Prof. Dr. Maren Hartmann

I would like to start with a call for more research: This may sound like navel-gazing and self-assurance on the part of the scientific guild, but it is driven by content. I'm not talking about research per se, but research in relation to the digital divide, as it is often called in German. At the same time, I would like to call on people not to allow this research to run under this label and the basic assumptions associated with it. This also applies to the title of this article: Mobile media, lifestyle and digital divide/exclusion; because the connections that are established here (between lifestyle and media use on the one hand and these two and social exclusion on the other) can only be outlined to a very limited extent with the term digital divide (see below). At the same time, the concepts of mobile media and lifestyle are reasonably directly accessible, but the concept of the digital divide requires a more detailed definition. I will therefore briefly outline the concepts in the first part, combined with a look at the development of digital divide research over the last two decades. I will also address the question of the extent to which mobile media represent a new direction of development. And finally, I will come to a case study that emphasises the ultimate ambivalence of the developments.
Origins of digital divide research: the have-nots

Firstly, a brief working definition of the digital divide: in principle - building on the question of the above-mentioned connection - it is about the opportunities of access to and use of information and communication technologies (ICT) or the internet. The basic assumption is that these opportunities are unequally distributed in society. This unequal distribution in turn leads to different access to information and knowledge and therefore to negative social developments in the long term. The question of possible connections between exclusion and inclusion processes and media use has become more relevant than ever thanks to developments in the ICT field in recent years.

In principle, the question of the digital divide is a question of how we want to shape our society in the future, because it is a question of participation and its framing. And we have to ask ourselves this question again and again. Do we still have the framework conditions that we originally assumed? The basic assumption behind the concept of the digital divide is, among other things, a normative idea of democracy and corresponding forms of participation, as the following, somewhat older quote shows: "Due to the distance of offline users to the new developments in the field of communication, they are increasingly and possibly in the long term decoupling themselves from important forms of social information transfer and communication" (Grajczyk/Mende 2001: 409). A possible prevention of such decoupling processes was or is the overriding goal. However, in addition to the changes in the framework conditions, the forms of information transfer and communication and thus social participation in general may also change.

But let's go back to the beginning: when digital divide research began to take shape under this term, the initial focus was on differences between first and third world countries. It was feared that existing inequalities there would be exacerbated by the lack of Internet use. The main reason for this was initially seen as the lack of information and communication technologies locally, i.e. a lack of access opportunities. This problematisation of access was relatively quickly extended to other areas. For example, Pippa Norris (2001), one of the better-known theorists of the digital divide, distinguished very early on between a global, a social and a democratic divide. The global divide refers to the aforementioned differences between industrialised nations and developing countries, the social divide in turn refers to the difference between the information-rich and the information-poor in each nation, while the democratic divide refers to a further possible difference between users, i.e. between users who use the technologies to participate politically and those who do not (Norris 2001: 1-4). A problematisation of the concept of access already becomes clear here, as access alone is by no means sufficient as an indicator of opportunities, especially when it is narrowed down to the question of political participation. In this way, Norris moves from a differentiation of the original distinction between haves and have-nots to a distinction between information-rich and information-poor. More pointedly, one could create a differentiation between democracy-rich and democracy-poor. This in turn emphasises that media use alone cannot be the decisive differentiation.

Over the years, however, we have increasingly moved away from this binary distinction anyway and now speak more of a coupling of several indicators: "The dynamics of Digital Divides vary, but they generally refer to the way that various combinations of levels of digital access, digital skills, and digital literacy within situations of social exclusion result in individuals experiencing different levels of access to material, social, and digital goods" (Wessels 2015: 2802). A combination of (learnable) skills, but also ways of using or applying those very skills, expands (and complicates) the spectrum of access.
Further differentiations: Of rainbows and spectra

Along with these differentiations, the term digital divide was also gradually criticised, as it still reflects binary thinking. The counter-proposals range from an access rainbow (see Fig. 1) to digital distinction (Zillien/Hargittai 2009) and digital exclusion or inclusion (e.g. Helsper 2012).

The Access Rainbow is an early attempt to differentiate between various basic technological requirements, content, skills and, not least, actual participation. It works according to the principle of accumulation, i.e. the more of these aspects are present, the higher the degree of digital inclusion (at the same time, the parts are not equivalent, but build on each other). However, it is precisely this principle of accumulation that is neither suitable for contextualising usage nor for mapping possible user choices or the like. This means that even this breakdown is ultimately under-complex. Here too, however, the following applies: The direction of travel is helpful.

Another - albeit also older - suggestion can be found in the idea of the spectrum of internet usage (see Fig. 2). While the 2002 version shown here would certainly have to include further differentiation points and renaming in the meantime, the basic idea of the spectrum initially enables a less strong evaluation of use versus non-use, but also does not demand pure accumulation. At the same time, the spectrum indicates different forms of utilisation and motives. An extended basic orientation, which combines spectrum and rainbow, could offer itself as a basic model for the complex field of the connection between exclusion or inclusion and media use. While there is little that is radically new in the modelling, the empirical foundations of the differentiations have become increasingly better - and are therefore also problematised differently.
An emphasis on voluntary non-users

The phenomenon of want-nots, i.e. those who have used the internet and then deliberately left it again, or, as Sally Wyatt (1999) put it at the time: "They came, they surfed, they went back to the beach: why some people stop using the internet", is already being hinted at in the spectrum with the net dropouts. The reasons for non-use in particular must therefore be differentiated. Non-use can be a conscious decision or even a privilege and would then not easily correspond to the assumption of social exclusion mentioned at the beginning. Zillien (2008), for example, points out that a distinction must be made between utilisation or non-utilisation as such and the reasons for such non-utilisation. According to Zillien, socio-demographic factors (age, gender, education, income, etc.) actually play a major role in the distinction between utilisation and non-utilisation as such, but the reasons for non-utilisation show no clear relationship to socio-demographics. At the latter level of reasons, non-use is as diverse as use (ibid.), suggesting that too simple conclusions regarding causality are not helpful. Zillien crystallises the following as the main reasons for non-use: "1. lack of motivation: no need/interest; 2. lack of skills: low computer/Internet skills; 3. material barriers: Lack of technical/financial resources; 4. Doubts about usefulness: rejection of personal use; 5. Rejection: criticism of content and social impact of the Internet" (Zillien 2008: 18).

The last two points in particular - the lack of appropriateness and undesirable content - are similar to internet and (the older) television non-utilisation patterns. However, this form of non-use has been and continues to be assessed completely differently by society and has also tended to be neglected in research. Apart from a much-cited communication science study on the subject (Sicking 1998), the few other studies can be found in publications, some of which are remote. These include the US Fire Prevention Study from 1975 (Folkman 1975). This study investigated the extent to which media can be used to prevent forest fires. In this study, the author found a differentiation in media usage patterns, which emerged as a difference between radio and television users on the one hand, but also as a difference between education levels on the other. Non-use of radio was thus a phenomenon in less educated households, while non-use of television was definitely associated with higher levels of education:

"Non-use of radio was highest among the lower educational levels; for television, non-use was highest among higher educational levels." (Folkman 1975: 4)

Non-use of television is known to be a conscious decision against the medium. However, this has never been problematised as non-participation/not taking advantage of social opportunities, but rather the non-use of these media is actually an accepted social standard. This has to do with the distinction between entertainment and information, but also with the question of the activity of the audience or reader.

If we now update this idea again to understand non-use not only as part of a rainbow-like spectrum of use (see above), but also as a potential privilege, we are no longer so far removed from the idea that non-use in its pure form hardly exists today anyway. Kirchner (2014), who is investigating the phenomenon of non-utilisation in her (as yet unfinished) dissertation, also supports this thesis. In a podcast on the subject, she talks about the fact that there are no offline users, but - on a much smaller scale - non-users of certain applications. She therefore goes right down to the level of content - and thus arrives at a further differentiation in (non-)usage patterns.

Another study can be interpreted in a similar way (Knapp 2009), in which selective non-use was outlined as a practice of media literacy. Particularly in times of surveillance, non-use is also a progressive form of use - or at least a conscious form of want-not. Overall, such results indicate that one should go into much more detail on the question of what someone does online (and in what form, in what contexts, etc.). This is a slightly different focus to understanding non-use as such: it is more about learning to recognise patterns of (non-)use and being less quick to judge.
Normalisation?

Another indication of the need to change the question is the normalisation that has occurred. There are indications that classic patterns of the knowledge gap in Internet and mobile communication use are indeed continuing, i.e. that some users (who have certain socio-demographic or socio-economic backgrounds) are better able to utilise these forms of communication for themselves than others: "Those already in more privileged positions are reaping the benefits of their time spent online more than users from lower socio-economic backgrounds" (Zillien/Hargittai 2009: 287).

In a large-scale study, van Deursen and van Dijk show that patterns of exclusion or social stratification known from outside media use are now also very clearly reflected in media usage patterns in relation to online use: "The findings suggest that as the Internet becomes more mature, its usage reflects traditional media use in society; Internet use increasingly reflects known social, economic and cultural relationships present in the offline world, including inequalities" (van Deursen/van Dijk 2013: 15). Van Deursen and van Dijk call this a development from a knowledge gap to a usage gap, similar to Kirchner's argument (see above). We therefore see a reflection of social reality in the form of media use, and accordingly a normalisation of use and thus also a reduction in the potential for change. However, Van Deursen and van Dijk also explicitly point out that the inequalities are not simply structured - on the contrary. Very similar to what has already been mentioned above, the authors of this study show that a quantitative increase in use (including frequent use) can also be observed at the less educated level: "in their free time, lower educated individuals use the Internet for longer periods of time than those who are medium and higher educated" (van Deursen/van Dijk 2013: 11). However, the communicative activities that took place online were of an above-average quality that could hardly be translated into broader social skills or other points of contact. This brings us to a further need for differentiation. It is not enough to simply look at whether someone is online or not, but you have to take a closer look at what kind of communicative action is being carried out and how this is also dealt with in the wider social environment, i.e. what benefits are derived from it. The fact that support from the social environment is a basic prerequisite for participation is also shown by Friemel in his study on the online (non-)use of older people (Friemel 2014). This indicates not only usage patterns, but also possible tools for increased participation. This form of non-use (i.e. non-use of central sources of information etc.) in turn indicates reduced social participation.

However, normalisation also means that internet penetration is currently no longer increasing, but that the growth here is only minimal (Frees/Koch 2015). The latest ARD/ZDF
online study indicates that a level has now been reached (around 20 per cent non-usage) from which usage is unlikely to increase significantly. We are actually dealing here with a stabilisation of not being online. However, whether these 20 per cent of non-users are socially excluded (see above) or consist partly of deliberate non-use is a question that has not yet been answered and is also linked to the types of use that online users engage in.

However, what is definitely different - and this finally brings me closer to the question of mobile media - is that there is an increase in Internet use while travelling. In other words, 55 per cent of online users now use the internet on the move to some extent and also have the highest usage intensity there, i.e. they have a higher average usage duration and frequency than those who do not use it on the move (Koch/Frees 2015). Accordingly, the authors of the study also ask whether there may be a mobile divide (ibid.: 382). I would like to take a perhaps seemingly strange turn in terms of content and use another binary distinction to frame these current developments in a different way.
Outsiders vs. insiders?

The distinction between outsiders and insiders is a classic sociological distinction (Merton 1972). Here, the outsider position is usually outlined as problematic, while insiders are described as privileged. However, in a 2008 study by the Institute for Advertising Practitioners on out-of-home advertising - the advertising we encounter every day on the streets or on public transport, for example - this explanation of being an outsider is reversed (Cronin 2008). Outsiders are the target group that advertisers want to reach, because in this definition they are the young people with more money; they are the more active, the aspiring early adopters of new technologies and they are the ones who are actually much more active in the outside world. This means that they are mobile in terms of content and physically and are therefore a clear target group for Out of Home Advertisements. The author initially assumes a structural change in the ratio of time spent at home to time spent outside the home. While on average 6.5 hours a day are spent outside the home, this figure is significantly higher for outsiders (9.6 hours or more a day). Accordingly, according to the study, traditional media use at home is increasingly being reduced to insiders, who in turn spend an average of twelve hours a day at home. In contrast, outsiders have significantly shorter attention spans, as they consume media outside, on the move and increasingly on convergent mobile technologies (Cronin 2008: 100).

What is interesting about this is precisely the reversal of these concepts. The outsiders suddenly become the ones who are attractive to the industry. This is one of the reasons why mobility can be seen as a new form of segmentation. This initially refers only to target group segmentation, but also to a new form of social stratification. This type of mobility - far more than the possession of mobile devices - is far too little considered in research, according to this plea. The following should always be borne in mind: "[M]obility was not invented by the mobile phone" (Cresswell 2012: 646).

I would therefore like to emphasise these two tendencies - and thus challenges to research - which are in principle contradictory (or at least in tension with one another), and at the same time to explain them in more detail using a case study. As a reminder: On the one hand, this is about a better understanding of the micro level, i.e. focussing on the question of what is done online, in which contexts and with which references and motives. On the other hand, I am calling for us to move further away from media use, to broaden our perspective, to consider social contexts and mobility in general in order to be able to understand which (mediatised) communication patterns are reflected here. Both are relevant for the following case study.
Homelessness as a case study

In the following, I would like to turn my attention to homeless people and their use of media. They certainly represent a small marginalised group that tends to live in a social extreme. However, it is precisely this context that reveals a number of things that may be helpful for further research into the digital divide and emphasises the problematic nature of the concept's normative nature. As far as the fundamental issue of access is concerned, Jackie, a 61-year-old homeless woman in a study, describes the problem very succinctly: "So you need to have a cell phone. Most people can't afford it. People go out and pick up cans just so they can have something to eat, but these other things are necessities too" (quoted in Roberson/Nardi 2010: 447). Many homeless people do not have basic media technology or access to it. However, if the need to communicate is seen as a basic need, then the basic equipment (keyword: universal access) must also be provided. And this includes not only technologies, but also other types of infrastructure (electricity, pre-paid cards, etc.). But even if this were the case, the question is whether the fulfilment of the basic need for communication also requires social inclusion?

Studies from the USA or other English-speaking countries (there is hardly any research on this topic in Germany) show that even the existing barriers to access are surprisingly often overcome because, for example, Internet use is highly valued by the homeless (Pollio et al. 2013: 174). This certainly does not apply to all different homeless populations in the same way (and also not in all countries). However, another US study emphasises that most homeless young people use the internet at least once a week (84 percent), i.e. very frequently depending on their circumstances. In such cases, access takes place primarily through and in public places (libraries, youth clubs, etc.). At first glance, the behaviour of homeless young people is not very different from that of their peers because they mainly use social network sites (SNS) to maintain their social networks - both those at home and those on the street (Rice et al. 2010). Another study emphasises that the topics that take centre stage online are also similar to those familiar to young people: "Exploring identity (...) cultivating and exploiting social ties (...) interpersonal tensions, brought online and amplified (...) managing audiences, adjusting profiles (...) shifting affiliations and transitions" (Woelfer/Hendry 2012: 7-9).

However, what is already indicated here is not only the facilitation, but also the problem. If the focus is on everyday organisation and contacts and at the same time the "street-based social network ties" are strengthened, then the everyday patterns of the street can also be found here: people look for food, accommodation or the opportunity to earn money, and security should also be guaranteed (Roberson/Nardi 2010: 446). In a Scottish study, Bure (2005) once rightly summarised this as "digital inclusion without social inclusion". In her research, she found evidence of everyday petty crime that could be better organised with the help of ICT, such as the buying and selling of drugs. Bure sees homelessness as a kind of subculture, which in some cases is simply replicated online. In principle, these users do what we do every day: they organise their everyday lives with the help of these media. Digital inclusion as an automatism for social inclusion is thus questioned (and some of the literature suggests precisely this simplified basic assumption). This in turn ties in with what van Deursen and van Dijk have also pointed out: a reproduction of social inequalities despite increased internet use.

The question therefore remains as to whether digital inclusion is a necessary prerequisite for social inclusion - or whether this also equates (non-)users too indiscriminately. At least that's how we can read the information provided by the Rainbow Spectrum and Want-Nots. At the same time, we now know that insiders and outsiders are no longer the same thing. In a way, however, it is too early to think about the new forms of exclusion thanks to mobile media, because mobility as a social stratification phenomenon has not yet been researched enough. However, a look at one extreme of exclusion - homelessness - initially shows the need to continue thinking about the basic provision of digital media for all citizens.

However, to end on a somewhat more optimistic note, it is worth pointing out the potential of use for homeless people in particular (and this brings us back to the question of content): Because, of course, digital media offer opportunities for forms of privacy, for identity work, which can be very important for homeless people in particular: 'Therefore, Facebook provides privacy and a space where people experiencing homelessness are in control of how they are perceived by other people' (Yost 2013: 25). Especially for people who have little privacy in their everyday lives, who live on the street - constantly being looked at but never seen - this is a great potential. Nevertheless, Roberson and Nardi (2010: 445) also point out: "Many questions remain regarding whether (...) the use and ownership of these technologies put homeless at an advantage or disadvantage."
Communication as a basic need

Riehm and Krings already established in 2006 (Wyatt several years earlier) that there will continue to be a base of non-use (see above) and that it may make more sense a) to focus research on users and b) to continue to consider potential alternatives for non-users, i.e. not to reduce socially relevant communication to the Internet alone.

Furthermore, as the case study has shown, in addition to these basic orientations, it makes sense in certain niches to continue to proactively provide access and promote the use of the Internet, but at the same time not to raise expectations of possible socially relevant uses too high. Convergent media are simply too many media at once: they allow entertainment, everyday organisation, interpersonal communication, political information and participation and much more - but they also tend to be used only for what users consider relevant and necessary for themselves. Accordingly, the question of participation must be posed differently for each user. Everyone must also be allowed to deliberately not use them.