Throughout the day, we will identify themes andemerging lines of inquiry. From these themes, we will identify a journal to publish the output of the workshop or, alternatively, will investigate the possibility of an edited book. We will make this decision based on the perspective application exercise and the papers submitted, after discussion with participants.

1. Position papers: opinions, visions, perspectives, points of view

2. Design studies: designing and evaluating ML/AIsystems in organizational settings, challenges, lessons learned and best practices.

3. Methodological reflections: reports about fieldwork, reflections on actively involving theworker in the design process, best practices and lessons learned.

The workshop will have a multidisciplinary emphasis, including work on the technical details of the ML/AI system itself (designing the system technically in light of what is happening in the workplace), understanding and improving UX for collaboration and individual use of ML/AI based business systems, Understanding and designing ML/AI business systems within organizations.

The maximum number of participants is 25 workshop participants (both academic and practitioners). We seek submissions from participants involved in CSCW, HCI and professional technical or user experience capacities and focused on machine learning in a professional or business setting. Researchers and practitioners will be invited through a call for participation through the organizers various networks and relevant mailing lists.

The goal of this workshop is to examine machinelearning and artificial intelligence in the workplace, as a design space. We aim to broaden enquiry into the role of Ml and AI algorithms in the workplace by seeking contributions which are concerned specifically with contextual encounters with ML and AI, with an emphasis on such matters as:

1. The interpretive work that takes place around ML outputs. How do professionals and others make sense of these outputs and how do they deploy them in the context of their working lives? What is the role of the worker in ML based industrial settings and what are the organizational implications? How is work approached and accomplished in light of this technology? What strategies are used? What is missing in our current accounts and assumptions around this work?

2. Ethics and trust: what are the ethical dimensions of ML/AI in business settings? How can we design systems that look beyond the simple outcomes of immediately apparent design decisions and consider the full ecosystem. How can these systems engender trust of the work at hand and the system as awhole.

3. What narratives are produced when communicating the results to others? What communicative requirements might there be to best facilitate an understanding of Ml and AIapplications by those who might need it?

4. Design best practices: What are some guidelines for individuals and groups interacting with ML/AI applications in a professional, work or business domain? Howcan new ML/AI experiences be designed for, and integrated into the everyday goings on and existing technology in the workplace?

5. Research Methodologies. What new problems of access, research management and analyticprocedure need to be dealt with whenundertaking research in these areas? How dowe equip researchers to navigate theorganizational, legal, ethical and practical circumstances of work in light of these issues,and how can we best equip researchers take onthe perspective of workers and their methods while understanding this work, and design forthem [18]? What methods are currently used to study ML applications for consumer/personalconsumption and what are their applicabilityand shortcomings? E.g. how do we design for ML futures [16], develop and evaluate bestpractices [1], account for both normative and contextual considerations [e.g. 18] among other ML related challenges?

We will do so by connecting researchers and practitioners within the “human-ML interaction community” and include interested parties from theCSCW research community. We want to encourage discourses that raise awareness of interactions with MLin the workplace at the individual, group and organizational level. We would like to eventually understand how people are currently doing their work in these settings, the issues they encounter. We further seek to explore futures that involve better designed systems for that work. In the workshop we will begin to explore that design space by sharing our work, and point to new and unexplored opportunities for further research.

Public and academic interest in Machine learning (ML) and artificial intelligence (AI) has accelerated in recentyears [5]. In parallel, business application investmentin ML has intensified, with the marketing of intellectual property leading all categories of investment [3].

Despite prolific business application, efforts tounderstand the design space in the light of ML have focused, we suggest, on limited areas such as:

1. Better understanding approaches to the consumer [eg 1, 10].
2. Developing a foundational perspective of HCIand AI [e.g.7]
3. Business use cases that exclude workplace context [eg 9,19]
4. ML as a deterministic force in design, deemingprevious design approaches as obsolete [20]e.g. “ML is the new UX” [24, 22]

Moreover, despite (variously) largely optimistic, andsome dystopian views of ML and AI among popularmedia and the general public [6], questions remain onthe topic of trusting tools to “do their job” in theworkplace. ML/AI trust research has focused on publicperception of the transparency of systems [17,12,9, 5] and how consumers reason in various trust relatedways when faced with the adoption of ML basedautomated technologies [16]. To our knowledge therehas been little attention paid to trust of ML/AI from theperspective of the worker in the workplace, in particularas it relates to the tools they use as part of theireveryday work [e.g. 13, 2]. Further, trust in theefficacy of automation is an important thread here, andthis has remained largely unexplored. This is becomingincreasingly urgent, as ML/AI technology becomesmore pervasive in daily work. We have a timelyopportunity to understand multiple facets of trust andwhat this means in a workplace context.

In order to better understand, plan and design for afuture increasingly inundated by ML/AI technology it isprudent for us to move beyond examining some of thethreads above and explore the role it plays within theworkplace. That is, how can we consider AI and ML asone piece of a system includes the worker, theorganizational context, and the various technical andnon-technical collaborators, knowledge and otherresources that comprise will help us better understandtrust and design systems to support better work. Aboveall, this means an approach to understanding the placeand role of new technologies in the light of theirintroduction into organizational settings as opposed toconsidering them as separate entities doing work ontheir own. Because of this, it is equally important for usto investigate the process of developing and designingthese technologies and make them most appropriate tofit human and organizational needs.

Michael F Clarke is a Director of Research at Facebook leading the signals and delivery research groups. He helps engineers and designers build business and eCommerce products. His academic contributions have focused on collaboratively working with automated marketing systems and related data tools in the workplace.

David Randall is a Senior Professor at the University of Siegen. His research is mainly to do with CSCW/HCIand the use of ethnographic methods for design purposes. He is also visiting professor at Linnaeus University in Sweden.

Joseph Gonzales is a UX Research Manager at Facebook focusing on researching design spaces to better accommodate business use of ML/AI tools in e-Commerce.

Richard Harper is Co-Director of the Institute of SocialFutures (ISF) at the University of Lancaster and Professor of Computer Science at the same institution. He is the Principal Investigator on a Leverhulme Trustdoctoral training centre on Material Social Futures. This entails looking at the relationship between new material forms of computing and energy production and thesocial arrangements they enable. Key to this research is seeking a balance between material possibility and environmental impact.

Nozomi Ikeya is a professor at the University of Keio in the Department of Humanities and Social Science. Her main interest is in researching issues such as sharing, creating and handing down knowledge and expertise inorganizational settings and communities by examining and understanding how knowledge and expertise is produced and shared in and through visible practices (and in particular technology mediated contexts). Her research has resulted in the co-design of services andtechnology with stakeholders.

Thomas Ludwig is an assistant professor in the field of cyber-physical systems (CPS) at the University of Siegen. His research focuses on the human-centered design of cyber-physical systems as well as the impact of these CPS on current work structures and practices. Application fields range from manufacturing settings tocrisis management.

Michael Mair is a Senior Lecturer at the University of Liverpool in the department of Sociology, Social Policy and Criminology. He is an ethnomethodologist whose research falls into two main areas: politics, government and the state; and the methodology and philosophy of research. The focus of that work includes the politics of accountability in different settings and methodological practice in the social sciences, including qualitative, quantitative and digital methods.