In the Name of the Right to be Forgotten: New Legal and Policy Issues and Practices regarding Unpublishing Requests in Slovenian Online News Media

Abstract The goal of this study is to explore the Right to Be Forgotten (RTBF) as a specific legal aspect of the EU General Data Protection Regulation (GDPR) within the context of digital journalism and related media policy issues. We address this issue in cases where requests have been made to unpublish news items or other (visual) content from online media archives because they contain embarrassing, irrelevant and/or outdated, yet truthful content. To do this, we researched the editorial policies of five Slovenian online news media outlets in their responses to such unpublishing requests. First, we reviewed regulation and key legal decisions and then used in-depth semi-structured interviews with editors of these outlets. Our research showed that unpublishing requests from 2018 and 2019 cite or imply the RTBF as having an EU-wide legal basis, yet the media outlets analyzed have not established clear internal policies. This opens the door to inconsistent and/or arbitrary decisions. The legal foundations for unpublishing online news items are vague and, at least in Slovenia, subject to opposing interpretations which might lead to new restrictions on media freedom. To avoid additional potential for the manipulation of media, both legal and self-regulatory frameworks need to be updated and clarified.


Introduction
The issue of censorship and freedom of expression in the digital ecosystem has been evolving within the wider framework of "a conceptual reevaluation of a new communication technology" (Bollinger 1990, 103). This is how Bollinger defined the need to re-think the key aspects of media paradigm(s) when faced with disruptive technologies. The development of new policy framework for digital journalism calls for such reevaluation, as the new digital information and communication technologies (ICT) on which digital journalism is based present a number of challenges to freedom of When deciding whether to grant unpublishing requests, in principle news editors are faced with an ethical dilemma arising from the clash of free expression, historical integrity and accountability on one side, and harm reduction, privacy and redemption on the other (Shapiro andRogers 2017, 1101). Here, online media have to make important decisions about how they should respond to such requests while upholding journalistic principles and best practices (Acharya 2015, 89), as removing truthful information from digital news archives involves a "conflict between the traditional ethical values of reporting the truth while at the same time not causing harm" (McNealy and Alexander 2018, 401). Such unpublishing requests pose additional problems for the editors of digital news media in the EU since they must also deal with legal requests referring to the GDPR and RTBF. This means these requests can constitute both a threat to media freedom and a media policy issue.
The goal of this study is to explore the Right to Be Forgotten (RTBF) as a specific legal aspect of the EU General Data Protection Regulation (GDPR) within the context of digital journalism, freedom of expression and related media policy issues. The GDPR Directive is one of the EU's key new media and ICT directives for digital news ecosystems in the context of the evolving freedom of expression in digital media. We will explore the application of this based on analysis of editorial responses and potential policies in cases of RTBF requests by individuals to unpublish truthful information in Slovenia, an EU member state. The current editorial practices at five Slovenian online news media outlets will be reviewed, and their editorial policies thoroughly analyzed. In this way, we will explore fully the role of the RTBF tool in digital journalism practice in Slovenia as one of many countries facing this issue within the EU.

The RTBF and the News Media -Precedent Cases and Industry Practices
The RTBF was established by the CJEU in the 2014 case of Google Spain SL v. Agencia Española de Protecci on de Datos (Google v. Spain judgment) 1 (for more, see Alessi 2017, 145-146), as the right of individuals to request search engine providers to remove links to personal information about them. According to the CJEU ruling, delinking of search results can be granted when the data "appear to be inadequate, irrelevant or no longer relevant, or excessive in relation to those purposes and in the light of the time that has elapsed". The introduction of the GDPR, applicable throughout the EU from May 2018, was the first time European legislation recognized the existence of the RTBF, which had until then been rooted in case law (Di Ciommo 2017, 623-624).
The complainant in the Google v. Spain case also requested the newspaper withdraw published news, yet this part of his request was not granted. According to the CJEU judgment, "the processing of personal data carried out in the context of the activity of a search engine can be distinguished from and is additional to that carried out by publishers of websites". Nevertheless, the Google v. Spain judgment led to "wide-reaching implications for freedom of expression on the Internet" (Youm and Park 2016, 284) by empowering individuals and states to censor content (Oghia 2018). As a consequence, news organizations have received more and more requests from individuals asking that their names be removed from news stories (Sant ın 2017, 305), and some have gradually become increasingly willing to grant unpublishing requests (Shapiro andRogers 2017, 1101), even in cases of truthful reporting.
Since then, there have been signs of the RTBF being applied, or attempts to apply it, directly to news websites. In 2015, the Supreme Court of Cassation (the highest court in Italy) upheld a ruling in the case of PrimaDaNoi that after a period of two years an article in an online news archive had expired. The Court decided that, although the article was published lawfully and was in the public interest, allowing access to it disproportionately impacted the individual's privacy. "The Court accepted the importance of the right to report, but held that sensitive and private information should not be available to the public indefinitely (unless the publisher receives consent to do so from the concerned person). The Court upheld the EUR 10,000 in damages against PrimaDaNoi" (Columbia Global Freedom of Expression 2019) 2 .
In Belgium, the Court of Cassation found in 2016 that Le Soir had been properly ordered to anonymize an article containing information about the applicant, who had been convicted of a drink-driving offense that led to a fatal road traffic accident. Since the conviction had been spent the court argued that the continued publication of this offense twenty years later was likely to cause him disproportionate damage, outweighing the strict respect for freedom of expression (Agate 2018;Tomlinson 2016). Although the article was not removed from the media website but merely anonymized, the Court of Appeal held that the "right to be forgotten" was an intrinsic part of the right to respect for private life (Tomlinson 2016). The case confirmed the willingness of certain courts and legal systems in the EU to attribute the right to respect for private life greater importance than the right to freedom of expression, including cases in which the media report truthful information.
Both cases led to criticism in affected countries 3 and abroad 4 , including that the RTBF had become "the right to remove inconvenient journalism from archives after two years" (Matthews 2016). According to The Independent's executive editor Will Gore (2018), "it is important to note the right to be forgotten applies specifically to search engines, not to individual publishers". Opponents of unpublishing often argue that a generalized RTBF "would lead to the rewriting of history in ways that impoverish our insights" (De Baets 2016, 64).
The Google v. Spain judgment has led to several research articles and analyses, and prompted diverse reactions and discussions (for an overview, see Villaronga, Kieseberg, and Li 2018, 307) mostly related to removing links on search engines. Conversely, previous studies have focused mainly on theoretical foundations, legal frameworks and controversies regarding the RTBF and search engines (e.g. Andryushchenko 2016; Alessi 2017; Kelly and Satola 2017;Post 2018;Villaronga, Kieseberg, and Li 2018); the application of the RTBF in digital journalism and its understanding in news media policy have received less attention.
Research on unpublishing requests includes English's survey of editors from North American news organizations (2009), which found "little news industry consensus on how to handle and respond to public requests to unpublish news content from online news sources" (English 2009, 4). The Canadian Association of Journalists (English, Currie, and Link 2010) later recommended ten best practices for handling requests to unpublish digital content, but internal media guidelines have been rare. McNealy and Alexander (2018) provided a theoretical framework for news organizations to make unpublishing decisions, weighing the sensitivity of information against its news value. To reconcile conflicting principles in this dilemma, Shapiro and Rogers (2017) suggested distinguishing between truthfulness and relevance, and between the availability of information and the ease of its searchability. English's study (2009;2010) identified various reasons for unpublishing requests. However, this research was not conducted within the EU's legal framework, policies and practices, and as such its results do not relate to the RTBF as a specific European concept and issue.

Unpublishing Requests from the Standpoint of Slovenian and EU Legal Frameworks and Policies
Individuals who request that Slovenian media outlets unpublish news items from their online archives often cite the Google v. Spain judgment (Zakonj sek 2019, 35) to support their claims, even though it is not relevant to the subject in question, as it applies only to the obligations of search engines. In Slovenia, three laws can be applied to the issue of unpublishing news items from media online archives: the GDPR as an EUwide directive; the Personal Data Protection Act (ZVOP-1) as the key Slovenian legislation on data and privacy; and the OZ-UPB1the Obligations Code.
The GDPR in Article 17(1) specifies that the data subject has the right to obtain from the controller the erasure of personal data concerning him or herself without undue delay, and the controller is obliged to erase personal data without undue delay where one of the grounds applies, including, inter alia, in the event that "the personal data are no longer necessary in relation to the purposes for which they were collected or otherwise processed" (Article 17 [1][a] of the GDPR). Article 17(3)(a) provides that erasure within the context of paragraphs 1 and two of the same article shall not apply to the extent that processing is necessary for "exercising the right of freedom of expression and information." The Slovenian Information Commissioner (2019) stated that the GDPR provision could also be applied to requests to erase journalistic items from online archives. The Commissioner took the view that, on the basis of Article 17(1)(a) of the GDPR, a person who participated in an interview nine years ago could request its removal from the company owning the website on which it was published if the information it contained ceased to be relevant or essential due to time elapsed. At the same time, the Commissioner warned that the data controller can reject such a request if, for example, it can prove that by posting the content it is exercising its right to freedom of expression and information: "[t]he media enjoy a special status by law because they act in the public interest, and informing the public is part of their freedom of expression." ZVOP-1 contains a vague formulation in Article 32 (1): "upon the request of the data subject, the personal data controller shall supplement, correct, block, or erase personal data which the individual proves to be incomplete, inaccurate or not up to date". Yet, we argue that this provision does not apply to cases in which an applicant requests that complete, accurate and (at the time of publishing) up-to-date data be unpublished, even if these data are no longer found to be relevant over time.
Applicants requesting that news items be unpublished from online media archives sometimes refer to the OZ-UPB1. Under Article 134, all persons "shall have the right to request the court or any other relevant authority to order that action that infringes the inviolability of the human person, personal and family life or any other personal right be ceased, that such action be prevented or that the consequences of such action be eliminated". Namely, keeping a news item in an online media archive could be understood as an infringement of personal rights, lasting as long as a news item is accessible to readers. Therefore, the injured party may request that a media outlet cease the infringement of personal rights; that is, to unpublish from its archive a news item that infringes upon personal rights. However, such a request would only be justified when the encroachment on personal rights is unlawful, i.e. when a media outlet has abused its freedom of expression to encroach inadmissibly upon personal rights.
We may conclude that, on the basis of Article 17(1)(a) of the GDPR, interested parties could in principle require that a media outlet unpublish journalistic items from its online archives if these items encroach upon their personal rights, and if they are no longer current. Due to the protection of the right to freedom of expression and the public's right to information (Article 17(3)(a), such requests can be justified only exceptionally. It is not specified in the GDPR when such exceptions occur. These criteria, however, were established by the European Court of Human Rights (ECtHR) in its (scant) case law on the subject.
To date, the ECtHR has adjudicated on the right to unpublish journalistic items from online archives in Wegrzynowaki and Smolczewski v. Poland 5 and ML and WL v. Germany 6 .
In Wegrzynowski and Smolczewski v. Poland, the ECtHR decided on an application by two Polish lawyers who won their action for compensation, and an apology from the media for untruthful statements in a journalistic article before the Polish court, but lost the subsequent action demanding the erasure of published articles from the newspaper's online archive. The ECtHR upheld the findings of the Polish courts and dismissed the application. It is clear from the reasoning of the ECtHR judgment that online media archives are protected within the context of freedom of expression. According to the ECtHR, it is not the task of the judicial authorities to rewrite history by removing published articles from online archives.
In M.L. and W.W. v. Germany, the ECtHR adjudicated on a case in which the applicants were convicted in 1993 of the murder of a famous actor. In 2007, they filed an action against several media outlets with a request for the anonymization of personal data. The court in the first instance and the court of appeal both upheld their claims, on the grounds that their interest in not being confronted with their past convictions prevailed over the public's interest in being informed of the applicants' criminal offences. However, the federal court reversed this decision on the grounds that the previous courts did not adequately protect the right to freedom of expression.
The applicants lodged an appeal with the ECtHR against these judgments, which the ECtHR subsequently rejected. It concurred with the findings of the German federal court, stating that the media is tasked with participating in the creation of a democratic opinion by providing the public with old news items stored in their archives. A decision to ban the publication of information about individuals could have a detrimental effect on the media's freedom of expression. Regarding the request for the renewal of criminal proceedings, both applicants contacted the media themselves, and gave them information for publication. This is another reason the court did not find their application admissible.
In the judgments described above, the ECtHR did not deviate from its established position that the protection of the right to freedom of expression is an essential human right that can only be limited in exceptional cases (Harris et al. 2009, 443). The introduction of digital technology in the way the media functions, which brings with it the possibility of storing journalistic items that are accessible to the public for an extended period, is thus clearly not, according to the ECtHR, affected by the relationship between freedom of expression and colliding rights. 7 In this context, the ECtHR draws attention to the chilling effect that judgments requiring the deletion of articles from online archives has on the media, and to the risk of rewriting history which might arise if courts were to order media outlets to unpublish news items from their online archives. According to the ECtHR, as is clear from Wegrzynowaki and Smolczewski v. Poland, the media can also keep news items in their online archives that constitute an abuse of press freedom: for example, items containing untruthful references to individuals, and thus constituting an unlawful interference with their human rights. Based on this judgment, we may assume that the ECtHR would find an application to unpublish a news item containing truthful data even more unfounded, and thus inadmissible.
As is clear from the judgment in M.L. and W.W. v. Germany, the assessment of the merits of the claim for erasure must also take into account whether the matter is in the public interest, and whether it is still current after time elapsed. It is also important to ascertain whether or not the persons who request the erasure have exposed themselves to the public. In M.L. and W.W. v. Germany, the applicants became known to the public by their own initiative after attempting to renew criminal proceedings; therefore, they themselves had turned to the media, a fact that speaks against their request for the erasure of controversial journalistic items. Additionally, the content and form of controversial news items are relevant. If the facts were objectively presented, and if identification of the person requesting the erasure was difficult due to the passage of time, the media would unlikely be obliged to unpublish the contribution from their respective online archives.
In this respect, in the cases of Wegrzynowaki and Smolczewski v. Poland and M.L. and W.W. v. Germany, the ECtHR only complemented its criteria for defining the limits of freedom of expression arising from its settled case law. In line with these criteria, the limits of freedom of expression must primarily be set considering two criteria: whether the journalist was reporting on a topic in the public interest; 8  However, the potential threat to freedom of expression when the RTBF is applied to news media was confirmed in a Slovenian case in May 2019, when an applicant sued the newspaper Dnevnik for not removing a (truthful) story about him. The applicant claimed he was no longer a person in the public interest, and thus the RTBF should be applied. The District Court in Ljubljana ordered Dnevnik to remove the article from its website and place it on a specially produced website where only users with specific approval and passwords have access to it 10 . The defendant has filed an appeal, and the case is pending. However, the District Court's decision confirms the potential threat of the RTBF to news media by demonstrating the confusion around its interpretation and implementation. This confusion is enhanced by statements made by influential Slovenian lawyers. A former Information Commissioner and now private-sector lawyer says: "It is significant that Slovenian courts in analogy to the decision in the case of Google v. Spain also recognized the search engines of particular online media as those that enable the production of personal data collections. This of course means that the GDPR also applies to the media/ … /particularly from the perspective of the RTBF" (Ropac 2019, 8).

Methodology
In 2018 and 2019, online news media in Sloveniaa member of the EU and thus part of the EU legal frameworkreceived a growing number of requests for a certain news item or piece of content (such as a photograph, or other visual element) to be removed from a media website or archive. This demonstrates a shift in how the application of RTBF affects digital journalism, and thus represents a significant research issue. In 2018, the Journalistic Honorary Arbitration Court adopted recommendations regarding unpublishing news articles in online media (N CR 2018). These stated that as a rule media outlets should not remove news items, but are permitted to do so in exceptional circumstances. In such cases, the reasons for removal should be explained to the public. To this end, media outlets should prepare, and make publically available, clear unpublishing guidelines.
To research how Slovenian online news media outlets respond to unpublishing requests implying the RTBF, we used the methodological approach of in-depth semistructured interviews. This qualitative method was considered appropriate because it is "sufficiently structured to address specific topics related to the phenomenon of study, while leaving space for participants to offer new meanings to the study focus" (Galletta 2013, 24). We adopted the following definition of an unpublishing request implying an individual's right to be forgotten: An unpublishing request is a request to unpublish a news item or other (visual) content from an online media archive because of embarrassing, irrelevant and/or outdated (yet truthful) content about an individual, regardless of whether the applicant explicitly refers to the right to be forgotten. Requests referring to allegedly false, incomplete, misleading or offensive information are thus excluded from our study.
The research question is: RQ: What are the policies and practices of online news media for dealing with unpublishing requests implying an individual's right to be forgotten?
We examine the decision-making process for resolving unpublishing requests, by carefully reviewing the processes that occur within a newsroom once an unpublishing request is submitted to a news media organization. We pay particular attention to: whether any written or informal procedures have evolved within newsrooms; who in a news organization is involved in the process of deciding whether to grant a request; who makes the final decision; what arguments applicants make for unpublishing; what criteria decision makers use; and which particular circumstances of a case justify unpublishing.
We conducted in-depth semi-structured interviews with editors of five Slovenian online news media outlets: the editor-in-chief of rtvslo.si, the news website of the Slovenian public service broadcaster (Editor A); the editor-in-chief of 24ur.com, the news website of Slovenia's biggest commercial broadcaster (Editor B); the editor-inchief of slovenskenovice.si, the news website of Slovenia's biggest daily tabloid newspaper (Editor C); the editor-in-chief of Delo and delo.si, the news website of Slovenia's biggest daily broadsheet newspaper (Editor D); and an editor from finance.si, the news website of Slovenia's biggest daily business newspaper (Editor E). We interviewed these editors as key gatekeepers, responsible for both publishing media content and unpublishing it. All editors were interviewed in person, in their offices, and all interviews were recorded and then transcribed. 11 We interviewed the editors on two occasions: first in February or March 2019, and second in Autumn 2019. When we first approached the editors, we asked them whether anyone else in or outside their newsrooms had a relevant role in the process of resolving unpublishing requests, and could therefore give us additional information about the procedure. Based on their answers, we performed one more interview with a data protection officer (DPO) at Pro Plus, the media company that publishes 24ur.com (DPO A).
In addition, we asked the interviewees to provide documentation on their media outlet's unpublishing cases before conducting the interviews. This information was to be as complete as possible, including individual requests, the news items in question, decisions made by the media outlet, and all other correspondence related to the case. Since unpublishing cases are not systematically archived by all the media outlets involved in the research, the interviewees had to search their correspondence. This resulted in a relatively random sample of cases, some of which contained incomplete documentation. Altogether, we acquired documentation on over 30 cases that corresponded to our definition of unpublishing requests. This documentation was fully analyzed, enabling us to partially reconstruct different cases, requests and demands through written correspondence and procedures. Because the documentation was incomplete, we were unable to make full and general statements on the issue in terms of complete document analysis. Nonetheless, the material that we gathered was useful, as it provided significant insights into discussions and procedures, and helped us prepare an interview guide.
For analyzing the interviews, we chose cross-case analysis (see Patton 2002, 440) for each question in the interview, that is, we grouped together answers from all interviewees and analyzed their different experiences and perspectives on our central issues.

Online News Media Policies and Practices for Dealing with Unpublishing Requests Implying the RTBF in Slovenia
The decision-making procedure for the unpublishing of a particular news item begins upon an individual's contact with the media outlet. Email is the formal (and typically the official) path. A general newsroom email address is used for this communication (Editor E), but the public broadcaster has a specific email address for issues involving or implying personal data. Personal contact is the least formal path; it was mentioned by Editor B, who stated that he is sometimes contacted through a common acquaintance asking for a favor.
The applicant contacts the newsroom independently or through a lawyer. The content of communication regarding unpublishing requests can be classified in three categories: informal requests; legal or quasi-legal requests referring to or implying certain legal argumentation; and official demands from an applicant's lawyer, threatening to file a court case.
Informal requests are mostly genuine, polite and reasonable ("These are above all attempts; they just try to send them, they don't make threats" -Editor A); demands, on the other hand, are sharper and threaten legal action in case of noncompliance ("Highprofile people, usually assisted by their law firms, warn us that if their requests are not granted we will be confronted with legal procedures" -Editor E). These requests frequently lack a legal grounding, proceeding rather from a place of common sense.
Legal or quasi-legal requests stem from a strong belief that it is the applicant's right to have the news item in question removed. Some of these requests refer explicitly to the RTBF, as the Google v. Spain case applying to search engines has made it well known, while others state that unpublishing a news item containing their personal data is their right according to the GDPR. When referring to the GDPR and/or the RTBF, the applicants (or their lawyers) often do not quote a precise article, precedent case or similar. "They have heard about the GDPR and then everything is mixed-up … They even think that court reporting falls under the GDPR … " (Editor C). Some applicants do not mention a specific legal source, but just request deletion: "One lady did not refer to any articles or directives … she just said that she had been acquitted in court and asked if we could remove [the articles]." (Editor A) Finally, there are official legal demands made by applicants' lawyers, quoting the GDPR in detail. An increasing number of these cases refer to the RTBF as their legal precedent, believing that it also applies to news media.
Whether they use legal argumentation or not, the applicants claim that the continued publication of the offending news item is damaging to them. Yet their reasoning is modest and superficial: "They request and explain. They don't give a lot of detail, they just write that it's harmful to them now, when they're looking for a job." (Editor A). This applies to both news articles and visual content, particularly photographs (Editor A, Editor C). Applicants usually find the news items embarrassing, and detrimental to their professional career and/or personal life. As Editor A established, "they mainly plead that in their present lives [continued publication] damages them, mostly when they're looking for a job".
In some cases, the applicant's appearance in a news item they now want removed was consensual, even helpful to them at the time. This request was from an individual who appeared in a news article ten years ago: "There was a column 'this is my work, this is my education, this is my profile, I am looking for a job, and I am ready to do any job'. He claims that he has changed his job twice in the meantime, that he is now successful, but this article prevents him from progressing in his career, as people see him as a loser because he was looking for a job in 2009" (Editor D). Another situation involves giving consent once, but now wanting to withdraw it: "The most typical case that I have at Novice, at least once or twice a month, is a girl who at the age of 18 to 20 agreed to be photographed in a bikini,/ … /and it was written that she was a hot Taurus, who liked to go to the cinema and was in love with Leonardo DiCaprio. Now this Taurus is all grown up, she got a job, became a mother, but this publication is still among the top hits under her name. I have several requests to withdraw such publications." (Editor C) In other cases, the applicants' past media appearances were not voluntary. They became objects of reporting based on their newsworthiness and the public interest recognized by journalists. These include court reports, particularly when an applicant has either been found innocent or has already finished serving his/her sentence. The applicants cite not being convicted in court, and assert that the publication "is irrelevant today or not in the public interest" (Editor B).
The procedures within newsrooms are not precisely defined, and are frequently vague. The editors did not quote any written company or newsroom guidelines or clear internal policies on procedures and criteria for dealing with unpublishing requests. Editor E mentioned some "informal guidelines", while Editor B referred to the fact that their company "as part of the company CME 12 has adopted codes of ethics of reporting". However, these are general codes that do not apply to the GDPR or the RTBF explicitly, and thus do not provide specific guidelines for unpublishing cases.
When describing procedures, the editors interviewed used words such as "mostly", "generally", and "in principle", thus revealing potential differences and inconsistencies. The decisions are, as mentioned, usually made by the editor of a news websiteyet, they are often assessed on a "case-by-case basis" (Editor E), and the editor "eventually" consults with the media outlet's legal representatives (Editor C, Editor D). Other people within the media outlet can be involved: e.g. The Web editor (Editor C, Editor D), the director of the news program, or the daily content editor (Editor B). Although the media outlet's legal representatives and advisors can be consulted (Editor A, Editor B, Editor C, and Editor D), the editor does not always get official information or feedback from them (Editor A).
Responses and procedures can also depend on the formulation of a request, and the persons involved: the informal option is usually dealt with by editors who frequently decide in an informal manner, but sometimes consult the legal department or DPO. However: "If we receive a letter [from an important Slovenian lawyer], then this is a serious matter and we immediately send it to the legal department, which deals with it on its own" (Editor A).
The procedures are not precisely defined and leave space for editors to decide based on conscience, friendship, or other personal or subjective criteria, rather than official proceedings within the media company. Different practices have thus evolved, without definite criteria for granting or rejecting a request, which leads to somewhat inconsistent decisions.
The interviewees shared a general opposition to the removal of news items from online archives. Some claimed they had never removed an article or other content (Editor E), and others that they were prepared to do so only in exceptional circumstances, although some were more indulgent than others. Editor D, for example, said: "Our standpoint is that we interfere with the digital archive only when/ … /a court decides that an article must be removed. But certain life situations can fall into a grey area." The main argument for refusing an unpublishing request is public interest. Finding an adequate solution when balancing the public interest on one side and the RTBF on the other is not always easy or without doubt.
The dilemma of whether the public interest outweighs an individual's RTBF also appears in cases where individuals have undergone one or several investigative or legal procedures, yet have never been found guilty in court. Editor E explains: "We have a banker, for instance, who has been involved in a series of weird businesses/ … /legal procedures were taken against him, including in court, but they later failed, or were stopped for different reasons. This man is legally completely clean./ … /And then this person contacted us through his lawyer, saying 'you mentioned me in 30 news articles, in a negative context, and I demand that you erase them'/ … /, and we decided not to unpublish the articles, as we will also not go to the National University Library to burn old copies of Finance." In situations where the continued publication of a news item cannot, in the editor's opinion, be justified by the public interest (i.e. when information is deemed irrelevant and/or outdated) it is usually removed. Editor C tries to be empathetic in situations where a news item is outdated and irrelevant to the public: "If it is a completely human request, such as 'they tease my children because three years ago there was an article about us needing help, and now my children are upset', this we absolutely grant." The editors said they sometimes rely on "additional options" when formulating their responses. These are in line with the options mentioned in the Google v. Spain judgment: "Publishers of websites have the option of indicating to operators of search engines, by means of particular exclusion protocols, such as 'robot.txt' or codes such as 'noindex' or 'noarchive', that they wish specific information published on their site to be wholly or partially excluded from search engines' automatic indexes". Thus, while not unpublishing the story or content itself, editors have the option of excluding it from search engine indexes.
A number of editors used this option: "The most we can offer is some kind of compromise: that is, we send a request to Google to remove the articles from their index" (Editor E; similar statements regarding re-indexation were made by Editor C and Editor D). Certain editors also mentioned "anonymization" within content: removing names, particularly of "victims or unrelated persons", and either replacing them with initials or removing them completely (Editor C, Editor D). As a result, editors seemed more understanding and willing to grant (informal) personal requests than legal/quasi-legal ones, or official demands.
The editors stated that the number of unpublishing requests and demands based on the GDPR and the RTBF is increasing. They also agreed that this places additional pressure and potential limitations on freedom of expression. Several editors interviewed (Editor B, Editor C, Editor D) were glad that this issue was being researched in Slovenia as they are increasingly worried about how to respond to such requests and demands. Editor A said that the research was being conducted "too early" as "the real explosion of demands will come in future months and years", and Editor E was glad that the research would be published in a scientific journal rather than in a newspaper or magazine with massive circulation, as the latter "would actually promote the RTBF and more people would probably start sending legal demands regarding their stories and cases".

Discussion and Conclusion
The analysis of media procedures and policies regarding unpublishing requests based on the RTBF identified three key potential media policy issues in contemporary newsrooms: unclear self-regulation: a lack of defined in-house policies and procedures in the companies analyzed; an unclear regulatory framework for news media: a lack of sufficient clarity and precedent cases regarding the application and implications of the RTBF for the news media; and an unclear regulatory framework for wider digital and analogue archives and other potential distributors of specific content.
Each of these three areas contains specific dilemmas regarding freedom of media, freedom of expression and potential threats.
Regarding the self-regulatory framework, our results showed a lack of consistent procedure and criteria for handling unpublishing requests based on the GDPR and RTBF, and a large maneuvering space for editors to use their discretionary powers when deciding whether to remove certain content. These blank spaces within the editorial decision-making process (including informing the public about such cases) increase news editors' influence and power, and the potential for abuse and manipulation of information through the non-transparent unpublishing of potentially controversial or negative (yet truthful) news items about themselves.
There is some confusion surrounding the legal framework for how (and whether) the news media should implement the GDPR and RTBF. The claim that news media are also search engines since they offer a search function on their news websites runs counter to the decision in the case of Google v. Spain. In that case, the court clearly separated news media from search engines, even though news media websites have offered a search function since the 1990s (years before the court's decision), meaning that this judgment excludes news media. However, public statements by some influential lawyers in Slovenia confirm and increase the confusion that the RTBF has brought to news media and legal systems in the EU. This confusion might allow potential new cases to claim that news media are search engines, and should thus implement the RTBF in relation to their own news stories and content. This not only endangers the transparency of media content and its potential deletion and manipulation, but also threatens freedom of the media and freedom of expression.
Regarding the wider regulatory framework for archives and other databases of news content, one practical problem of implementing the RTBF should be considered: unpublishing a news item from an online media archive usually does not erase all traces of its online presence. Even if it is deleted elsewhere, the erased personal data will remain accessible in the digitalized archives of print media, which are generally stored and provided by national libraries (Ov cak Kos 2019).
As with many other aspects of digitalization, (attempts to achieve) the unpublishing of news items in digital media, and the application of the RTBF to such content, confirm the "dual-use" approach to new digital technologies, and to digital journalism: the schism in the perception and use of digital media and digital journalism. While digital tools enable news media to enhance their key journalistic aspects (collection, production, distribution), they also pose new threats to its authenticity and credibility, while potentially decreasing public trust. This confirms the old schism and presents new examples of dual use of new technologies, not only by showing their empowering potential within newsrooms, but also by confirming claims that they can "be deployed for beneficial purposes as well as exploited for/ … /repressive ends" (Feldstein 2019, 50). Within this context, it is highly relevant that legal and self-regulatory frameworks are updated, providing clear, consistent and comprehensive set of rules for the requests involving or implying the RTBF for news content in news media. The relative lack of precedents contributes to opposing interpretations of the RTBF and its potential use in news media, and clear and more extensive decisions, descriptions and explanations of particular legal tools such as the RTBF should offer greater clarity to editors and the public regarding legal implications. At the same time, an updated self-regulatory framework of policies and procedures needs to be established in individual countries as well as individual media companies so as to reflect their specific editorial polices and values.
This article is based on interviews with media stakeholders in a single country (Slovenia). Other countries might provide different insights into these editorial practices. In addition, further interviews with other stakeholders and the use of different methods could provide a greater insight into legal practices and arguments. However, our analysis still shows that the legal foundations within Slovenian and EU frameworks and policies for unpublishing online news items upon requests implying an individual's RTBF are vague at best, and often non-existent. This creates a dangerous combination of circumstances, as any vagueness (or perceived vagueness, seen in attempts to apply the RTBF not only to search engines, but also to news media) makes room for new requests, and thus opens the door to potential limits on freedom of expression. This could have a chilling effect on digital media and digital journalism.
In order to avoid the possibility of manipulating media history and eroding journalistic authenticity and credibility, both legal and self-regulatory frameworks need to be updated. A new, more precise framework is required that will sufficiently and efficiently support journalistic autonomy and editorial independence, and prevent further non-transparent digital manipulation of the past. Clear criteria for the eventual removal of news content need to be defined and made available to the public. This is needed to ensure media transparency and responsibility, as well as to prevent arbitrary interventions by either the state or individuals, through their networks and through their personal (economic, political) influence. If not, freedom of expression in the digital eco-system will be under threat, and the potential for new forms and shapes of digital manipulation will continue to develop.