2.1. Evolving Concepts

Sources are the backbone of journalistic activity. In this field, “trust” and “credibility” inevitably intersect - a topic often debated in the literature, whether these concepts align or stand apart. According to ^{Hovland and Weiss’s (1951}) classic approach, information garners credibility when sourced from a considered trustworthy origin. Other authors argue that trust in a source is established through repeated displays of credibility. ^{Van Dalen (2020}) admits that, while the two concepts should not be conflated, they can overlap. The author revisits the notion of “public-perceived credibility”, which can be assessed through the characteristics of the sender, the channel and the message itself. However, as highlighted by ^{Lecheler and Kruikemeier (2016}), theoretical studies on the credibility of digital sources are scarce. The impact on citizens’ trust in media due to the use of material from the digital sphere, often lacking scrutiny akin to that applied to offline sources, remains an open question.

There is no unanimous agreement on the terminology for social media. This article adopts the conceptual framework systematised by ^{Bechmann and Lomborg (2013}), which identifies a set of characteristics. Described as de-institutionalised communication, users also assume the role of producers. With the ability to transition between these roles, they actively engage by creating, filtering, and sharing content. This interactive process operates within a “network society” (^{Castells, 1996/1999}), eliminating any intermediation or hierarchy.

The influence of this type of platform is widespread. In the United Kingdom, for instance, Facebook has emerged as the third most popular news source, following BBC One and ITV (^{Jigsaw Research, 2022}). The escalating demand for social media as a news outlet is prompting media outlets to view them as sources. A Pew Research Center survey of American journalists substantiates this trend: 94% of respondents incorporate some form of social media into their work. Among the professionals who perceive its inclusion in their practices as positive, gathering reliable information is one of the lowest motivations (49%; ^{Gottfried et al., 2022}).

Journalists show interest in DSN material, particularly content created by citizens for publicising purposes, known as user-generated content (UGC). According to a literature review by ^{Naab and Sehl (2017}), there are numerous academic contributions exploring the concept of UGC and its integration into news stories. Research by ^{Grosser et al. (2019}) suggests that this integration has a slightly negative effect on the public’s perception of information reliability. The growing involvement of non-professionals in news production has prompted the development of normative instruments to guide journalists. As per a directive from the Portuguese state regulator regarding UGC edited and/ or incorporated by journalists into news space, it warns that “such content was not, ab initio, conceived in accordance with the rules and codes of conduct inherent to the exercise of journalism” (^{Conselho Regulador da Entidade Reguladora para a Comunicação Social, 2014}, p. 2).

The concept of “private territory” is constantly changing. The traditional notion that each individual preserves an inviolable space - intimate, confidential, shielded from others’ intrusion - is fading. The process, not entirely new, has gained momentum thanks to new ways of expressing sociability. In a digital environment, the boundaries between the public and private spheres are blurring (^{Loosen, 2011}; ^{Martins, 2019}; ^{Primo et al., 2015}; ^{von Pape et al., 2017}; ^{Whitehouse, 2010}). Rather than the erosion of privacy, we are witnessing the dissolution of the private into the public sphere, with DSNs as instruments, since they encourage what ^{Suler (2004}) describes as “online disinhibition” (p. 321).

DSNs create an illusion of privacy.

Even if someone restricts access to their Facebook profile, they can never be certain about who is actually reading their posts. The private nature of a message in a closed group can be compromised as it circulates among other people, detached from the original context in which the content was produced. (^{Primo et al., 2015}, p. 529)

In a landscape of public hyper-exposure, “self-disclosure seems to be a condition for being present on networks. Furthermore, the voluntary relinquishment of informational privacy serves as a means to authenticate an individual as a social actor” (^{Martins, 2019}, p. 15). Journalists tend to downplay the private nature of content, as evidenced by a survey on the attitudes of Brazilian professionals: “social networks are like open shelves where you can take what you want, a kind of public domain exempt of permission or authorisation” (^{Christofoletti, 2019}, p. 194).

Research exploring the relationship between journalism and DSNs has delved into issues such as interaction with the public, engagement with institutional and political sources, and the influence of networks as gatekeepers, agents qualified to select what can be turned into news. Two valid criticisms emerge: an overemphasis on the positive aspects of networks and insufficient attention to their negative implications for the journalism profession (^{Hamada, 2018}; ^{Lewis & Molyneux, 2018}). The literature review by ^{Lewis and Molyneux (2018}) underscores the scarcity of research focused on the specific issue addressed in this article: the use of DSNs by journalists as sources of information. However, the professional field has made its way. Reuters has adopted a general policy of refraining from using photographs from Facebook (^{Wardle et al., 2014}). For the BBC, journalists should abstain from using content that is clearly not intended for widespread dissemination, and the use of content from networks featuring children requires an assessment of the impact of dissemination.

Several studies have demonstrated that journalists acknowledge the utility of digital media for their work (e.g., ^{Gulyás, 2016}; ^{McGregor & Molyneux, 2020}; ^{O’Sullivan & Heinonen, 2008}; ^{Weaver & Willnat, 2016}). However, a critical attitude persists, materialising in daily operations aimed at verifying the authenticity of DSN accounts or determining whether they are managed by the person or someone else. Editorial managers are often unaware of the complexity of the ecosystem, and newsrooms are not adequately trained to navigate these platforms. In an investigation examining the use of photographs and videos produced by non-professionals or individuals unrelated to the news organisation - analysing content from eight international television stations (on air and online) and conducting 64 interviews with professionals from 38 media companies - one journalist highlighted the challenge of building a relationship with someone through Twitter. “You see people just hammering [uploaders]. ‘Call me. Call me. Here’s my number’” (^{Wardle et al., 2014}, p. 109).

Digital media inherently influence practices not only in terms of distribution but also in the production of information. As highlighted by ^{Colombo et al. (2023}), among others, changes are constant, requiring continuous adaptation. ^{Brandtzaeg et al. (2016}) learned that DSNs are frequently used as a primary source of news. Verification strategies encompass five main categories: “(1) Trusted sources. (2) Access to eyewitnesses and authenticating sources. (3) Traditional journalistic methods. (4) Multimodal verification and verification tools. (5) Workaround methods” (Brandtzaeg et al., 2016, p. 328).

The existing organisational contexts within the media do not foster adherence to basic rules. The downsizing of newsroom staff makes all members generalists and requires functional versatility. Permanent news updates, characteristic of what is described as the “24/7 online era”, significantly affect journalistic work. A study on Dutch journalists’ practices in handling digital sources (^{Lecheler et al., 2019}) demonstrates a tendency to prioritise the quantity of information at the expense of quality.

Immediacy has become so deeply ingrained in professional culture that it has evolved into the primary news value. A new paradigm has surfaced: “in the digital era, users are looking for real-time publication, and publishers feel pressure to get out information now” (^{Elliott & Spence, 2017}, p. 21). However, the pursuit of being the first and the fastest detracts from the time available for assessing, interpreting, and confirming information. The fixation on exclusivity, exacerbated by market logic, negatively impacts online coverage, giving rise to sensationalist and even alarmist coverage.

2.2. Digital Social Networks in News Dynamics

Stories on the platforms have the potential to be newsworthy. However, tweets are often considered in editorial decisions primarily based on topicality, sidelining essential news values such as credibility, objectivity and context (^{McGregor & Molyneux, 2020}). When the potential for virality is also factored in to align with algorithmic criteria, it opens the door to prioritising the interest of the public at the expense of the public interest.

The mediatisation of private matters may be legitimate, but it requires additional caution in order to reconcile the rights - to information and privacy - consecrated in democratic societies. The risk of journalists being exposed to manipulation or misinformation must also be considered. Especially in the tabloids, tweets seem to be taken at face value. There are no signs that the source or other sources were contacted to verify information that was twittered”, ^{Broersma and Graham (2013}, p. 461) note. ^{Vinuesa and Nicolás-Sans (2023}) remark: “social media is the fastest way to spread fake news or any other element of post-truth” (p. 3).

Journalists use DSNs for information gathering rather than validation or interviewing sources, although they also resort to them for breaking news (^{Weaver & Willnat, 2016}). Additional research - such as ^{Belair-Gagnon’s (2015}) examination of practices at the BBC - indicates that coverage of unscheduled events is more likely to involve the use of DSNs as sources. Twitter, in particular, proves to be a valuable tool and, in certain instances, the sole available to cover disasters, crimes, or political unrest (^{Moon & Hadley, 2014}).

Given the unfolding events, journalists’ social responsibility is being more visibly tested. It proves effective for them to concentrate on the information available in DSNs rather than solely identifying the person who posted it, considering that the individual may be in a state of shock or distress. In such situations, the testimonies of victims and those directly affected take precedence. Contrary to previous research, ^{Mayo-Cubero (2020}) learned that, under these circumstances, editorial managers tend to equate the credibility of official and unofficial sources. However, the uncritical or reckless collection of content has damaging effects. Making characterisation errors - converting an accident into a terrorist act, for example - is likely to cause panic. It increases the likelihood of disseminating content that is false; authentic content but only showcases one perspective or part of the incident; or pertains to an entirely different event. A video from YouTube showing footage of combat cars being shot down, presented by RTP in 2022 in a report on the use of molotov cocktails by ordinary Ukrainians against Russian troops, actually represented the Maidan Square protests in 2014, according to the Regulatory Council of the Portuguese Regulatory Authority for the Media (^{Conselho Regulador da Entidade Reguladora para a Comunicação Social, 2022})¹.

The immediacy of posts fosters the extraction of all information, even when there is no way of assessing its integrity. Lacking the time for critical scrutiny, journalists risk overlooking the fact that they are only accessing fragments of reality and making a comprehensible and coherent narrative impossible. An American study concluded that journalist do not consider incremental updates to be significant, “but they do them anyway” (^{Usher, 2018}, p. 28). This behaviour is motivated by factors such as the “obsessive quest for traffic” and the fear of becoming irrelevant to the public - essentially, the fear of losing professional authority. However, “the changes journalists are making may ultimately destroy what little claim to authority they may have left” (p. 32).

The 2004 tsunami in Southwest Asia marked the first systematic effort to seek out UGC to access eyewitness testimonies, as reported by a participant in the UGC Hub, a project initiated by the BBC to monitor this phenomenon (^{Wardle et al., 2014}). During the 2005 terrorist attacks in London, the BBC incorporated images not collected by its professionals for the first time. This marked a challenge to the traditional concept of journalism. In recent years, there has been a profound shift in dynamics in monitoring ongoing events. The final phase in the coverage of the 2011 terrorist attack in Utøya, Norway, was characterised by “a hybrid authority ( ... ). Rather than primarily framing social media as a ‘source’, journalists sought to use it as a means of involving the audience in the actual news production” (^{Konow-Lund & Olsson, 2017}, p. 1201).

Comparing the media coverage of the American school massacres at Columbine (1999) and Sandy Hook (2012), ^{Joffe (2004}) states that, in the latter case, “what began as unverified social media posts carried over into daily newspapers, perpetuating rumour and erroneous information” (p. 53). A university professor of Communication noted, especially in relation to the “Boston Marathon” bomb attack in 2013, that “Twitter, especially, has put a lot more pressure on news organisations to get it out fast” (^{Shih, 2013}, para. 12).

The misidentification of suspects is a recurring issue in cases such as Sandy Hook and the Boston Marathon. In the latter, mainstream media accepted as verified the name of a student unrelated to the crime, as pointed out on Reddit, leading to the student’s family receiving threatening phone calls. The reproduction of images not only impacts the personality rights of those involved, as seen in cases like the broadcast of a video from DSN depicting an alleged sexual assault on a bus in Portugal (^{Martins, 2019}) or the use of an actor’s photograph as that of a dead soldier by a New Zealand newspaper (^{Smith & Sissons, 2019}). There is also the possibility that journalists, even unintentionally, may contribute to terrorists’ agendas or provide useful information to perpetrators - for instance, revealing the position of police officers during an attack - should not be ruled out as a potential consequence (^{Rauchfleisch et al., 2017}).

Guidelines on the use of digital media, issued by regulatory bodies or media organisations - ^{Associated Press (2013}), BBC (^{Hulin & Stones, 2013}), Los Angeles Times (^{Standards and Practices Committee, 2009}), Österreichischer Rundfunk (^{Christophg et al., 2012}), The Washington Post (^{Washington Post’s Social Media Policy, 2009}), among others - manifest duties that are already encompassed in general codes and recommend procedures contingent on the context. These guidelines serve as measures to address risks, as identified in the Code of Ethics of the News Broadcasters & Digital Association (United States), which is part of the corpus of this research: “‘trending’, ‘going viral’ or ‘exploding on social media’ may increase urgency, but these phenomena only heighten the need for strict standards of accuracy (para. 10).

The safeguarding of privacy is an overarching concern. The ^{Independent Press Standards Organisation (2022}) notes that “journalists must not assume that the absence of privacy settings means that information can be published” (p. 9). The impact of reusing images of the deceased is mentioned by this organisation and echoed by the BBC, which also emphasises the importance of ascertaining the original intention of the authors of the posts. The ^{Associated Press (2013}) outlines measures focused on verifying the authenticity of the source and mitigating the risk of endangering people. The Source Media Group, according to the American Society of News Editors, mentions approving the use of material by the senior management (^{Hohmann, J. & 2010-11 ASNE Ethics and Values Committee, 2011}).

3.1. Reporting and Discussion of Findings

By enumerating duties, ethical codes embody journalists’ social responsibility and their allegiance primarily to the public. This allegiance is evident in their commitment to principles such as the protection of privacy, among the most important guiding principles of professional practice.

Authors such as ^{Hamada (2018}) and ^{Ward (2018}) have raised questions about the feasibility of establishing universal values that could guide journalists across different countries or platforms. While ethical foundations are expected to remain consistent regardless of the communication channel, proposals like ^{Whitehouse’s (2010}) advocacy for updating codes gain relevance. This is especially pertinent due to intrusion into private life facilitated by the current media ecosystem, particularly with the growing influence of DSNs. The content of the corpus studied here suggests that such a strategy of updating codes is being pursued.

This research originated from an analysis of deontological instruments - specifically self-regulation rather than legal frameworks - that explicitly reference DSNs as journalistic sources. The aim is to understand how the use of content from social networks is outlined within codes, manuals, guides or other types of documents that provide ethical guidelines for journalists, an aspect included in some studies (^{Ballesteros-Aguayo et al., 2022}; ^{Fidalgo et al., 2022}), although at a secondary level.

The objective is materialised by identifying the presence of values and principles specifically applicable to the collection of information (text or images) on this type of digital platform, particularly those sourced from personal accounts. For the qualitative analysis, a coding table was created encompassing eight categories reflecting values commonly regarded as essential in professional practice: “rigour”; “verification/validation of information”; “identification of the source/origin of information”; “protection of privacy”; “public interest”; “vulnerability/pain”; “news treatment of victims”; and “handling of cases involving underage citizens”. Additionally, “generic references” is included as a category, encompassing elements not covered in the preceding ones.

The corpus includes 32 deontological instruments regulating journalistic activity (Table 1), characterised by their broad scope. These instruments were selected from a database created by the author, which includes approximately 250 codes. The selection was based on the following cumulative criteria: (a) inclusion of recommendations on information collection in DSN or UGC, (b) current validity, and (c) availability online. The codes are sourced from various entities (self-regulatory bodies such as press councils, trade union organisations, employers’ associations, specific bodies and business groups, state or private). The covered codes are national - sometimes international, such as the Global Charter of Ethics for Journalists of the International Federation of Journalists - or exclusive to a particular media outlet.

Several codes advocate the adoption of procedures typical of offline operations in a digital environment. Some codes allocate specific sections for using the internet or, more specifically, social media. Notably, the Canadian Association of Journalists and the BBC have dedicated instruments directly addressing professional practices on the web, which serve as complements to their general guidelines.

The Guia de Ética e Autorregulamentação Jornalística of the Brazilian group RBS emphasises that “in digital culture, everyone needs to continually learn, update, and share” (p. 35). The warning is all the more relevant given the frequent changes in this field - especially in technology. The code of the Greek Online Publishers Association of Greece underscores that “tweet or a Facebook post can be visible to ten, hundred, or thousand people” (p. 8), acknowledging the diffuse characterisation of the recipients, as academic research also suggests: “it’s more appropriate to think of Twitter as a public, rather than the public” (^{Lewis & Molyneux, 2018}, p. 17).

The Independent’s Code of Ethics highlights that the internet “can present an ethical vacuum, and you should not assume that simply because something appears online, it can be published by us” (para. 120). The NRC (Netherlands) approaches the issue from a similar perspective, stating that the popularity of a subject on the internet can be a consideration but not necessarily an argument for publishing a news item.

In this context, transparency in the relationship with sources and consumers of information becomes even more crucial. A subject of general interest broadcast in a digital environment must be clearly attributed “so that our audiences can assess the information and its source” (TG4, p. 47). In its Ethics Handbook, American National Public Radio pledges to disclose both confirmed and unconfirmed information and to challenge those who publish information on social media to provide proof. Explaining to witnesses of events who produce UGC where and how the content will be included is part of the work of Agence France Press, whose Charte AFP des Bonnes Pratiques Éditoriales et Déontologiques embraces the principles established by the Eyewitness Media Hub. This organisation investigates the media’s broadcasting of images captured by witnesses of events.

While the risk associated with spreading rumours through the internet is explicitly mentioned in only six codes - Agence France Press highlights that the provision of accurate and verified information “becomes more important than ever as ‘noise’ and rumours grow on the Internet” (p. 5) - this does not mean that the others neglect indispensable procedures. Both technical and deontological procedures are highlighted across most deontological instruments in the corpus. These measures aim to prevent journalists from being conduits for disinformation and to ensure the production of an accurate narrative. In fact, around two-thirds of the deontological instruments that make up the corpus fall into the category “verification/validation of information”, and half relate to the category “identification of the source/origin of the information” (Table 2). Collectively, these measures aim to achieve the desirable “rigour”, even if the term is not explicitly used in the codes.

The concern for privacy, which is explicitly mentioned in nine documents, is given greater prominence than the public interest, which, despite its centrality in professional culture, is only noted in the Code de Déontologie du Raad voor de Journalistiek of the Press Council of Belgium, in a directive on the use of information and images from DSNs and private websites; TG4’s Journalism and Contentious Content Guidelines; and the Editors’ Code of Practice (Independent Press Standards Organisation) and Code of Conduct (The Independent), both from the United Kingdom. Other sensitive topics - the treatment of victims and minors and situations involving vulnerability or pain - also intersect with dimensions of personal life. When combined with the “privacy” category, references to these issues are found in 20 codes.

The Guide de Déontologie des Journalistes du Québec (Canada) notes that “the speed of these networks should not blind us to the need to ensure that the facts are true and to put them into context” (p. 10). Accuracy and context are overarching concerns in the codes. It involves safeguarding against exposure to propaganda or hidden agendas circulating in digital media while also upholding values such as privacy. Journalists are thus required to assess the potential harm to individuals’ reputations and image, particularly when dealing with people in vulnerable situations - such as minors, victims of crime, disasters and accidents, as cited by the Belgian Press Council. This ethical provision limits the use of personal information and data on public websites, acknowledging “the differences in context, scope and impact of the dissemination of journalistic information” (p. 10).

Concerning the conduct of individuals sharing personal information online, the Independent Press Standards Organisation incorporates digital communications into the privacy section, emphasising the need for appropriate protection. In line with the British model, it mandates editorial executives to justify invasions in this field. The assessment of legitimate expectations of privacy includes considerations of self-disclosures and the fact that the material is already in the public domain or will be. This approach, used in the United Kingdom to assess media conduct, prompts journalists to ponder essential questions as outlined in The Independent’s code: what is the nature of the material (is it intrinsically private)? Who uploaded it and why? Did they intend that it be widely published? How widely has it been/can it be seen? Has that person waived their right to privacy in whole or as regards this particular aspect? Is there sufficient public interest to justify a possible invasion of privacy? The scrutiny underlying these questions is irrespective of the status of those involved. Nevertheless, the Global Charter of Ethics for Journalists, endorsed in 2019 by the International Federation of Journalists, and the Swedish Media Ombudsman code both refer to “non-public persons” who expose themselves on social media. That underscores an awareness of the potential involvement of anonymous citizens in the news process, whose repercussions are not always properly understood by them.

A more detailed examination of the array of professional practices outlined in the codes shows that the recommendation to exercise particular caution in scrutinising the information obtained is a recurrent theme in the six codes that refer to the UGC. The imperative to identify the author is explicit in both the German Press Council and the BBC Editorial Standards. The British public broadcaster permits the use of anonymous posts on social media for information purposes but prohibits quoting and demands clarity about what is or is not confirmed.

The use of content extracted from unauthenticated accounts and profiles on digital platforms is explicitly forbidden, according to Agência Lusa. Expresso, in the version updated in 2019, mandates a meticulous accuracy analysis, while the BBC demands the verification of platform manager identities along with confirmation of material authenticity. However, merely citing the source may not suffice to ensure the reliability of the information. “how do you prove that the person who apparently tweeted, commented or uploaded a photograph actually did so themselves?” (p. 87), The Independent asks. The fear of manipulation prompts the Group New Zealand Media and Entertainment to underscore in its code the fundamental principle of multiplying sources, for “in this era of text messages, social media, and emails, our journalists are required to make sure their sources are legitimate and not an ‘imposter’” (p. 5).

The fact that websites, even reliable ones, have not been updated for years raises the need, as TG4 notes, to resort to trusted sources who are qualified to corroborate information. Direct contact with sources circulating on the internet is advised in six codes to verify their identity, confirm the authenticity of the material, or gain a better understanding of the meaning of the material concerned. These steps become particularly crucial when a post on a DSN is considered news in itself or when seeking consent to extract material, especially photographs.

In the continuous coverage of events, reports on unfolding situations vary constantly. “Speed and accuracy are fierce rivals” National Public Radio (United States) summarises in the section of its manual dedicated to “Accuracy” in a digital context. In a news agency, this “rivalry” can have disastrous consequences. Hence, the charter of Agence France Press (France) intensifies the verification rules to be applied to content disseminated by DSNs - including the authenticity and authorship of videos, the date, and the place where they were produced. The key moment or the triggering of an event can be captured in images by witnesses, who share them online. The agency highlights that “these images often play an essential role in our account of the event, and sometimes they are the event itself” (p. 22). Such content cannot fail to be scrutinised, not least to understand the motivations of those who disseminate it. According to the BBC, information provided through social media by alleged victims must be carefully considered because incorporating it into reports could put them in greater danger.

Listed in various codes, securing consent for using social media material (i.e., DSN) is rarely feasible during the monitoring of breaking news, a situation in which the speed of broadcast is of the utmost importance. This constraint does not rule out - but rather reinforces - the need for verification as thoroughly as possible. In the case of content that is not considered open access - that is, protected by privacy settings - the ethical considerations take on new contours. According to the Belgian Press Council, there is no doubt that only the demonstration of significant public interest or the consent of the individuals themselves can override the rule, inhibiting the use of information or images from private platforms.

Sharing personal information, possibly recklessly, reduces the expectation of privacy of those who have done so, making it permissible to extract content. According to the BBC, this is applicable when individuals demonstrate an understanding of the impact of their posts or adopt privacy controls. However, it may not apply to other citizens, especially children whose image is publicised. When individuals under the age of 18 submit UGCs online, the BBC does not categorically exclude them. However, it imposes the condition of obtaining adequate parental consent, considering the subject matter and the age of the individual. The charter from Agence France Press recommends exercising particular caution concerning content such as film footage and photographs featuring children. Additionally, images taken by minors using smartphones should be subject to legal advice before dissemination. Age, therefore, emerges as an unavoidable factor, as minors may not fully comprehend the public nature of their posts (Ethics Guidelines of the Canadian Association of Journalists).

Given the risk of spreading rumours on DSNs, some codes explicitly address news about people’s deaths - echoing the sentiments expressed in the guidelines of France Media Monde and Agence France Press. Mistakenly identifying a deceased person is so detrimental to a media outlet’s reputation that it cannot forego absolute confirmation obtained either from the family or another authorised source or with direct knowledge of the situation. In a digital environment, where information is often disseminated rapidly, obtaining such confirmation becomes more challenging. However, that is not the sole dimension to consider: TG4 emphasises the negative impact on family and friends when using images from DSNs to illustrate news of someone’s death should be assessed. Journalists should exercise a similar level of caution before linking to an online video related to the deceased.