Workshop Report: “Public-Oriented Mobility Through Digital Technology”

By Karol Kurnicki

On 9 March 2021, Karol Kurnicki organised an online workshop on the theme “Public-Oriented Mobility Through Digital Technology”. The workshop was part of Kurnicki’s PUTSPACE Fellowship at the Leibniz-Institute für Länderkunde (IfL).


Kirill Galustov (IfL), Wojciech Kębłowski (PUTSPACE), Meike Levin-Keitel (TU Dortmund), Indra Lukosiene (PUTSPACE), Ibrahim Mubiru (TU Dortmund), Egor Muleev (IfL), Mariya Petrova (IfL), Lela Rekhviashvili (IfL), Wladimir Sgibnev (IfL & PUTSPACE), Louise Sträuli (PUTSPACE), Tonio Weicker (IfL & PUTSPACE)


Along the general trend of platformisation of cities and societies (Poell, Nieborg, van Dijck 2019), Mobility as a Service (MaaS) technologies promise innovations in how transportation is provided, accessed, and experienced by travellers. Their implementation reignites questions about the public nature of transport, while the technologies themselves are presented as one of the main avenues of mobility transformations, necessitated by climate catastrophe and recently in response to the COVID-19 pandemic. Simply put, the future of mobility appears to rely on digitalisation, which uses data to connect old modes of transport with inventions such as autonomous vehicles or drones.

The questions raised by the development of MaaS platforms are not only technological or economic but include a strong social component. What changes will they effect in the provision of transport? How will they affect privacy? Will there be different levels of access depending on geographical location, financial means, or any other criteria? What will be the role of public transport in the digitally mediated system and will it continue to be public at all? Can MaaS platforms, supported by and integrated with similar digital technologies, be a good answer to challenges in transportation?

The workshop intended to delve into some of the issues raised by the emergence of MaaS platforms in cities. Inspired by the approach taken in the PUTSPACE project, which looks at the intersecting areas of public transportation and public space in both spatial and temporal perspective, the workshop was concerned with possible consequences of such technologies for the public nature of mobilities. It aimed to interrogate some of the ways in which new technologies influence social aspects of mobility, for instance differentiation of users by different means, accessibility of transport, or representations of urban spaces.

To critically engage with these issues, the starting point of the workshop was whether it is possible to envisage a MaaS platform that strengthens public values of transportation. Instead of focusing on the critique of existing solutions, whose shortcomings have been already identified (e.g. Pangbourne et al. 2019), the workshop supposed a situation in which the development of digital technology is based on principles of public good, mobility justice and the support of equality in transportation. This approach was reflected in the methodology of the workshop, which focused on the design of different elements of a platform, such as interfaces, data processing, regulations and its part in a wider “MaaS ecosystem” (Kamargianni et al. 2018).

Themes of the Workshop

The introductory presentation recalled the description of digital platforms from the perspective of social sciences, defined by (van Dijck, Poell, de Waal 2018: 9) as a “programmable architecture designed to organize interactions between users (…)”. The argument that the MaaS platforms, just like other technologies of the sort, deliberately design what kind of interactions (and by whom) will take place as a result of their operation, provided a background for the tasks taken up at the workshop.

The social and interactive effects of MaaS are inherently built into the platform and can be discussed and selected during the design process. This relates to the discussions within the PUTSPACE project and beyond around the nature of the public (both in spaces and transportation), which has been always shaped partly by deliberate design decision (Paget-Seekins & Tironi 2016; Korn et al. 2019), such as locations of stops, setup of trams’ interior or physical access to transportation. Currently, one part of this discussion is the fact that in the mediated digital environment the public nature of transport is already reinterpreted through technology and needs to be put under scrutiny by social research specifically in this digital context (Marres 2017).

Accordingly, the reference to Sarah Barns’ definition of a platform as a “highly participatory ecosystems of interaction, shaped by particular software design tactics” (2019: 1) was made to highlight the potential of inclusion and exclusions that platform technologies embody in their design. It was also a basis for the argument that if certain commercially oriented, exclusionary or discriminatory elements of platforms exist, it is possible to use “software tactics” for opposite objectives, for instance to envisage, build and support socially equitable mobility transitions. Therefore, one of the objectives of the workshop was to make use of these “specific but not always visible design tactics” by, first, making them visible and, therefore, open for discussion, and, second, by interrogating the “relational dynamics operating within platform ecosystems, which incorporate code, commerce, and corporeality, and engender and reshape everyday selves and spaces in vital ways.” (Barns 2019: 1).

The role of the workshop was therefore not to build on existing platforms and improve their operations but to think whether it is possible to repurpose the existing technology for the creation of public-oriented mobility that would have transformative potentials. The question whether it is at all possible was therefore open – perhaps the future mobility should not depend on digital platforms at all? This would be in line with some of critiques already voiced around platformisation (Srnicek 2017) as well as MaaS more specifically (Mulley 2017).

Some problems from the user perspective were mentioned early on by workshop participants, who were asked to download and try out one of the applications available in their city of residence (or, alternatively, in Berlin). This experience brought comments about data sharing, when one of the apps required quite detailed personal data at the point of log in. It was also noticed on the example of Berlin that applications do not necessarily include all operators in a city – in this case some of the e-scooters were not available through the integrated system. This shows the exclusive and potentially monopolising character of MaaS as a point of access to urban mobility. Moreover, it was noted that the functionalities of apps do not allow for fine-tuning searches and can imply “standard” users, therefore excluding a wide range of other users who do not fit the automated functionality.

Design Objectives

The themes around platformisation of mobility and components of platforms, which were a focus of discussions in small teams during the workshop, have been approached from a set of design principles proposed to participants and briefly discussed during the introduction. They were the point of departure and the context in which the design exercise unfolded.

These principles were shortly debated among participants, who – while agreeing with the general orientation – raised concerns for instance about the understanding of the term “public”, the tension between enabling and limiting mobility in a systemic way and the responsibility of technology. The first point reflects a more general discussion about values of transportation (e.g. how to define the public and who defines it?) as well as different models of ensuring its public nature, for instance by ownership, regulation, rights of access, or fare policy. Second, there is a growing recognition that increase in mobility of people and goods can itself be a challenge for the environment and planned transformations should include ways of limiting movement. This can be a role of technology but might as well mean that the development of platforms that increase mobility by making it easier should be completely abandoned. So, there is a question mark and a tension in (im)mobility itself. The last point can be represented by the question about “MaaS without digital technology” – perhaps the aim should be to ensure the provision of transportation to travellers is good enough to make the use of digital technologies unnecessary?

With these questions in mind, we moved to design objectives, that is to listing the conditions that the publicly oriented mobility platform needs to fulfil.

The general task formulated for the participants read as follows:

This call was meant to free the imagination and open up thinking about mobility platforms rather than tying the participants to already identified problems with platforms and finding solutions to those. Such a strategy helps participants to take a more general view on technology and pick on many different elements at once. It also allows them to be less restricted by real-life technological and social conditions and highlight – by juxtaposition – the limitations of existing platforms in the area of social responsibility and, in this case, public values and mobility justice. At the same time, the groups were asked to reflect on the structure of platforms and come up with ideas about certain elements that platforms consist of.

The workshop built on the identification of the six areas of “platform anatomy” described by van Dijck, Poell and de Waal (2018): data, algorithms, interface, ownership, business models and user agreement. These were the starting points of the work in the design stage of the workshop, which was conducted in two groups. The groups were asked to provide solutions in three broad areas: front-end, back-end and platform integrations (such as with providers of transportation of services or payment systems). These solutions were meant to make data productive for social good; algorithms strengthen mobility justice; interfaces accessible; price packages that ensure public access to transport; ownership secure democratic values; various providers serving a broad range of modes of transportation inclusive; and finally to make additional elements (labelled as ‘externalities’) which considered how the operating platform would impact on urban spaces and cities in general.

This task was supported by a set of questions in each area, which was however not limited and did not have to be answered in detail.

Results of Group Work: Some Ideas for Mobility Platform Design

The description of the results of the design stage and ideas from both groups will follow the areas described earlier. The results are not a comprehensive blueprint for how to build a public-oriented mobility platform, but a set of ideas and solutions that should be taken into account while implementing one. The description can therefore serve as a resource for designs that stem from public values of transportation in digitally mediated settings. So, what would the socially and environmentally responsible platform look like?


The application should provide options and settings that allow for better inclusion of different users and needs. The example is a person who travels with a bicycle and for whom the automatic selection of routes is not appropriate as s/he might not be allowed to take many means of transport. This example can be extended to different groups of users and their needs, for instance to people with various disabilities, whose travelling can in principle be made easier by the platforms if their specific needs were better recognised in the app. The ability to create selectable and adjustable profiles for individual users is a way to make the mobility provision better.

The issue here is not only the accessibility of the application itself (i.e. its look on a screen of a mobile device) but the flexibility of the transport provision in the platform, which can be increased with the use of the app. This stands in contrast to most of the existing apps, which streamline the process of route selection and automate decision-making. This clear preference for convenience and speed of use has a cost in the form of limiting choice for users with non-standard needs.

The interface can make a better use of the data already in the system and visualise them for the purposes of information. For instance, the hidden costs of mobility can be surfaced as information about pollution or CO² emission generated by modes of transport during journeys. In current systems this opportunity to increase awareness about such environmental effects of travelling is lost. This also pertains to social costs, such as salaries of drivers, train operators and other people employed in the transportation system. This is a way of increasing knowledge about social and environmental issues relating to travelling, which currently is not visible in the app.

The interface of the system must also have a non-digital form to avoid exclusion of people who do not have use mobile devices. This can be organised as a call centre or have some kind of material form, perhaps as information points located throughout the neighbourhoods. The digital exclusion, which is always a danger in mobile technologies, can be thus counteracted by the expansion of different interfaces. It is also possible to imagine access to the platform on three layers: through a simple digital application, telephone service and a more sophisticated, multi-option application.

From this point arises the question if the platform’s main access interface should be primarily an app? It is possible to imagine a platform that is built up from the level of a neighbourhood and therefore community-based and increasing inclusivity from this level. In such a platform, the application on mobile devices is only one of many points of accessing networked transportation.

Plans and Packages

The main thing to decide in this area is the extent of coverage by a standard payment package. In practical terms, the fundamental question is what modes of transportation are subsidised and in what form. This also implies answering questions about what is covered by public expenses and what is considered additional services that has to be paid for by passengers themselves (or higher fares). The objective should always be to include within the standard public package as many passengers and means of transportation as possible.

We can imagine that in a comprehensive system of public transportation, different modes are under the umbrella of a public provider and include, apart from traditional means such as metro, buses and trains, also scooters, shared cars or bicycles. The pricing and provision of packages should be regulated by an accountable public body and aim at limiting the role private vehicles, especially cars, in urban mobility.

Making payments should include the possibility to do it anonymously, for instance by depositing money on a “mobility account” which would not require sharing information about the passenger. Special allowance based on the needs of different passengers can also be included in the payment plans to support social equity and universal access to transportation (lower fares for those who need it and selected categories of passengers).


Data collection must be minimised, collected data should be protected and used strictly for the purposes of improving services and infrastructures. In the socially oriented platform, data is also used for assuring equality of access for different groups.

The algorithms run by the platform must be made more understandable to the public and under democratic control. This can be achieved by the organisation of public forums bringing together, for instance, regulators, passengers, and service providers. The aim of making algorithms public is not primarily to explain technical aspects of data processing within the platform but to open up decision making about how the entire system of transportation works. This would be, in effect, a forum of discussion about principles of operation of digital platforms as well as on public transportation in general.


Apart from ensuring a strong presence of publicly owned means of transport, the system should allow private owners to share their vehicles through the system. This kind of schemes for sharing private cars through a platform already exist and should be included in socially responsible MaaS, thus challenging the strong public-private divide existing in transportation.

Participants in the system, in particular commercial ones, should contribute financially to the operation of the system. In other words, to support the leading role of public provision of services, any entity wanting to offer their services must do so through the publicly controlled platform and support it financially. This goes against existing arrangements, in which companies are often subsidised from public funds to provide a service considered beneficial for society, for instance shared bikes included in the transportation network of a city.

There are also dangers and responsibilities that come with such centralisation of the system. So, oversight mechanisms need to be built in the system and discussed publicly. As a result, perhaps some flexibility should also be provided by the system, whereby smaller operators are included to provide specialised, tailored services to passengers. In such case, standards will have a strong role in ensuring that the system is operational and that it fulfils the role of a comprehensive public transportation well.


This area relates to integrations that platforms initiate and maintain with the outside and the possible consequences of MaaS (even publicly oriented) in a broader context, such as urban spaces, publics and other actors. In fact, part of the responsibility of the platforms should be to consider the possible impacts and design mechanisms that both prevent negative effects and envisage possible benefits.

One broader issue that reflects the discussions about environmental impacts of mobility is the fact that platforms are always built on a logic of expansion and growth, which means increases of mobility are part and parcel of their operation. However, the socially responsible platforms would take into account that mobility has to be in some instances minimised, or at least certain journeys that now result in pollution will have to be switched to active travel such as walking and cycling. The role of community-supported parts of mobility system will be crucial. The operation of the MaaS platform must also consider the environmental effects and include various ways to decrease them as part of its operation.

Possibilities of stronger integration of the entire system with neighbourhood or communal level of urban life were identified as part of the work in this area. They aim to address the tension between the need (and right) to move and necessity of limiting movement to reduce pollution and negative environmental effects. The platform strongly integrated with the areas where it operates has a better chance to be suited to the needs of the inhabitants and to react to existing socio-spatial relations instead of just “disrupting” them by bringing about total restructuring of the mobility systems. Another aspect of neighbourhood integration is a recognition of mobility needs and a possible pooling of them that can guide how transport is organised. To exploit these positive potentials of MaaS platforms, it has to be from the beginning integrated with the needs and resources existing in communities and cities. Therefore, it has to be sensitive to differences and changes within its area of operation.

Bridging interface and externalities, there is a strong argument for the platform to involve three main layers of transportation: public transport/infrastructure; communal and collaborative enterprises (providing mobility services that would not be based on profitability, but rather community based); and lastly commercial services, after a democratic discussion establishing what is essential and has to be covered and what is non-essential (i.e. bringing market oriented solutions only in the selected, controlled areas). This kind of structure spans between the most general public sphere, mid-level of community resources and lower level of preferably small-scale free market activity.

Implementation of MaaS also has to consider the infrastructural aspects of mobility and envision ways in which systemic underpinnings of transport will be maintained and provided. This is an issue related not only to digital infrastructures on which the platform relies but also to the possible impact of MaaS on existing and future infrastructuring of transportation.

“The Best App Is No App”, “Too Much Mobility” and Some Other Challenges for Digital Platforms for Transportation

In the course of the workshop, the groups took different strategies: one followed the “platform anatomy” suggested in the introduction and the other imagined a platform as built for the imagined socialist city. The latter strategy was grounded more decisively in a particular political economy of an imagined socialist city, but both strategies developed their solutions from the question about how a socially oriented public institution can use a digital technology for social transformation. A discussion on a range of issues followed: how to identify and reveal different needs, how to highlight intersectional inequalities that are part and parcel of mobilities, how to overcome them, how to support values of an equal society? This section presents some arguments from the final discussion, without providing definite answers. It is rather a composition of issues that have been identified around MaaS platforms that can be taken up in further discussions.

The main point confirmed by the workshop’s discussion is that the MaaS platform, imagined as a comprehensive system that changes transportation in large cities and regions, must include in its very design mechanisms that counteract inequalities of various sorts: infrastructural, spatial, social or cultural. It also has to be socially responsible and public oriented in two ways: (1) by upholding democratic, inclusive values and transforming them into design decisions and internal mechanisms for operation, and (2) by being responsive and responsible to different levels on which mobility functions: from the most general, to communal and personal.

One of the contentious arguments that came up in response to MaaS platforms at the workshop can be summarised as “the best app is no app”. This means that if the transportation in a city is good, reliable and accessible enough, no platforms is needed. In the imaginary situation when trains come regularly, buses are frequent, cabs can be called on the street, it is easy to find public bikes and walking is safe and pleasant, the need to rely on digital technologies for planning and accessing transport greatly diminishes. So perhaps instead of trying to solve problems of transportation by the introduction of new layer of technology, and highly contested one, cities should simply try to improve their services and deliver transformation in a different way, for instance by designing better neighbourhoods.

At the same time, applications as they already are, can be helpful to different categories of users less familiar with a city. For them, planning and route-finding functionalities are very useful. Even for those who do not have to rely on mapping systems to find their way in a city, there are advantages in using the application, for instance for getting updates on the current state of the system, information about possible delays or planning trips to unusual places. The better the app will be able to recognise the variety of needs, users and uses, the more justified its use will be. There is also an argument related to these deficiencies of apps that advocate for adding more data, developing better algorithms, making the app work better internally, so that the system improves. But maybe sometimes no data is the best kind of data and without platform intermediation the transportation systems can be developed with increased clarity in mind. In other words – do not overcomplicate, just improve what’s already working.

The apps might also be seen as creating the needs they are trying to solve. The obsession with time they bring to the surface, for instance, might not be so important without real-life apps and should actually be counteracted, not strengthened, on the way to sustainable cities and neighbourhoods, not to mention quality of life. In other words, time-saving orientation build into apps and mobility platforms has social or environmental costs and, for instance, it might be better to extend the duration of trips if it means benefits for the society and environment.

Another potentially negative aspect identified during the workshop is the depersonalisation of the city, when it is mediated by a digital technology. Normally the transportation system is one of the main urban interfaces, which shapes experience of people in public urban spaces. When intermediated by a platform, this experience is by definition filtered and not with positive consequences. The use of mobility apps and platforms bring back fundamental questions about how we want to interact with city infrastructures, mobility and each other. While often convenient, platforms also bring the danger of depersonalisation and exclusion as part of their digital character. This is why – as proposed by both groups at the workshop and described above – MaaS platforms must have non-digital interfaces and points of access.

Another set of issues relates to how platforms work as regulatory mechanism. The takeover by platforms of many urban processes might be seen not as a positive opportunity and an answer to climate challenges but rather as a danger of centralisation and corporate colonisation of infrastructures and public services, including transportation. They can in the long term prove counterproductive and their purportedly “disruptive” potential can turn out to inhibit necessary transformations.

In this light, operating a platform might be a way for a municipality to preserve control over the public services, develop cutting edge technology and keep advantage over commercial competitors. But there is another answer to the technological challenge, at least in the area of mobilities: great public transport that does not require an app is also a way of keeping on top of technological competition, which platformisation cannot undermine.

The underlying question is again about values. As democratic deliberation and public responsibilities of any technology and infrastructure are necessary, the issue is how to put them on the forefront of any innovation, both digital and not.

Imagining mobilities outside of capitalist system is a completely new sphere of ideas and possibilities. In such circumstances speed, punctuality, and efficiency are not necessarily the main criteria of travel, in which “use value” become more important – for instance, the general satisfaction with travel or meeting others along the way increase in importance.


The platform cannot be built outside of the conditions already existing in society. And yet many of those, which are actually in operation or preparation, seem to completely disregard territorial differences and social, cultural, or environmental values that are present in areas where they operate and intervene.

Apps and platforms bring normativities: in interfaces, in time calculations (e.g. assuming probably male, able bodied, often white person), they flatten differences, they are “rational” but do not correspond to actual experience across different groups and sections of society. Platforms not only grow from particular social conditions but also bring with them a particular vision of how societies should function. They organise interactions and relations between various human and non-human elements and as such imply a vision for the organisation of the world.

The workshop proved to be a very good exercise in identifying, in a critical way, the potential social issues that arise when a complex digital technology is introduced in already contested area of transportation and mobilities. It was a reminder that digital interventions must not overlook social, cultural and environmental aspects that will be involved in them. This especially matters for transportation, as it is already operating between the public and the private and is enmeshed with a broad range of issues around public spaces, urban planning, social and spatial justice, sustainable transitions and future of cities.


Barns, S. 2019. “Negotiating the platform pivot: From participatory digital ecosystems to infrastructures of everyday life”. Geography Compass 13 (9), 1 – 13. doi: 10.1111/gec3.12464

Kamargianni, M., Matyas, M., Li, W., Muscat, J., Yfantis, L., 2018. The MaaS Dictionary. MaaSLab, Energy Institute, University College London. Available at: www.maaslab.org (downloaded 10.08.2020).

Korn, M., Reißmann, W., Röhl, T. and Sittler, D. eds. 2019. Infrastructuring Publics. Wiesbaden: Springer.

Marres, N. 2017. Digital Sociology. The Reinvention of Social Research. London, New York: Polity Press.

Mulley, C. 2017. “Mobility as a Services (MaaS) – does it have critical mass?” Transport Reviews, 37:3, 247–251. doi: 10.1080/01441647.2017.1280932

Paget-Seekins, Laurel and Manuel Tironi. 2016. “The Publicness of Public Transport: The Changing Nature of Public Transport in Latin American Cities”. Transport Policy 49: 176-183.

Pangbourne, K., Mladenović M. N., Stead, D., Milakis, D. 2019. “Questioning mobility as a service: Unanticipated implications for society and governance.” Transportation Research Part A: Policy and Practice, January 2019, 1 – 15. doi: 10.1016/j.tra.2019.09.033.

Poell, T., Nieborg, D. & van Dijck, J. 2019. “Platformisation.” Internet Policy Review, 8(4). DOI: 10.14763/2019.4.1425.

Srnicek, N. 2017. Platform Capitalism. Cambridge, Malden: Polity Press.

van Dijck J., Poell T., de Waal M. 2018. The Platform Society. Public Values in Connective World. Oxford: Oxford University Press.