Is design dead — again ?
How the emergence of participating peers impacts design praxis & the prospect of ‘agency design’.
People are the problem.
14 years ago, I was doing my internship at Paris public transport company, quite an engineer minded industry. I can clearly recall a discussion between a in-house designer and an engineer, the later explaining that trains were running smoothly, trafic information was accurate, automated doors right on time, escalators ready, everything was running like a giant well-lubricated machine. The only sand grain that was disturbing the whole system were people. They would run, hold back the doors, leave trash, etc, and the full system ends up delayed, if not stuck. The designer just asked back : “but then, what’s the value of that huge perfect machine if not for people to use it daily ?”
Eventually, both of them were right : yes, that transport system has not much value if only carrying air around, and yes, people do not always behave according to the rules, thus ruining the collective effort required to keep the system running optimally.
Even in countries like Japan not everyone play by the rules. And hopefully.
Since a decade or so, quite a few industries started to see people as problem. Not only because people would have mis-used a product or mis-behaved in a service, but rather because these amateurs successfully created by themselves and delivered services, knowledge or products, that compete with those delivered by so-called professionals. Think about the news and music industry, cartography and knowledge production, intercity transportation, hotel and travel industries, or goods delivery, to name a few. And envision what may happen in the coming years in banking or the energy sector while the blockchain technologies evolves.
If these people in the form of amateurs or makers have become a problem for quite some traditional industries — and an opportunity for all business challengers — it is because they did not stand by the role they were intended to, being users, if not simple consumers. People’s creativity and abilities to contribute have often been outlooked, if not down-played, stating that only experts knew what was good for them. But at a time where decentralised technologies turn organisational costs so low, it has become a great opportunity to build people’s capacity. (Amartya Sen)
Looking back at what it means to design practice, it has evolved from industrial design, when designers first worked for industries targeting consumers and helped corporations to increase sales. Design was about aesthetics, branding, style, built around usability, desirability, feasibility. Then came the moment when designers’ job was not only about the point of sales, but rather about the lifetime of use. They targeted users, and worked on user research, user experience, service design, thinking about gamification, learnability or stickiness for instance.
But once we acknowledge that “the new rule is that if you are a participant, you are, by default, a moderator, a curator and an editor for others.” (E. Kilpi), design practitioners may develop new tools and relevant methodologies to work with the abilities of participants to produce rich and accurate content, share knowledge, answer people questions or contribute to services. As an example, collaborative economy platforms enable end users to take part and develop hundreds of roles, as sharers, contributors, connectors, promoters, curators, co-designers, drivers, hosts, money lenders, makers, etc. and therefore do increase participants’ agency.
In other words, from consumer, to then users, they — most of us actually — behave as agents.
But then, do agents need designers ? Or how to design for agency
Facing the disapearance of the ‘user’ will impact the way we think and do design today. But similarly to what happened when moving from consumer to user, when designers started to pair up with ethnographers and ergonoms, undersatnding how system of agents works will require to bridge with new disciplines and develop the approrpiate tools and methodologies to reinvent design.
Design is a “fast-changing discipline, showing high responsiveness to societal, market and economic transformations”. I highly agree with Gjoko Muratovski’s words — the chief editor of the Journal of Design, Business and Society — and when design as we know it dies, it is to better evolve.
As a source of inspiration, I could quote Murtovski once more. According to him, one of the most important characteristics of design is perhaps “the ability to build capacity by ‘empowering people’”. So designing a car would less be about designing an object to be owned than “a tool to empower people to move”. Or an open source software “a tool to empower people to rewrite or improve the code”. That driver or that coder then may been seen as capacities that ‘agency designers’ could work with.
Agents may be understood as all “knowledgeable touchpoint”, interacting with each others within a specific “environment”. Among these agents account professionals as much as amateurs or random users. Also shall be taken into account within their direct environment humans and non-humans as well, such as eco-systems (think about bees or trees in the food production process) or algorithms (think about Uber algorithmic management of car drivers).
Following that systemic approach, designers may assess an their brief differently. It is often too quickly assumed that people are the problem and design is the solution. But adding a new object or pouring a new service within the system may not be the only way to fix the system. If the system we are part of is limiting our agency instead of fuelling it, it means that we shall look at ways to rebalance that system we are part of. And eventually, “understanding how to act to change the systems we’re in is arguably the biggest meta-challenge of our age.” Dr Dan Lockton.
From what I have read so far, this agent thinking may apply at many different scales, such as work or learning environments, or project-driven collectives, local communities or smart city crowds for instance. At each level, from micro to macro, some form of agency is at stake, and design methodologies have already started to blend with capacity building activities, such as teaching, knowledge sharing, human ressources, policy making, in order to enhance participants’ agency.
So far, here are a few hypothesis agency design has to deal with :
- Capacities assessment : map capacity building and interactions among stakeholders and their environment.
- Feedback loops modelling : model the system wired by feedback loops to identify the system potential trajectories.
- Targeting emergences : build upon expected emergent properties and set up the system and its initial conditions to make these patterns ‘happen’.
1. Capacities are everywhere
Capacities may have the shape of tools, like a smartphone, or a fablab space, or algorithms, compute power, money, knowledge distribution, language, rights or even regulations allowing non-professionals to take part in the labor work force without having to comply to heavy regulations for instance.
From this rough list, capacities appear to lay within the agent’s intricate abilities, its know-how and ability to act, as well as among its direct environment.
But remember that capacities could also reside in void or vacancy. As an example, look at the empty parts embedded in Alejandro Aravena’s Quinta Monroy house design ; they allow inhabitants to update and customise their house through time and according to their needs and abilities.
Capacities may also reside in participants who ignore themselves. For instance, SOSH — a Orange network provider intended for Millennials that only exists online, with no touchpoint in the physical world such as shop or call center — teams in charge of the web forums realised that around 80% of queries were replied online by non-team members, and not even SOSH customers. Those people had some knowledge and felt like sharing it, but never thought of the value it may have for the overall customer relationship system. After acknowledging that, SOSH teams decided in a clever move to make sure those amateur helpers had all the required knowledge to be able to deliver an accurate information, while these amateurs would have never asked for it.
From a private sector stand point, it means that users or amateurs (if no pro-am) may be offered a ‘seat’ next to company employees, blaring the frontier of where company starts or ends. For instance, when Oxylane launched Open Oxylane, it meant assessing amateurs’ ideas and putting some effort to collectively building up solutions with them. That company employees became a capacity for amateurs-with-ideas to develop a new product, all together.
An other example could be the shiny pianos that SNCF put in major train stations for passengers to play freely. By providing this tools for free, SNCF enabled passengers-with-piano-playing-abilities to entertain themselves while waiting, and offered for other passengers free improvised piano concertos. By deploying capacities, SNCF turned its passive passengers into active quality-time-providers.
To build upon Saarinen’s famous recommandation to designers — one should always design a chair in a room, a room in a house, a house in an environment, an environment in a city plan — relevant design shall assess the systems among which it will be part of, and even take into account all participants abilities at each environment scales, of the room, the house, the environment or the city.
2. Systems have trajectories
As complex system theories would state it, agents are always « nested » within a larger system. It means that the agent-environment relationship is an active one ; an agent is responsive to its environment as much as the environment is responsive to agents’ activities. This environment shall be understood through several lenses, such as the direct physical surrounding, as well as the legal and cultural background operating there, or the technical milieu for instance.
For instance, at WoMa — a makerspace I co-founded in Paris — some people come over for co-working, others come for the workshop facilities, or for the community of skills, some companies come to hold events there, or get some initiation about fab. All of these ‘users’ come to the same space but not for the same reason. The space keeps on welcoming their activities in order to promote the creativity of WoMa, may it be by exhibiting one architect’s model or a student’s prototype, or throw a party for the community. But as we did so, we had to remain aware that the workshop activity or events are also producing annoyance for coworkers’ activity, while the coworkers don’t bring as much money in, but create an access to the community skills and know-how. All of the clients, the users, the community and the facility are co-evolving as a system.
In early 2016, we had yet an other agent coming into the picture : contributors, people from the community who would give time to clients and get free access to working environment in exchange. The question we had to answer was to know whether contributors were supporting or annoying the current system ? How would the system and the environment response to them ?
In order to answer that question, I built an agent-based model to assess the WoMa community & WoMa space co-evolutions. The learnings turned out to be not much about the physical space (what space to allow them) nor the time scope (what moments in the week are free for contributors), but rather on the rules level. What it means, from a designer’s perspective, in this context of agents-environment-responsiveness, sometimes it may be less relevant to work on the environment to make it easy for people to fit in (spacial and time organisation), but rather to design capacities to increase people’s fitness to that responsive environment (knowledge, know-how, rules, …).
So, within an agents’ environment, all stakeholders — may they be human or non-human — are interacting through feedback loops. Some of these feedback loops are fuelling some system dynamics, while some are inhibiting them. The overall outcome being the system evolution through time — its trajectory — this evolution affecting all nested elements, including agents. Like for the WoMa example, it means that the question of spatial co-existence — think about the issue of co-existing populations at the scale of urban space for instance — is rather a question of the co-evolution of people and their built and legal environment. It is a dynamic view of the same issue.
I recently came across the Evolutionary Learning Laboratory framework, from Ockie Bosch (Adelaide University) and Takahashi Maeno (Keio University). That method brings that whole idea further and from the interactions among stakeholders deduces the parameters upon which to build a “social algorithm”.
And following that idea of social algorithm, I would recommend to read Deborah M. Gordon’ works about collective wisdom and ants colonies. “130 million years of ant evolution” she writes “have produced many useful algorithms that humans have not yet thought of and that could help us figure out ways to organise data networks using simple interactions involving minimal information”.
3. Properties are emergent
When speaking about an object, a service or a platform, we often tend to highlight its features and inner properties. Designers, like engineers, are matter oriented specialists. But when we speak about capacities and network, we then also have to work with emergent properties.
The most common way to explain emergent properties is the use case of swarms. If one were to code the trajectory of each birds to create the complex evolving shapes birds end up doing while flying together, it would be almost impossible. Instead, by implementing 3 simple rules in birds behaviours — keep flying next to your pals, escape forward if you are followed, don’t collide — you soon will see emerging the amazing swarm effects. These patterns ‘happen’ as an emergent property from a set of initial conditions.
Playing with that swarm effect, engineering and architectural schools conduct research with beaver-like robots interacting together to forage or build structures. Each time they would run the protocole, a slightly different and unique output would come out of robots interactions. The same thinking could apply at the scale of bacterias, if you are looking to build nano-structures or work on medical protocoles. Or at the scale of a city district, the HyperVoisin initiative in Paris aims at increasing connections among neighbours in order to test what could come out of these neighbourhood boosted interactions.
From the study of collective behaviours and swarms, it has long been proven that ‘wisdom’ and other valuable outcomes could emerge from free agents’ interactions. As corporations and institutions aim at playing with these behavioural topic, it is up to designers and architects to step in and find ways to “craft” outcomes by implementing the proper rules or setting up the correct environment.
Intuitively, I think it may lead us to consider that the « smart cities » may only ‘happen’ if there are first smart communities, or smart co-working behaviours, or even smart agents (bottom-up and not top-down thinking). In order to make sure we avoid turning service design into serf-design, we have to design capacities for agents to swarm, not the end product only to be ‘used’.
At last, some design questions
Capacity building, system trajectories and emergent properties are the first structural questions when trying to design for agency. But that remains a broad field of research — the Carnegie Mellon University, Keio University and many more already work on these issues for sure — and it quickly opens up upon other questions, such as :
- How to collect data and model the existing situation ? As Bryan Boyer from the Helsinki Aalto Design University explains, user centred tools are good to improve life of individuals, but don’t necessarily scale when we think of the societal level or urban level (at that scale, systems are not human centred only). The ELL framework or the WoMa Netlogo experiment are two examples out of many more approaches to be developed.
- How to trigger enduser’s agency ? Think about endusers, how to make them realise they have some abilities, and could benefit from using them like amateurs, or students. There apply the reflexion about how to develop participants’ literacy and feeling of ownership over the system they are part of.
- How to prototype and test ? Agents are a human material, and the test protocols maybe would learn from medical experimental protocols.
- How to implement and monitor ? Or how sustainable is a system with no single leader ? How to evaluate power distribution and monitor knowledge distribution ? When implementing social algorithm, one has to assess different types of efficiencies and created values (not only production capacities).
- Who owns agency and/or its by-product ? Are the system and its outcomes a common ? How city dwellers produces self-organised living spaces is well documented, specially around the formation of slums. But for external observers, because slums communities produce culture and space ‘organically’ (may it be as various as circulation, energy water networks or even education, music or lifestyle), somehow the by-product of their interaction is not seen as ‘belonging to them’ (T. Williamson).
- What does it means for an ‘agent system’ to grow ? By spreading out ? By growing its connections only ? By becoming denser ? What are the different types of growths?
- What is then the role(s) of the designer ? If agents are a new material to design with or design from, what is designers’ role next to human ressources, teachers, psychologists, etc. ?
Many things have already been written about agency ; some interesting lectures here, here, here, here, here, here or here. If I were to run for a PhD in design today, this is definitely the topic I would love to spent time on researching. What about you ?
On the 5th of July 2017, the OuiShareFest will be an opportunity to share some thoughts about “how to design for agency” ; after this OSF.Paris, the video will be made available for sure.
