 Welcome to this talk, all of you. My name is Senua Sandberg, and I'm a PhD student at the IT University of Copenhagen, where I'm a member of the Technologies and Practice Research Group. And my work is on the sort of social infrastructure of digital payments, as some might call it, which is looking at the people and networks that enable telecommunication. So I look at use of technology and how technology affects social economic relationships. This talk is pointing to three things, very briefly. Cover some findings from a recent research project on digital payments in Indonesia. So find out if you want to share more about it. So for context, at DevCon in Copenhagen, the blockchain community, we talk a lot about onboarding users, finding use cases. I don't keep a long scripto. And ideally, I need users who really know that they're using crypto at the end of the day. And so this talk focuses on commercial platforms or a particular proprietary platform and looks into how users are actually interacting with this mainstream digital wallet. Specifically, digital wallet that emerged relatively recently. And specifically, I'm interested in the relationships between intermediaries that emerged in that space and the potential peers of what we refer to as the period of payments. So my research took place in Indonesia, which has a population of over 260 million people and currently with the newly re-elected president, a very specific profile about increasing digitalization or what we refer to as Indonesia 4.0. Bring any bells or sound familiar? Yes. So there's this growing emphasis on e-commerce and digital platforms. So what did the digital payment look like in Indonesia? Well, fragmented infrastructures for connectivity and deregulation of telecommunications industry has led to a really rapid uptake of mobile phones because they're both more convenient and more affordable than traditional landlines. 80% of internet users are accessing internet by using their smartphones or tablets, which is kind of a problem out of things since even though many people have mobile phones, less than half have a smartphone, which tells you something about who is not accessing internet. Many of those who do access internet are also doing so with allocated gigabytes that are mainly targeting social media. So for example, Facebook users, so it doesn't come out showing that many people who are using Facebook don't even realize they're using the internet, which is something to think about when we talk about things like Libra coming out. And an important insight into that is that many of my friends pointed out that even though they can explain to me in detail why mobile payments are awesome, they don't actually use them because smartphones are such low-quality smartphones they simply don't have enough space for more than one payment app. And they would rush by the prioritizes social media apps and all the photographs they need for Instagram. So infrastructures of payment are also quite fragmented and have a huge lack of interoperability. So here in a regular supermarket you see seven different card readers. Five are from five different banks, two of them are from mobile-based payments. Less than half of the Indonesian population has access to formal financial services and that includes infrastructures such as ATMs and the ability to actually get cash in your local community. And of those, only an estimated 11% of account holders have debit cards. So systems like this are still for the relatively few. Low access to financial services and increasing use of smartphones means that digital payments bridge really important infrastructural gap for many people. So this is why we've seen this rapid growth of digital payments in Indonesia and particularly this app, Gojek, which is Indonesia's first unicorn for which they've been in part think Google's reputed one million dollar in Indonesia. So apps like these allow users to top up their infrastructure with cash at a designated agent and then they can use that app to make a digital transaction in a specific situation. So what many of us would use a debit card for. These bridges important infrastructural gap and provides sort of proxy banking or banking light feature in people's lives, something that you can upgrade with to get more features such as, for example, if you need to cash out again or send money to your friends. So this particular app sort of has a smear-caling service like a call-in and someone will come and pick you up and in 2015 formalizes an app. And in 2017, it developed this built-in wallet which now bridges an entire ecosystem of apps that range from go-car, to go-food, to go-massage, all of which you pay inside the app using their wallet. And the wallet, of course, is called GoPay, which is so real clear. So where am I coming out of this? Well, I've heard said a lot of times here that technology is not a mutual artifact. Reshape it and it shaped us. And the aspirations of technologies creators tend to be implementing gaps for other technologies that we make. So forms of money also convey meanings and they tell us how we make our communities. And this, I think, is a pretty familiar thought or ethic when we talk in the Ethereum community. It also means that nodes of mobile money transactions cannot be considered as isolating individuals. Decisions we make about our purchasing and our money behavior are not context-free. They are affected by external influences. So what does Gojek tell us about the communities we live in or the ones that they imagine us to live in? This is stills from an ad that was used in 2016. It's an advertisement that was made in the context of Ramadan. So it starts with a young girl in Hamam, Hamam is a single mother and she's working for this app. There's multiple clips, she's driving to and from, oh, she's left her daughter in the morning and said, I'll see you when we break the fast. And she gets her little package. So she's driving around all day, it's really hot, passengers are cranky, they're rude. And then the final passenger is taking food from the orphanage. Whereas the driver watches the happy children, she smiles. Meanwhile, her daughter opens the gift alone. And inside there's a card from a mother explaining that she can't be home to break the fast because she has to make sure everyone else gets home in time to break the fast. Finally, as she breaks the fast alone inside her road with a glass of water, she starts to receive messages through the app from happy customers, thanking them for her hard work and giving her a lot of ratings. So the app presents two categories of intended users. There's the service provider and the service user. Secondly, the company itself is hardly present. The role is intermediary only appearing with the phones at the very end of advertising. Emphasis is on users helping each other out through the sense of social responsibility rewarded by their thankful peers. The app presents a specific social understanding of the idealized user, performing centralized social infrastructural roles while minimizing the role of the company in governing those interactions within the app. These interactions are often being referred to increasingly as being peer-to-peer, or peer-to-peer. So I would argue that two of these contexts is much more than just a directional message. It's not just from peer-to-peer. It's also a placeholder for this intermediary company. And I would argue that the P's actually become a really inadequate signifier for different types of user categories. So originally a technical term, Peter P described a computer sharing files without a central server. Now it's increasingly seen use in gaining traction in a wide range of social economic activity, remittances, lending, blockchain, and of course digital wallets using proprietary infrastructure, governed in fact by central intermediaries, such as Gojek. In a social context, when we use the word peer, we associate it with some sense of social economic quality, something that the ad, for example, leans heavily into. And this, I would argue, is not really an accurate representation of the relationship between those peers that this app is trying to connect. So maybe more like this. The social economic, so yeah, finally, social economic, maybe the two in the P2P represents the relationship that the peer has to the company. So not only do they have a relationship to each other, but they also have a relationship to the two. So when I asked drivers in Indonesia about their relationship to the company, the drivers would explain to me that they are META leading partners. And during the focus group, one of the drivers looked at the same, but we've never met, just to be clear. And this relationship between driver and intermediaries is not in fact the equal partnership as they like to play with their publication material. For the company, it is the transactions made by the customer that are profitable. And thus, the drivers become tools to provide customers with these services. So in contrast to the idealized advertisement, an examination of the app itself reveals how specific mechanisms and interfaces, in fact, deliberately skew the relationship between the potential peers towards a configured inequality. There is incentivizing drivers to maximize tips in the first one. And these incentive tiers where they can achieve bonuses are called bonuses, but are in fact an important element of the daily income, otherwise it's going to be a sustainable job. So it's completely compulsory to reach them every day. Ashes are great drivers. So what is the question of milliliters from my thumb to slide upon can be the difference between continuing the job and being banned from the app if you, for example, get a one-star rating. The app also sets roots in price. And because of the mechanisms employed, they have few seconds to decide whether or not to accept them. So drivers don't really have a chance to make an active decision about it. In one case, one person pointed out, because my smartphone is so crap, this interface goes beyond where I can even see where the root is going. So I have to click before I know where this is going or how much I'm going to make on it. Finally, cancellations. Drivers can maximum cancel 25% of orders or they'll lose their daily bonus. This means that drivers typically will ask customer to cancel if they really don't want to take the trip. Just fine. The app that introduced a new button going, hey, did the driver ask you to cancel when you're trying to cancel? That, and it turns out, of course, if you click it, they'll be suspended. It doesn't tell you that, but they will be suspended. They later on, if they've all died even further to this entire list of pictograms, which say everything from, I can't reach the driver, the driver's really far away. I've, you know, a whole variety of options for you to cancel with. And basically, I'm mindful of the consequences between you and the driver. So not only does the user in the term PDP conceal these embedded and they quality, but it obfuscates the role of the two in defining the terms of the relationship between these user groups. So, it's important to stress that users of an app are seldom passive consumers. There are many examples of resistance to these advantages. I'll tell you a brief story about one of my opponents. His brother died and he went back to visit his village to be a participant in the funeral arrangements. When he came back to Georgia, where my research took place, he turned on his app and he found that he wasn't receiving any orders. And after riding around an entire day and receiving the orders, he called customer service and they told him there's no problem. The system's working just fine, please carry on. And after about 20 days of this, he finally gave up and sold his account for all of 8,500 dollars. This was a pretty cheap account because at the moment he wasn't receiving any orders, but also it was in his dead brother's name, which means he was able to change the affiliate and bank account and he also had to sell his ATM card to go with it so that the person would be able to extract digital earnings. Months later, he recounted to me that the buyer had gotten in touch with him to let him know that he'd finally managed to resuscitate the account, meaning it was finally receiving orders again. The buyer had given the account for the call of therapy and gradually nursed it back to health. So account therapy is this term used by drivers to describe a variety of practices deployed to train the algorithms that govern order distribution on the platform. So therapy is a form of intervention, but it's also continuing practice since maintaining the vitality of their accounts becomes a necessary part of the job when your job is government algorithmically. So aside from the relationship brokered by the app, customers and drivers work together to co-op the platform in this sort of informal arrangement. They make trips offline, so the drivers can avoid the 20% cut that the app takes on the trips. They make additional bookings so drivers can reach that last point here that they need to get their daily bonus. They buy drinks for drivers when drivers are waiting to take the order, to get their order, because drivers have to sit in line while the food's being prepared, so often they just come with like a, buy yourself a coffee. And this kind of an interesting enactment that was peer-to-peer that doesn't materialize on the app itself. So these are all strategies to make surviving within these proprietary platforms viable, maintaining a kind of flexibility within this imposed digital rigidity. Beyond the fact that there is this sort of inherent inequality here, there's also, it's also important to stress that these two peer groups are not monolithic categories. So not all peers on either side are equal. There are multiple inequalities within those groups as well. So not all drivers have equal conditions for participation, maybe lacking documentation or bank account. So many of them rely on third parties called vendors. So vendors are a semi-formal agreement with the app company, and they allow drivers that are unable to register with the main platform for various reasons to get access to an online account. And one vendor like this can have up to 1,000 drivers, and they take a cut of between five to 30%, ranging on depending on what kind of evidence. They also store all driver data in an elaborate spreadsheet. And then at the end of the day, manually make a bank transfer from that account to the driver's bank account. Or alternatively, if you don't have a bank account, you can make a column, and then the cash will be there in three hours when the company is up. Aside from this, drivers are often members of communities or colonies which facilitate communication about how to access accounts. So this is the biggest community in Georgia, which has over 1,000 members. And here you came for your membership sticker, and sometimes it looks like money for different development projects like building a space camp, or for members in need or struggling, for whatever reason. They broker deals between driver services, such as power repairs, that are going to get a deal on how to access these kinds of services. They show up in person to help out if they have accidents by communicating extensively through WhatsApp communication groups. They also do outreach if passengers have got into cars because they're really afraid of getting bad reviews. For example, being accused of stealing something. And of course, they're a market place for a countering, even though they're woe to a bidder. And this particular head of this community even talked to me about making their own payment system when they eventually upgraded their stickers to membership cards so that they'd be able to connect all of these services that they now organize for themselves. The picture gets muddy even further on the other side. So also people who are using the services and the other apps are not equal. So people who don't have smartphones or for whatever other reason are not accessing these accounts or whether they're in a family or friends and strangers to provide access to the platform. As mentioned, not everyone has access to the infrastructure so I've talked to multiple people who don't have a Wi-Fi at home. So they will rely on going to public access points for getting Wi-Fi, so they can make bookings but also to charging stations, such as this one. And many of my informants share their phones, share their accounts, their cards, pins, and many of the other resources that they're using these apps to bring access. And an example happens to me myself when I was trying to register an integration I just changed my law about when I could get a SIM card. So I found myself in a really frustrating situation of not being able to get a SIM card because I only had a passport with me that was foreign. No problem. This is the guy I found her. I'll register in my name. I'll be dealing. Okay, so I got a SIM card registered in the strangers family card ID because they can have up to three, which is great for kids because that's a lot of SIM cards you can register. One and a half months later, my access completely denied because apparently he found someone else that he wanted to give that number to. So, both service users and service providers rely on agents to facilitate caching in and caching out. So agents, there are several ways to caching out and you can make a transfer in ATM if you've connected your account to the bank. You can go to local mini-mark and pay a fee and alternatively the driver fleets themselves operate as agents that sell the digital balance for cash. And money there ends up circulating according to the incentive streams that the apps provide. So in one model, Gojek, drivers will get if they sell 100,000 rupees, which means they have to have minimum 200,000 rupees available each day to make the point viable. This means that if you ask for less than that, they're probably not going to sell it to you. In the other system, they are given a 20% cash back on whatever they sell you. Many drives are going to be really quick to be like, hey, would you like 100,000? In which case you should usually say yes. One driver also showed me how he had managed to connect his bank account with his app using what he called his digital token. And basically what he had done was enable himself to always have access to money on his phone, meaning he had converted himself into sort of a mobile digital ATM for users, but also he had converted his customers into cash ATMs for himself because drivers need cash to pay for fuel and for food and all the other things that these providers perhaps don't actually give them access to. So all of this and piece-to-piece transactions. On the surface, digital transactions end up seeing some streamlined or clean the messiness of a cash exchange replaced and confined to the affordances of the user interface and the terms and conditions set by the companies providing the platform. However, the exchange of digital money is about far more than the financial transaction itself. It contains much more than what we summarized when helling it back then. And this raises questions about how all of that current messiness will fit into seeing these ideas about immutable ledgers, disintermediation, and trustlessness, all of which are heavily part of this. To summarize everything, because I think it's time to conclude, a few key takeaways. Intermediaries configure the relationships between users. So configure for a peer, but not the opposite. Access is not a binary nor a static category. So contexts are not universal. So you have to consider the lived experiences of the people you're willing for. People will find ways to access the things that they need, however many intermediaries it's gonna take. So success should be defined in terms of people's actual use, not the designer's intent. So technologies that don't interface with existing infrastructure will be repurposed and appropriated to meet the real needs of users to address the problems that they themselves define. And finally, given all of that, the thing that would be most critical for the people in this case study, support the flexible transition between digital and tangible forms of mind rather than necessarily talking about the illumination of cash or, yeah, the idea of living in a cashless society. That's much better than this. This was a very brief presentation of what I've been working on. Please come and talk to me if you thought it was interesting. Thank you.