Dr Bahareh Heravi

The future of journalism is more stable than we imagine. The search for innovative solutions always requires that we take a look back at the past, even if it is to address the opportunities and challenges posed by artificial intelligence in the realm of journalism. Computational algorithms are capable of streamlining such diverse tasks as content discovery, filtering, analysis, production, publishing or distribution, although the financial investment of implementation represents one of the main impediments and generates a visible divide in the present and future media system. Furthermore, the ethical dilemmas it incites will depend on the choices made by its creators, whose mixed profiles will be in high demand.

News organisations have longstanding practices for archiving and preserving their content. The emerging practice of data journalism has led to the creation of complex new outputs, including dynamic data visualisations that rely on distributed digital infrastructures. Traditional news archiving does not yet have systems in place for preserving these outputs, which means that we risk losing this crucial part of reporting and news history. Following a systematic approach to studying the literature in this area, this paper provides a set of recommendations to address lacunae in the literature. This paper contributes to the field by (1) providing a systematic study of the literature in the fields, (2) providing a set of recommendations for the adoption of long-term preservation of dynamic data visualisations as part of the news publication workflow, and (3) identifying concrete actions that data journalists can take immediately to ensure that these visualisations are not lost.

This study introduces a synthesised framework for the analysis of data visualisations in the news. Through a close examination of seminal content analyses, their methodologies and findings, this article proposes a framework that consolidates dimensional components of data visualisations previously scattered across this body of research. To transition from incidental and essentialist examinations of visual data artefacts towards a systematic and theory-informed exploration, we consider the diagrammatic dimensions of data visualisations. The offered synthesized framework can serve as a starting point for both theory-infused descriptive purposes as well as more theory-guided explorations. The framework is put to the test by analysing 185 visualisations drawn from award-winning data stories. Findings generated through the application of the framework highlight the varied composition of components of data visualisations, though certain combinations of components are prevalent, leading to static categorical comparisons or interactive spatial localization. After all, data artefacts can be understood as problem-posing elements that are the outcome of diagrammatic thinking that journalists employ to communicate claims.

The cultural heritage sector has traditionally been concerned with sharing resources and furthering human knowledge, with particular interest to the issues associated with metadata and interoperability, especially when it comes to the use of technology. These goals and interests in the cultural heritage sector are natural alignments with those of linked data; hence. there has been an increasing interest in the application of linked data in this sector. This article studies the implementation of linked data in the cultural heritage sector, through a systematic literature review of case studies of linked data implementation projects in this sector. The results reflect on the parties involved, the level of collaboration, and the motivation behind these projects. The study suggests that universities and national institutions were the main players in implementing such technologies in the cultural heritage sector, suggesting that there may be some barriers preventing smaller GLAM institutions from implementing linked data projects. The results further suggest that many linked data projects in this sector were primarily exploratory projects, and often performed in a collaborative manner. They further indicate that the most common motivating factors behind these projects were research needs, a desire to contribute to linked data as a movement, and other specific user needs. Reflecting on this systematic literature review, this article makes a set of recommendations for future work to increase the use of linked data in the cultural heritage sector and to remove barriers to adoption.

Reviewing the existing and long-established storytelling structures, this paper examines the use of the storytelling structures employed in data storytelling, specifically in the context of data journalism. For this, a large set of data stories from a variety of news outlets was collected, tagged and analysed. Accordingly, and reflecting on the results, the paper proposes a new storytelling structure for data storytelling, which addresses the unique requirements of this emerging area of study and practice, called the Water Tower structure. This proposed structure is an addition to the existing storytelling structures, and is specifically designed for and targeted at storytelling with data, with a particular focus on data journalism. While this paper is primarily focused on data storytelling in journalism, the contributions are believed to be of use and value to other domains such as Business.