The piercing bell from her phone alerts Marina, a driver for mobility platform Gojek, that she has been matched to a food order. (Marina is a pseudonym.) She has 12 seconds to decide whether to accept it, so she goes through some quick calculations. The pickup location is Grand Indonesia, a mammoth Jakarta mall known for its elite customer base and complex layout. Even though its restaurants rely on motorbike delivery services to keep them ticking, Grand Indonesia has no convenient parking spaces for motorbike taxi drivers.
Luckily, Marina knows some local motorbike taxi groups in the area who could look after her motorbike as she picks up the order. But the route to the customer would take her past Monas, a national monument in the center of Jakarta. A political demonstration has closed off a lot of the streets in the area, according to the news floating around driver WhatsApp groups.
Nevertheless, she decides to accept. En route, she shares her live location and destination information with her driver community’s WhatsApp group—a common practice for online motorbike taxi drivers in Jakarta. Hundreds of highly organized WhatsApp groups form the backbone of the platform driver communities in Jakarta, becoming a space where drivers get advice on their work strategies, ask for help in the event of accidents, and, as in Marina’s case, solicit information about the neighborhoods they are about to venture into for order completion. There, she is informed that the drop-off point is near a zonah merah—a red zone where online platform drivers cannot enter due to long-standing agreements between conventional motorbikes taxi drivers and their “digital” counterparts. She messages the customer on the app requesting that they collect their order 500 feet away from the zonah merah. Grudgingly, the customer agrees, but Marina worries the request may cause the customer to leave her a poor rating.
The multitude of decisions Marina had to make for this order is representative of the limitations of an algorithmic vision of urban space the mobility platform deploys—a flattened, idealized geography where frictions do not exist, only supply and demand. In this world, the former appears to move easily through mapped streets to the latter.
Drivers in Jakarta, though, know better. Over the course of my multiple fieldwork visits to Jakarta between 2019–20, Marina and other Grab and Gojek drivers shared with me their understanding of urban space. It is infused with social relationships and infrastructural hurdles. To do their jobs, they must think every day about which routes have the most potholes and which traffic signals stay red the longest. Their mental maps of the city note what places have unfriendly security, where they might encounter violent traditional motorbike drivers, specific agreements they have to comply by, friendly roadside restaurants that would let them rest. They must compensate for inaccurate geolocations caused by GPS signals blocked by nearby infrastructure.
Much has been written about the frictionless technology of ride-hail platforms celebrated by customers and technologists alike. Startups like Gojek and Grab have become decacorns (companies with valuations of $10 billion or more) on the basis of finally providing a simple technological solution to the chaotic mobility markets of the developing world. Yet their elegance is powered by and relies on the human mediations of the drivers on the street. It is the local markets they claim to replace that have often furnished drivers with the knowledge of local physical and social constraints.
In Jakarta, the seamlessness of the operation of both digitized and nondigitized bike taxi markets depends on particularities of street network morphologies, traffic conditions, and spatial clustering of bike taxis on streets. It is the task of the driver to bring together the two visions of urban space: the abstract and the grounded. Yet the celebration of digitization renders the driver completely invisible—even though it is their knowledge and ingenuities that allow the technologies to appear frictionless. Even as algorithms become more complex, that local, granular knowledge is difficult to replicate. There will always be unknown optimizations algorithms will miss and real-time hurdles that tech firms, no matter how organized, will remain unaware of.
This deployment of techno-solutionism via a contextless algorithm is particularly pernicious when we consider that the assumed view from nowhere is indeed a view from somewhere: Silicon Valley. Even homegrown startups in emerging markets inherit Silicon Valley’s belief in “technological fixes” independent of context. Western biases can then seep into assumptions about mobility in different places. For example, until recently, Google Maps did not discriminate between impassable streets of informal settlements and wide main roads. (I became well acquainted with this particular design flaw when it ended up with me getting someone’s car stuck within the narrow confines of a winding informal settlement.) These gaps eventually have to be resolved by the on-ground workers whose livelihood is suddenly dependent on such incomplete algorithmic visions.
Digital platforms, then, have not removed frictions—they have shifted them on to someone else.