Emotions, behaviours, and context-aware systems for journalism

New system ideas drawing on Affective Computing for journalism training include synthetic players with behaviour attributes, conceptualised as sub-systems and integrated with emotion engines, but they need to be designed from participatory data. System models could also look to advances in EmotionML (mark-up), but elicitation and articulation of emotions contextualised to professional interactions is the first step. Trigger documents from the Semantic CAT Method are used for these ends. Emotion-based vocabulary can also be integrated into logic models to consider machine reasoning for future journalism systems, but the conceptual models of Finite State Machines (FSMs) for Robo-Journo and Promo-Robo include health and emotion attributes assigned to transitions between states in professional processes. These automated systems are for verifying a story and treatment of press releases in the hands of journalists. Future context-aware systems can also include sensors and wearable systems, with potential to protect journalists reporting in conflict zones.

Drone journalism and aviation laws, systems, training, and tech trends

The professional use of civilian drones involves training in air safety and law set by aviation authorities, like CASA, which modified its laws, licences, and procedures in 2016. By 2019 media producers in countries like the UK, Australia, and the US, had used drones for almost a decade. Amidst the rules and deterrents, there are mixed benefits in using drones for news media. Prior to 2015 drone training in Australia began with PPL (Private Pilot’s Licence) theory, followed by an alternative pathway of a RePL (Remote Pilot’s Licence). The firsthand experiences of PPL training and subsequent training covers many aviation topics, from flight controllers to OzRunway apps. Beyond training, recent tech trends include networks for drones and swarm systems already used in the US and Korea. However, tracking and registration systems are only just emerging in Australia and drones, regarded as disruptive technologies in the UK, are complicated by Brexit.

The backend of news as a juxtaposition of data and human costs

Indexing and semantic code in news draw on a base of well-defined vocabulary from classification systems used by news editors for search tags, but journalism also uses leaked data, mobile metadata logs, and datasets for visualisations. The tagging systems in news, like NewsCode, are embedded in CMS and help to bind data for cross-referencing purposes. The defined concepts have an ontological base that relate to “news” and they are structured in hierarchical and logical ways. For many years social media tags were unstructured, but folksonomy approaches do not exclude semantic methods, and vice versa. Media cloud tools can also be used by journalists to generate lightly interactive graphic visualisations or to integrate data onto maps. However, data and metadata should also be used to develop new semantic systems to better protect journalists in conflict zones and to embed the values and ethics of journalism into algorithms for journalism training systems.

Formative digital and online technologies in news and journalism

Online news systems share some affordances of Turing’s universal machine, especially configurability, but the early generation of web standards enabled data sharing, interoperability, and ultimately frameworks to reasoning about digital resources. At the backend of online news, indexing, mark-up languages, and applied logic, provide a base for machine intelligence that ultimately extends to cloud servers and big data. However, XML languages, like RSS, enabled the first phase of sharing stories in the form of newsfeeds. Specific mark-up for online news, such a NewsML, also defined layout and other features of news sites. Tim Berners-Lee established the W3C for online standards in the 1980s, and then on the cusp of the 21st century he proposed semantic and structured approaches for meaningful data sharing online. However, in subsequent years entrepreneurs have appropriated semantic approaches for different ends. The atomisation of data also introduces “personalised” data preferences to pitch news stories.

Game spaces and the transformation of language for new journalism systems

The Semantic CAT Method, informed by participatory design, game design, and language attributes, is used for the elicitation and articulation of domain vocabulary and concepts in journalism and public relations, for the design of new semantic learning systems for journalism. Data in any new systems for journalism will require explicit labels extending to behaviour attributes. For these ends, drawings, creative lists, and game messages were created by journalists using Trigger documents integral to the CAT Method. Explicit domain concepts from doorstop interviews to online noise emerged from participatory drawing as well as game messages for meaningful language statements. The VerbIT technique, conceived by the author, was applied to statements from journalism and public relations, to turn language into imperatives for action, amongst other applied language features for online systems. UML diagrams, including Activity diagrams for logic pathways in professional tasks are also integral to the Semantic CAT Method.

Game design and the Semantic CAT Method for journalism

Future semantic learning systems for journalism should aim to integrate the values of the domain by using an ontological approach and a participatory design method like the Semantic CAT Method (Dowd). This method draws on game design and contextual approaches, as well as language structures. A focus on language and game design methods can work for semantic ends as well as modelling game-play. Ambiguity in design is also informative, but an ontology approach sorts language ambiguities, such as the same word with different connotations for journalism, social media, and public relations. It also helps to reveal domain characteristics that put journalism in a new light. The Trigger documents in the CAT Method include a focus on potential data and are scaffolded in participatory workshops. They include tasks for drawing and labelling typical scenarios in journalism, as well as UML diagrams for logic in processes, producing good results for an ontological base for journalism.

Decoding search and partnerships across news, journalism, and social media

Semantic news tags processed via cloud servers are in amongst big data and machine learning systems. The latter may have influenced Murdoch’s acquisition of a ‘social media news agency’, and other partnerships, as a mix of new roles across journalism, analytics, and search emerged. Some editing roles in journalism focus on SEO, but Murdoch’s Storyful, which started as a verification business created jobs for cloud operations engineers, viral video editors, and trends editors. Data-mining techniques were a lure for news and social media partnerships circa 2013–2016. In the name of verification, access to big data was matched by social media gaining credibility, evident in Facebook Newswire and other journalism projects. Deep learning methods in search, referrals, and automated tagging have also produced mutual benefits, mostly via third party agreements. However, data sharing for political ends by targeting particular users, and verification projects, have not stopped fake news.

Smartphones and social media in reporting the asylum seeker crisis in Europe

During the European asylum seeker crisis, circa 2015, asylum seekers used social media and smartphones for communication alongside journalists using disposable and short-form media, streaming media, and civilian drones for real-time stories that changed practices in journalism. Some journalists uploaded live video whilst others stitched together documentaries with short video clips. Google collaborated with the International Rescue Committee to develop an information site for essential services. A data visualisation developed by independent producers also showed the extent of the crisis. A former refugee, as well as media producers, used drones for aerial filming to help rescue people at sea, but as the crisis worsened, drones were used by authorities to stop people from crossing borders. The crisis also exposed journalists to trauma, even without working directly in conflict zones, revealed by the Dart Center for Journalism and Trauma, which plays an important role for the safety and protection of journalists.

The impacts of integrated media systems and social media in news

Integrated media systems are not only content management systems for production and publishing of online news, they are also hubs for mobile connectivity to remote servers and conduits for search and social media, but verification and analytics have also spawned data jobs intersecting with journalism. These layers of technological convergence across social media and media systems are like tunnels to data sharing on cloud servers. The latter has also presented opportunities for intermediary agencies, like Storyful, owned by Murdoch, to access big data and the potential of linked data via social media. The potential of cloud servers, mixed with social media, has also spawned new roles for news verification and roles in online trends, as well as cloud engineering jobs. Big data has indeed inspired new data sharing partnerships to boost online traffic and advertising through data insights. The result is more analytics that impact on the focus in journalism.

Language in data processing and ontology engineering for journalism

Trigonometry in algorithms with NLP (Natural Language Processing) can sort word connotations. The Triple structure of grammar in the RDF also extends to semantics in machine learning and big-data processing, but ontologies and a metamodel are essential for meaningful relations across data. They should inform the design of new journalism systems. Major processing platforms used by Facebook and Yahoo are distributed systems, like Hadoop, with resource negotiation features and computations applied to text. NLP used by Google also uses cosine vectors for connotations of words. Data processing already works across structured data for online news tags and unstructured data, like social media tags, with folksonomy characteristics, but social media also uses structured data. However, journalism is yet to come up with semantic systems, from an ontological base. For that end, ontologies across journalism, social media, and public relations and a little OWL to reason about resources can inform AI sub-systems and wider system perspectives.