Molo.news: Experimentally Developing a Relational Platform for Local Journalism

In this articlewe present a research project that experimentally develops a local news platformbased on empirical research (interviews, group discussions, a survey) and a co-creation approach. What is presented here is not a typical empirical social science research study but the culmination of an entire approach that is oriented toward software development. This article’s aim is to present the project’s conceptual ideas, its interdisciplinary character, its research-based development approach and the concept for a local news platform that grew out of our preliminary work. At each level we focus on the relationality which arises in the figurations of the actors involved and their various perspectives. First, we illustrate how relationality already shaped the objective of our project and how this results in its interdisciplinary structure and research design. We then discuss this idea with reference to our empirical findings, that is, the paradox of the local public sphere: While all the actors we interviewed—those who (professionally) produce content and those who use it—have a high appreciation for the idea of a local public sphere, themediated connection to this sphere is diminishing at the same time.We understand this as the real challenge for local journalism and the local public sphere at large, and not just for individual media organizations. This is also the reasonwhywe argue for a fundamentally relational approach: from a theoretical point of view, it can be used to grasp the crisis of the local public; from a practical point of view, relationality represents the core characteristic of the platform in development. On this basis, we will then show how the concept of the experimental local news platform evolved through the use of a prototype as a relational boundary object. This development lead to the conceptualization of the platformmolo.newswhich itself is characterized by a fourfold relationality. Our concluding argument is that approaching relationality in a more rigorous way could be the key to exploring the future of local journalism.


Introduction
The quote above comes from a composer from the German 'New Music' scene during the 1920s. Paul Hindemith experimented with new sounds (rugged rhythms, harsh dissonances, inclusion of jazz elements) and worked closely with an experimental public radio sta-tion in Frankfurt am Main, Germany. While the quote refers to experimentation in the field of music, it says something about the point we want to make here: We want to introduce the 'experimental' platform for local journalism that we are currently developing-an experiment, that was only made possible exterior to but still in cooperation with media corporations. This platform is called molo.news, where 'molo' acts a truncation of the phrase 'moving local': Moving local news for people in motion and those who want to bring movement to the urban public. Thus, 'moving local' represents both aspects: moving in local space and being moved by local news. The most important lesson this project has taught us so far is that we should think of such a platform as relational: Relational in the way it had been developed, relational in how it should work, and relational in regard to what kind of discourses it might support.
The aim of this article is to present the platform's conceptual foundations and to show how it was developed in a combined approach of empirical research and cocreation with various stakeholders and user groups. This remains a rather untypical research and project design in the social sciences and this is also the reason why we want to present the overall design of the project including the platform's framework instead of only focusing on specific components (such as the empirical research, the co-creation approach, or the front-end of the platform). This article is, if you like, a report on experimental research and software development.
The starting point for our attempt to develop a local news platform was a call for proposals from the German Ministry of Education and Research which was aimed at research projects with practical components that address 'social cohesion. ' We submitted a proposal for a project called "Tinder the City" which proposed trying out new ways of developing a local news platform. We called the project "Tinder the City" because we were thinking of exploring the possibilities of developing something that could bring about a similar movement in the field of local news platforms as Tinder did in the field of dating apps. The special character of the project results from the fact that it combines empirical social science research with software development. In other words, everything is aligned towards developing a 'real' functional product.
Our overall theoretical framework is characterized by a "figurational approach" (Hepp, 2020), an approach that has made us particularly sensitive to questions of "relationality" (Emirbayer, 1997)-questions that closely relate to the discussion about a 'relational thinking' in sociology, as can be identified in, for example, recent network approaches (Castells, 2009;Fuhse & Mützel, 2010), assemblage (De Landa, 2006;Latour, 2007), and systems theory (Holzer, 2010). The figurational approach that we put forward here goes back to the process sociology of Norbert Elias (1978) who argued that we should not position the individual against society but understand society as being made up of different 'interweavings' of 'interdependent' individuals which he called figurations. From this perspective, a family is a figuration, a group or a community is a figuration, as are the relations between journalists and their audiences. These figurations are defined by the shared frames of relevance of the individuals involved, a characterized constellation of actors between them (including particular roles), and are continuously constructed through the practices of those involved which, contemporaneously, are invariably entangled with a particular media ensemble.
Taking this theoretical framework as a point of departure and relating it back to city publics we become sensitive to questions of relationality-that is: the relations of individuals in and across particular figurations. We can describe the city itself as a "figuration of figurations" (Couldry & Hepp, 2017, p. 72), that is, a complex web of figurations of different organizations and communities implicated in an urban cross-media public (see Hasebrink, 2017, andespecially Hepp, Simon, &Sowinska, 2017). This highlights the importance of asking which relationalities exist within each respective figuration and how a platform can create, or at least reflect, relationality by bringing together the different figurations of a city or city public.
The idea of such a platform refers closely to the changing figuration of journalists' relationship to their audiences. Like any social connections, this one is relational: characterized by more or less congruent mutual expectations about what journalism should deliver and what audiences might expect, and the more or less mutually visible practices that emerge as a result (Lewis, Holton, & Coddington, 2014;Loosen & Schmidt, 2012). Within the culture of journalism, the shifting notions of "the audience factor" are also the expressions of a changing media environment, of new media practices and, finally, of the ways in which journalists engage with audiences in a digitally networked media environment (Loosen, 2019). Just as with the city and its public sphere, we can also understand the relationship between journalism and its audiences as a communicative figuration that transforms with the media ensemble and the communicative practices on which it is based (Kramp & Loosen, 2017). This development is, however, ambiguous: On the one hand, over the past few years news organizations have offered a growing number of participatory spaces and features; on the other, journalists are often reluctant to engage with audiences while users differ to a large extent with respect to participatory practices and motives (Costera Meijer & Groot Kormelink, 2017;Loosen & Schmidt, 2017).
In a project aimed at the development of an experimental platform, a figurational approach can also be switched around in a more self-reflexive way. Then the question turns to which concrete figurations can such a (relational) platform be developed. At this point, our previous experience in software development was important for the project. This includes initial experiences with software development in the field of jour-nalism  and more recently with approaches to "co-creation" in research software development (Berg & Hepp, 2018;Hasebrink & Hepp, 2017). This interest in the development of technologies goes hand in hand with our own empirical research on "pioneer journalism" , which isas with data, robot and sensor journalism-becoming more and more technology-oriented and in which new approaches to software development are increasingly being pursued. The experience gained from these earlier projects demonstrated that the development of software is a social process that happens within certain figurations. It is possible, indeed it is expedient, to 'create' certain figurations situationally-and this is exactly what the 'co-creation' approach stands for: the inclusion of different stakeholder groups in software development from the beginning and across different stages of the process.
To discuss in more detail how important relationality in and across different figurations is for the development as well as for the functionality of a local news platform, this article is not so much about the empirical research we carried out, rather, its principle aim is to discuss the process as well as the concept of the experimental local news platform we have developed-and still are developing-with a particular focus on the significance of relationality. We want to proceed to this end as follows: First, we want to describe in detail the paradox of the local public sphere that our empirical research revealed, a paradox which we see as a problem of shifting relationality. On this basis, we then want to show how the experimental nature of our project has resulted in the prototype emerging as a relational boundary object. This then leads to a presentation of the basic concept of molo.news as a relational platform. In conclusion, we will then argue why, in our view, relationality might offer an opportunity to explore the future of local journalism.

From the 'Crisis' to the 'Paradox' of the Local Public
In many Western countries, there has been emerging what we can call a 'crisis' of the local public. The reasons for this are manifold and intertwine in complex ways. With the advance of deep mediatization-the increasing saturation of the various domains of society by digital media and their infrastructures-journalism and its production and distribution routines as well as practices of use are comprehensively changing.
There are various concrete examples when it comes to local journalism and its digital transition (Jenkins & Nielsen, 2018;Nielsen, 2015;Van Kerkhoven & Bakker, 2014): Increasingly, in addition to actual reports, local journalists try to reach younger audiences through social media channels and are often now expected to deliver news by means of digital photographs or even video clips; local news is expected to appear without delay on a newspaper's website; and the work process itself has changed fundamentally with the advent of the digital news desk. At the same time, the relationship between journalism and its audiences has evolved as readers enter into direct communication with journalists via Twitter, Facebook, WhatsApp, forums, and other platforms or services, communicatively challenging journalists in completely new ways (Ekdale, Singer, Tully, & Harmsen, 2015;Loosen, 2019;Wenzel, 2019).
The once comparatively 'protected' position of local and regional newspapers has come under pressure: When national and global media are readily available online, much smaller, local newspapers are forced to compete. Sections on national and global news make less and less sense, since at this level a local newspaper can hardly succeed in competing with the large national and international media houses (Abernathy, 2018). At the same time, this once protected position is being undermined locally when local parties, groups and movements begin to communicate directly with their supporters and critics online, weakening journalism's intermediary role (Neuberger, 2018). In addition, the boundaries between local media become blurred when, for example, radio stations offer local news on their websites for 'free.' Additional financial pressure is applied by the fact that advertising is increasingly shifting towards online platforms and the websites of local newspapers no longer hold the status of a "premium environment" for advertisers (Jenkins & Nielsen, 2018).
There have also been considerable shifts from the users' point of view. Particularly with younger people, their bond with local newspapers is crumbling (Tang & Lai, 2018). If they do read local news they tend to do so online while also receiving a considerable share of their information via platforms such as Facebook, Instagram, and YouTube (Swart, Peters, & Broersma, 2019;Weichert & Kramp, 2017). In general, their media use takes place across a broad "repertoire" (Hasebrink & Popp, 2006;Kobbernagel & Schrøder, 2016) of media (platforms, services, offers) grouped around digital devices such as smartphones, laptops, or digital television sets. The expectation of at least being able to interact in principle is increasing while interest in traditional local journalism appears to be declining (Barthel, 2018;Ytre-Arne & Moe, 2018).
The consequence of all these dynamics is that local newspapers in many countries are coming under pressure and in some urban centers and regions they have completely disappeared. This poses a problem to the extent that the primary medium through which a local public sphere has been created over many decadesthe local newspaper-no longer exists (Jenkins & Nielsen, 2018;Leupold, Klinger, & Jarren, 2018). Here, a local political exchange and dispute has taken place, the outcome of which is not necessarily consensus, but solutions for living together can still be found. Robert E. Park (1967) already pointed out the necessity of local media's mediating role back in 1925 in the context of the increasing social and cultural differentiation occurring within cities. With the loss of local newspapers, the city public seems to be descending into 'crisis.' At this point, we argue that we can understand these changes and resulting challenges as a shift in the relationality of the figurations of local news production, the figurations of the journalists' relationships with their audiences as well as those of the city as a local space and public: The practices of local news production as well as the roles of local journalists transform as does the figuration of local news production; from the emergence of social media platforms and other kinds of media technologies the relations between journalists and their audiences shift; and the relations of the different figurations that make up a city transform if all communicate with each other in new ways. Therefore, a transforming relationality becomes the broader frame for an understanding of what is called the 'crisis' of the local public sphere.
We have taken this diagnosis as our starting point to investigate the communicative figurations of Bremen in Germany. For this purpose, we have worked with a mixed method design that triangulates various means of data collection and analysis: Qualitative interviews and focus groups as well as a representative survey were conducted which were then analyzed statistically and by means of qualitative coding according to the procedures of Grounded Theory (Glaser & Strauss, 1999). In total, we have collected the following data: • Citizens: Seven focus groups on the local public with people from different social backgrounds (aged 15-42); a representative survey on local media use (n = 727); • Collectives: Eight focus groups with representatives from sports and cultural clubs, citizens' initiatives, etc.; • Media and information providers: Qualitative interviews with thirteen journalists, seven party representatives, three representatives from district councils and one member of a business association.
This is not the place to present the results of this empirical research in detail. Rather, our aim here is to condense our findings in response to the diagnosis of a local public crisis and its underlying shift in relationality. In the course of doing so, it seems appropriate to not speak simply of a crisis of the local public, but of a 'paradox' of the local public. Essentially, in referring to the 'paradox of the local public' we mean to say that a high level of interest in the local public and its general esteem are thwarted by a simultaneously decreasing connection to it. Our survey as well as the focus groups and interviews demonstrate that there is interest in local topics and local events which becomes more pronounced the closer they encroach on an individual's everyday life (neighborhood, work life, topics of personal interest, etc.). Local media are considered in high esteem in order to ensure an appropriate information flow and generate discourse surrounding (public) urban life. In this respect, the groups of actors we researched barely differ.
Despite this relative lack of differentiation "public connection through media consumption" (Couldry & Markham, 2006, p. 251) decreases, especially among the young. While older people generally feel well informed by local media, and young people have the impression that these traditional local media do provide good information, the latter hardly use them at all-or if they do only very infrequently. An increasing proportion of people are, therefore, untouched by local media. At the same time, in the case of most users we investigated this gap is not filled by platforms such as Facebook or Twitter as their relevance to local information is negligible. Rather, citizens often become aware of local news through WhatsApp or other messenger apps, but above all from personal conversations with their fellow citizens which are highly focused on particular topics. There exists, therefore, a considerable gap between the desire for local information and exchange on local issues and the available possibilities for local, mediated communication.
In the case of Bremen, we can argue that-in the context of media change-relationality fundamentally transforms within the figurations of news production, in the relationship between journalists and their audiences, and that of the different groups and communities. However, in the city a new integrated communicative relatedness-what we typically refer to as the (local) public-only exists to a minor extent. This is in part caused by the "path dependency" (Garud & Karnoe, 2012) exhibited by current local news media. In the case of Bremen, local newspapers, television and radio stations, and local media enterprisesespecially newspapers-are primarily oriented towards developing digital offerings (websites, apps) that serve to aid the sale of their previously printed and broadcasted content which, in essence, opens up an additional distribution channel. They act and are caught up in the patterns of their own organization. It is this 'path dependency' against which we position our idea of the local news platform molo.news: If we plan to overcome the paradox of the local publishing sphere through a platform like molo, it must work as a tool for offering a new kind of relationality across figurations instead of reproducing the path dependency of existing media organizations' figurations. The idea is to develop a 'relational platform,' in which collectives (associations, citizens' initiatives, social movements, etc.) can have a say besides journalists and, ideally, in which the content of all local media are available.

Experimenting or Prototyping as Co-Creation
Keeping in mind our search for a relational answer to the paradox of the local public sphere, we understand our software development of a local news platform as 'experimental' because our aim is to sound out the 'scope of possibilities' in a way that local newspapers or local news providers are unable to owing to the path depen-dency of their organizations. We are able to act on the basis of empirical research and without economic pressure in order to develop a platform that comes as close as possible to the idea of relationality. Ideally, this platform would be established in the City of Bremen (with the possibility of expanding to other cities), but first of all we wanted to experimentally demonstrate what form such a platform might take. In pursuit of this goal, we have used an approach from software development called cocreation. This method of development is in itself relational because its underlying idea is to integrate the various figurations of actors that will use the platform into its development from the beginning. For this, situational figurations of co-creation workshops are set up which create the space to 'relate' different expectations and requirements. This is achieved by using a stepwise developed prototype as a 'relational boundary object,' that is, as a shared point of refence to build up the platform's detailed concept of relationality.
The current research discussion on co-creation has two main origins. First, it represents a reorientation in business informatics towards the customer or user and is accordingly expected to support market success (Piller, Ihl, & Vossen, 2010). This kind of user orientation has also become increasingly relevant to public institutions and (local) authorities. It refers to the fulfillment of tasks in their provision of suitable offers for citizens and the enhancement of citizens' participation in the definition and provision of such services (Aichholzer & Strauß, 2015;Nambisan & Nambisan, 2013). References to the challenges of strengthening cohesion in the community are quite clear here-even if the business informatics perspective and approach are different and more directed towards fostering cohesion between users and a product.
Second, and also highly relevant, are approaches that involve users in the design of IT solutions even before the definition of requirements phase. Here, three different "participative design" (Simonsen & Robertson, 2013) approaches can be identified: The ETHICS method developed in Great Britain (Mumford, 1981;Mumford & Henshall, 1979), the Scandinavian approach of the DEMOS and UTOPIA project (Ehn, 1988) and the US "cooperative design" approach (Greenbaum & Kyng, 1991). In these ways of developing co-creation processes, future users and their "collective creativity" (Sanders & Stappers, 2008, p. 6) are ideally included in the entire design process.
These reflections on co-creation in software development are also conducive to our 'experimental' development. Participation in journalism is, to a large extent, still limited to the journalistic post-production phase, that is, offering users the possibility of commenting on and discussing the news (Bergström & Wadbring, 2015). This is not to say that established news organizations and particularly journalistic start-ups are not also experimenting with more sophisticated ways of actively engaging audiences (Hansen & Goligoski, 2018) but we believe our ap-proach goes further in this respect as it is directed toward including both potential users and content providers of a news application before it even exists. In order to achieve this we felt that co-creation-in addition to empirical research-provides the most effective opportunity for the inclusion of a wide variety of stakeholders in the development and design phases of the software at its very initial stage. In addition, this kind of co-creation approach has the advantage that it can be used to sustainably promote acceptance of the platform from the outset against the background of the broad range of services available in the field.
In order to combine co-creation with the expectations of a high level of 'professionalism' for the platform, a degree of professionalism equivalent to the quality of other platforms such as Facebook or Twitter (a user demand which was repeatedly expressed during group discussions with users), we decided on a two-stage development process: In the first stage, we sought to develop a functional prototype of our platform in tight co-creation loops which were then to be implemented in a second stage of actual (re)programming for a platform for everyday operations. This procedure also corresponds with the discussion on the role of prototypes in software development (Turner, 2016). These take the character of "boundary objects" (Jarke & Gerhard, 2018;Star, 2010): A material artefact which is used to negotiate between different groups of actors, in our case, an understanding of the platform and its relationality.
In this way, and based on an analysis of the focus groups and interviews with the various actors under investigation, we developed both the name of the platform and its core rationale: molo.news, whereby-as already mentioned in the introduction-'molo' stands for 'moving local': Moving local news for people in motion and those who want to bring the movement of a new relationality to the city public. Relationality here means the following: Through this platform, content from established local media (local newspapers, district magazines, local radio stations, etc.) as well as content from city collectives (associations, citizens' initiatives, social movements, etc.) should be accessible in an easy-to-use app (relationality on the level of sources). All this news should be curated (relationality of curation) and provided on one news feed (relationality of content) through which the different users of the platform can anticipate how they would like or actually relate to each other (relationality of users).
We soon learned that such a solution would require a particular backend (an editorial system that is able to bundle a range of content and provide a uniform keyword system) as well as a frontend (the actual app and its user interface on the device). Apart from this basic structural decision, the concrete implementation and design processes were open and were developed step by step as part of the co-creation workshops. In order to illustrate this procedure, in what follows we will concentrate on the development of the app itself and omit a description of the editorial system (backend; for a detailed description of our practical approach, see work by Breiter, 2019, andRoeske &Heitmann, 2019).
A total of eight co-creation workshops were held between summer 2018 and spring 2019 to develop our prototype. The participants were acquired through direct contact at schools and clubs, via flyers, emails, and social media call-outs as well as invitations to courses at the university. The prototype development workshops focused on potential individual users and not on collectives or journalists (who are of particular interest for the development of the editorial system). We worked with people aged 15 to 39 from a range of social backgrounds and levels of formal education. The group sizes varied between five and ten participants. Our workshops had a fixed timeframe and a structured schedule which were outlined to the participants at the beginning of each workshop. The setting of the workshops varied greatly depending on the group involved and what stage of development we were at (see Table 1). In addition to the co-creation workshops we also conducted four feedback workshops in which we sought advice from experts on each respective stage of development.
In each workshop we applied a combination of methods from the input, interview, and creative phases as well as collaborative elements (Roeske & Heitmann, 2019; Figure 1). For example, we worked with the Think Aloud method which finds its origins in design thinking as well as with A/B tests. While the former describes an active articulation of impressions when testing an application (Van Someren, Barnard, & Sandberg, 1994), the A/B test is a comparative test between two variants of the same system (Fabijan & Olsson, 2015). All the co-creation workshops focused on specific problems and questions related to the app. Consequently, design elements and functionalities were conceived as ways of presenting different kinds of content. Here, we worked out design elements on paper with scissors, pens, and support material. After each co-creation session, the project team translated the results of the workshops into a so-called mockup of the app, i.e., a simple digital model, on the basis of which further in-depth work could be carried out in the next workshop while simultaneously informing the definition of the feature requirements that the software developers needed for their work. Each co-creation workshop was planned and conducted to reflect the results of those that proceeded it. To achieve this aim, we established an iterative process of prototyping.
Through this iterative process we were able to interrelate the different stakeholders in the development process. Basically, we can understand each co-creation workshop as an occasion of building up a situational figuration of joint development. However, it remains a somehow power-related and unbalanced relationality: The use of a co-creation approach in our experimental development does not dissolve the distinction between the roles of  'developers' and 'users' but it does make it possible to ease the rigid boundary between them. That said, across the workshops the focus on the prototype as a relational boundary object turned out to be central: Its iterative development has connected the sequence of the different workshops in a meaningful way and it also offered the participants in their practical work an orientation within each respective workshop particularly in regard to its position in the development process as a whole. The practical work with paper, pen, and scissors was used to playfully visualize ideas. This provided opportunities to stimulate the "practical consciousness" (Giddens, 1984, pp. 41-45) of the workshop participants: On the practical level of their 'everyday doing' they 'know' how they would (like to) act with such a news platform, while 'discursively,' for example, in an interview or by means of a questionnaire, this knowledge is not accessible. In concrete terms, we learned a lot about the participants' ideas for a simple, intuitive app, the desire for compatibility with existing applications, and their 'vision' of an open integration of diverse content with simultaneous data security.

Molo.news as a Relational Platform
So, what forms does the prototype that was developed in this co-creative process take? In essence, the platform we developed is characterized by the fact that it establishes a kind of fourfold relationality. Each distinct aspect of relationality operates at the level of sources (providers of content), of content, of curating, and of users (see Figure 2). We thereby associate a shift from the con- cept of social and societal integration through mass media (sources) to that of a platform relationality. Here traditional mass media are only one provider of content among many and all actors (in their roles as providers and users of content) and they have, in principal, the opportunity to perceive each other, to position themselves in relation to one another, and to establish relationships.
At the level of the 'relationality of sources,' the molo.news platform makes a distinction between established (journalistic) media such as local newspapers or local radio stations and non-professional providers of content. Content from established media complies with accepted journalistic standards and is automatically read into the platform's editorial system via RSS feeds. Non-professional content providers are represented as collectives, by which we understand "collective actors" (Schimank, 2010, p. 327) of all kinds who are involved in the urban public. Such collectives can be associations, social movements, citizens' initiatives, and so on. Collectives feed their content into the platform's editorial system for which an easy-to-use input mask was developed. To have access to these input systems collectives must be 'accredited' by molo.news, that is, recognized as reliable sources so that they might be granted permissions to upload content.
Consequently, the relationality of sources means, that in order to involve various local media and collectives in the platform, they must also support the platform through cooperation. Achieving this was first and foremost a social process in which mutual trust had to be established and appropriate cooperation agreements concluded. Currently, all major local media based in the City of Bremen, with the exception of one local newspaper, participate in the experimental platform molo.news. This alone has led to a different form of perception among the actors which, while likely still dominated by mutual competition, is also guided by participation in an overarching project.
The 'relationality of content' should also be seen socially as professionally produced content from local media and content from committed collectives are presented on an equal basis through the platform. The content that enters the editorial system in this way can vary wildly. At its core, however, three forms dominate: news on political and social life in the city; announcements of events such as concerts or demonstrations; and reports from the point of view of individual collectives, for example, on their own events or vocations.
The idea of building up a relationality of content corresponds to the findings from the empirical research and co-creation workshops from which we know that that collectives often feel powerless against the overrepresentation of professional media in the news space and that many users are dissatisfied with classic local journalism and are hungry for new forms of presentation and content. However, the development of the prototype quickly opened up a discourse on the character of 'more appropriate' local news that could be established around the molo.news platform involving both the local media and the various collectives with their individual perspectives.
The 'relationality of curating' has already been mentioned. With a platform that makes accessible content from different actors, it seems barely possible to let the selection processes run purely automatically or on the basis of the automated clustering of content. As recent research has shown, even on platforms such as Facebook, curating and moderation is far less automated than originally assumed (see Gillespie, 2018). The content must, therefore, be curated according to different dimensions, which in turn leads to the technical task of developing a backend system that makes curation possible at a social dimension, namely, the establishment of a hub editorial office for the selection and indexing of the content. Within the editorial system at the backend of the molo.news platform, all content is reviewed by an editorial team: We call it the 'hub editorial office.' It has a threefold task: Fundamentally, it is tasked with screening contributions to ensure their quality and to avoid publishing defamatory or legally dubious contributions. In addition, the hub editorial staff selects contributions, in that it decides which contributions are to be included in molo.news (to avoid duplication of, for example, agency reports) and marks contributions that appear to be particularly relevant for the community with the 'hot' signifier; 'hot' content refers to news that appears in a user's news feed regardless of a selection that a user may have made to personalize his or her own content stream. Finally, the hub editorial team allocates to each piece of content keywords for automatically imported news through which the user can gain access to content across different sources. The main task of the hub editorial staff is, therefore, to "curate" (Thorson & Wells, 2016) the variety of content that users can access via the app on their mobile devices (see Figure 3).
An editorial office like the Hub must be able to assume a neutral position in regard to the various content providers especially if different professional media are to be integrated. This also means that it should be located outside of the editorial offices and workspaces of the various individual providers. Establishing this position and creating the corresponding acceptance for such a metaeditorial office is considered as a social process.
Finally, there is the 'relationality of users' arising from the platform. Users have various options through which they are able to personalize their news feed: They can prioritize content that relates to a defined location and its environs. They can follow certain sources such as certain media whether they be local newspapers or certain collectives such as clubs, associations or initiatives. Finally, the news feed can be personalized according to keywords to select topics of interest. We have taken particular care not to simply reproduce current journalistic departments such as politics, business, and sport. Our considerations were instead guided by analytical considerations and empirical findings about user preferences as outlined in Section 2 on the crisis and the paradox of the local public which has led to the aim of delivering moving news in a dual sense. All news on molo.news is thus grouped along four questions: What moves the city? (moving news from the city); Where can I move to? (events that can be attended); How do I move? (the possibilities of mobility in the city); What can I move? (the possibilities of one's own commitment). However, as already stated: News marked 'hot' will appear in all personalized news feeds regardless of the locations a user prioritizes, as well as the sources and topics they follow. 'Hot' stories ensure that users will receive the information considered most relevant by the community itself.
The relationality of users is, therefore, fundamentally a social phenomenon as molo.news aims to bring together various figurations of organizations and communities as mutual audiences in order to foster cohesion in the city. At its current stage, we have made the initial steps towards this orientation through the co-creation workshops in which we have involved various groups of people as future users. The extent to which such a relationality can then also be realized in everyday practices will only become apparent when we go public with the app.

Conclusion: Relationality as a Chance for Local Journalism
So far in this article we have presented the development of and the conceptual idea behind molo.news. The connective notion across the project has been that of relationality: The relationality of each stakeholder's figurations that were involved in the development process, the role the prototype plays as a relational boundary object, and the relational concept of the platform itself.
For us, this focus on relationality was an important step in the process of developing our experimental local news platform-a step that was certainly carried in part by our theoretical starting point of the figurational approach. Beyond the specific role relationality played for our project, however, the idea of relationality seems to us an important basis from which a wider discussion on the future of local journalism can take place. Local journalism is often thought of as an institution that is supposed to maintain one inclusive public sphere for all actors within a city (politicians, communities, citizens). This idea has strong roots in the world of legacy mass media whose publics were constructed via certain distribution arenas and in which journalists played a prominent role as gatekeepers. This has changed with the progression of deep mediatization and the spread of platform media: Various actors from all walks of life are communicating on multiple platforms in addition to professional journalists such as the collectives outlined in this article (social movements, neighborhood initiatives). These collectives might even raise their voice in opposition to journalists who at the same time are losing sight of their role as gatekeepers and are struggling to reach younger local audiences.
As we have argued in this article, from this perspective we are confronted with the 'paradox of the local public': Various actors have a great interest in local news and information but at the same time notice that current structures of the local public no longer seem to function. At this point, relational thinking opens up a completely different space from which to think about the problem: Perhaps the progression of deep mediatization is about realizing the opportunity of platform media in terms of their relationality, making the most of diverse content from different sources accessible in their curated rela-tionality to one another and positioning local journalism within this space. This is not to say that a local newspaper article is the same as, for example, the information from a neighborhood initiative about its latest project. But perhaps these spaces of relationality across the different figurations of a city unlock the prospect of journalism repositioning itself in relation to other local voices and not simply equate itself with them.
Surely, then, completely new questions arise: For example, how can such relational platforms be financed in the long-term? What other business models for the production and distribution of quality journalism exist? Whose content is made accessible through these platforms and whose is not? We have not yet found definitive answers to these questions during the development of the momo.news project and we, too, are still looking for a sustainable business model for the platform. On the basis of the current academic discussion and research, the model of a cooperative currently seems to us to be the most sensible (see the discussion on "platform coops," in particular, Scholz and Schneider [2017]; similar thoughts are expressed, for example, in studies on the "cooperative regulation" of platforms like Helberger, Pierson, and Poell [2018]). The crucial point here is that a project like ours creates the space for these questions to be raised without adhering to the path dependencies of established local newspapers' and local radio stations' organizational structures. At least, in principle, there exists the possibility that some of these media organizations will disappear while making way for new, innovative organizational forms and business models. In some ways, we see our project as an attempt to experimentally explore spaces of possibility. In this sense, the experiment continues and we very much hope that it is possible for us to "create something enduring" as Paul Hindemith, quoted at the beginning of this article, also intended.