A Cloud Atlas for Eindhoven, and believing in society over unicorns

Strijp-S and Woensel Noord, top-down and bottom-up

This 2017 essay is the original edit of the recently published article in the collection Becoming a Smart Society (Het Nieuwe Instuut, 2018) available in Dutch and English. Further context below.

Throughout 2015–17, I was part of a project organised by Het Nieuwe Instituut in the Netherlands, called The State of Eindhoven, constructively ‘tracking’, addressing and interacting with the various smart cities/smart society projects in Eindhoven. Het Nieuwe Instituut (HNI) kindly asked me to be part of their active ‘advisory group’, leading workshops and projects for them, alongside the rest of the group, led by HNI’s Linda Vlassenrood, Klaas Kuitenbrouwe and Ellen Zoete, and which included Anab Jain, Albert Jan Kruiter, Chris Sigaloff, Tsjalling Swierstra and Evelien Tonkens, as well as various key local municipality stakeholders, and others from the wider city. It was a pleasure to be part of such a great team, as well as get under the skin of Eindhoven a little, a city I hugely enjoyed discovering and learned a lot from—and congratulations to Linda, Klaas and Ellen for delivering an array of meaningful, progressive projects, many of which are detailed at the The State of Eindhoven website, and in the publication. Below, an essay commissioned as part of the project, which includes a reflection on the Cloud Atlas workshops of late 2016, and associated ideas, my team and I coordinated in Eindhoven. It was initially written in 2017, after the Cloud Atlas workshop, though I’ve updated and reframed much of it in recent months.


A Cloud Atlas for Eindhoven

The smart city hasn’t happened—except where it has

A decade in, the idea of smart cities continues to be endorsed and promulgated by urban, national and transnational governments, and of course technology companies, for perhaps obvious reasons. Living labs, trials, demonstrators, pilots and, increasingly, actual projects abound. Whether they have had any true impact is another matter. Paraphrasing the famous aphorism of British architect Cedric Price, if smart cities demonstrators are the answer, what was the question?

Even now, in this apparently technology-driven age, and half a century after Price uttered his original phrase, we rarely address how technology truly shapes our cities, seldom explore carefully enough the impact of technology on our urbanism, on our urban economies, on the way we live, work and play in cities.

Partly this may be due to a form of prejudice about the lenses through which we view urban development. Asked by the EC about smart cities, Rem Koolhaas grumbled, “The city used to be the domain of the architect.”

Sort of, Rem. Of course, the city is the domain of the citizen. Through another lens, politicians and property developers have made the city their domain all too easily on occasion. But assuming we’re talking in terms of design, architects and planners have of course helped direct and shape cities, but apart from a few obvious exceptions, can we really say that the city has been “their domain” any more than technologists, engineers and inventors? Technology fundamentally shapes cities, and always has, whether that’s the elevator safety mechanism and the flushing toilet adding up to skyscrapers, or air conditioning and the automobile encouraging urban sprawl – these, despite planning and architecture, which tend to respond to those inventions by draping new forms of building around them, rather than create them.

As an aside, it may be far more beneficial for cities, and more importantly their citizens, if architecture did lead rather more of this invention; if indeed it had sufficient gumption to reinvent its business model to buy time for proper research and development, for prototyping, for absorbing the genuinely user-led design and development practices of other design disciplines. Then at least there is a chance that the broader ethical dimension of architecture, its unique quality compared to most of those other design disciplines — and certainly the tech sector — might ensure that new technologies are framed with the idea of the city as a public good. In the Netherlands particularly, where soggy geography requires an engineering-oriented technocratic pragmatism, perhaps we might strike a better balance between the intellectually rich Dutch architectural culture and the way its infrastructural thinking pervades everyday life?

Yet generally, most architects still tend to see technology as they do items in a product catalogue, absent-mindedly browsing Architizer perhaps. They fail to stand back and see how the collective impact of some of those items, in aggregate, has entirely transformed the ways in which we live together.

Yet being more sympathetic to Koolhaas—and to some extent OMA’s 2014 Venice Biennale show Elements could be seen as an attempt to engage with some of these building technologies at least—we could see his comment about ‘the city as the architect’s domain’ in a broader light, alongside an earlier comment of his noting that the city was the domain of architects specifically working to government ideals:

“The times in which architects were carrying out the good intentions of governments are long gone. There are no more ideals within governments; increased deregulation has strengthened the market economy to a fatal degree. The Universe is empty now or filled with companies. Progress is fragmented, completely scattered.” [Quoted here]

Indeed, despite the vestiges of 20th century oil-based infrastructures still present in our cities, the corporate inheritors of the deregulated landscape that Koolhaas described are now in the tech sector, led by the world’s most valuable companies. (At time of writing, the US-based corporations Google, Apple, Amazon, Facebook and Microsoft are five of the top six most valuable companies in the world. They are collectively now worth trillions, up from a combined valuation of about $2.2 trillion just two years ago, and are all based on the West Coast of the USA—at least until their Chinese counterparts of Tencent, Baidu and Alibaba expand their footprint.)

Just as importantly though, if not more so, these companies are arguably our greatest cultural influences, capable of hegemonic shifts in the political terrain and its emblematic social, and thus spatial, organisations.

And thus, of course, they now shape the way contemporary cities happen, too, not least via the associated second wave of Airbnb and Uber (again, also informed solely by that culturally narrow strip of coastline.) Yet these companies are markedly different to the mildly fevered, covetous smart city dreams of the last decade, with its 1980s ‘information technology’-led vision of control rooms and command centres. They are not ‘IT’ companies in the sense that IBM, Cisco and Siemens were, or are. Instead, they exploit an astonishingly widespread distribution of cellphones, which in turn enable vast social media networks, or so-called ‘sharing economy’ services, or rapidly emerging arrays of unsecured, flaky but intriguing Internet of Things devices, whether cars or washing machines or in-home speakers, lacing together their layered ‘skills’ via algorithms increasingly informed by machine learning capabilities.

This is where the real smart city is, and it is on the streets of Eindhoven as much as it is in Ensenada, Edmonton or Edinburgh. And as opposed to that more technocratic view of control, we can argue that this uneven distribution is shaping cities nonetheless—via an uneven, decentralised pattern, aimed at the entry level price point of everyday consumers. As a result, a different form of smart city has taken root, again utterly different to the Control Room Metaphor or the large infrastructure roll-out. Yet its potential for surveillance, exploitation and control is still present; just, to Rem’s point, “fragmented, scattered” and nimbly sidestepping governments.

Just what is it that makes today’s digital services so different, so appealing?

The change is tangible in terms of everyday interactions, the services people use, the infrastructures that enables these things, and the decision-making cultures and codes that increasingly shape how cities happen. This layer of urban activity happens in a space bigger than a cellphone and smaller than a building, suggesting that the latter is no longer the primary shaping influence upon urban life, but that this more malleable layer of infrastructure—mobility, temporary spaces and landscapes, events, devices—is where the action is.

https://medium.com/butwhatwasthequestion/the-battle-for-the-infrastructure-of-everyday-life-6c9b0572e57f

How should design respond to this progressively, in order to shape the direction of travel, whilst working with the grain of this layer, rather than against it? Many design-led projects over the last few years have been speculative endeavours, positioned to pose exactly these questions as much as supply simple answers. As noted before, this impulse has often come from outside architecture, with a few honourable exceptions, and instead from mutated forms of interaction design, service design, filmmaking, artistic practice or strategic design, such as UsTwo’s autonomous taxi, Helsinki Design Lab’s Brickstarter, Future Cities Catapult’s Quiet Tech, Superflux’s Buggy Air, James Bridle’s Autonomous Trap, Keiichi Matsuda’s HyperReality, Space Caviar’s RAM House, and so on. These often didactically explore what urban technology should do and should not do, rather than simply what it can do, leaving room for natural systems and human cultures to do what they do best.

Meanwhile, on the street itself, we see new service layers, such as mobility providers like Uber, Lyft or Zipcar; on-demand bus services like Beeline; post-grid infrastructure components like Tesla Powerwall; new co-housing models like Nightingale; new workspaces like WeWork; entirely new spatial conditions, like Airbnb; robotics like Google’s autonomous vehicles, Adidas’s Speedfactory, Amazon’s drones or Starship’s last-mile logistics robot; new forms of community practice, such as Peerby, or decision-making, such as civic crowdfunding platforms.

Some of these are emerging, such as the robots; some are here and now, demonstrated by Uber and Airbnb’s apparent ability to ‘disrupt’ some city governments and urban economies; some have already been and gone. For all the immediate allure, there are many aspects of this new service culture that are indeed uncomfortable and opaque, potentially damaging through reinforcing broader cultural patterns of individuality, inequality and intolerance.

The problem of ‘People Like Us’

For example, as George Packer memorably put it in the New Yorker, Uber appears to be largely concerned with “solving all the problems of being 20 years-old, with cash in hand.” In other words, the people that make Uber are making it for people like the people that make Uber—a characteristic sometimes known as ‘elite projection’; see also Elon Musk. This, despite Uber’s market valuation being based on a trajectory of global domination, which must necessarily include people entirely unlike them.

Though detailed aspects of Uber — the interface, the interactions — certainly provide a sketch of the way we might want to interact with urban mobility, the service also exemplifies the received wisdom that you make such things for people like you, and only pay on your own terms, avoiding wider responsibility for your impact on the city.

Indeed it is comparatively easy to imagine creating an urban car-sharing service for these conditions of “people like us”; say, as opposed to creating a new form of public transport that serves all citizens equally, irrespective of their economic or social status. Yet the latter is exactly the kind of design goal a city council in Eindhoven might have to set.

The private sector, after all, tends to cherry-pick the easy bits, the low-hanging fruit. Why wouldn’t they? They don’t really do difficult. That is ‘left to’ the public sector to pick up, or to drive. It’s easy to do Uber, a denuded form of shared transit that simply will not scale. It’s far more impressive, more challenging, to do genuinely public transport.

Yet the success of each subsequent ‘unicorn’ venture reinforces a hegemonic view that such technologies are the thrilling answer, and that the state’s role is largely to get out of the way.

This not only downplays the profoundly generative role of state-led innovation in generating the conditions for these technologies, and sometimes the technologies themselves, as Mariana Mazzucato has thoroughly described. It also makes it easier to further dismantle the apparatus of city government and civic society, leaving ever more fallow ground for startups to move into, portraying themselves as the answer to all Price’s unasked questions.

Margaret Thatcher famously said that she did not believe in ‘society’; yet we are now asked to believe in unicorns instead? The sheer lack of diversity in this thinking and practice is breath-taking.

However the paradox here is that such technologies also contain great promise, in terms of resource efficiency, say, or in terms of forming and reinforcing social fabric.

And intriguingly, while most of them currently simply overlay onto physical space, increasingly some indicate entirely new spatial implications.

How might Airbnb’s ability to re-program space change apartment design? How might vehicle-sharing change the form and volume of parking space? How might autonomous vehicles slowly dissolve the artefacts of the age of traffic engineering, enabling a complete rethinking of what a streetscape is, a form of shared space that Hans Monderman could only have dreamt of? How might retail spaces change form, or even disappear, based on ‘anticipatory logistics’, fabrication and autonomous delivery? How might super-local energy generation and storage change district design, and ownership models?

Perhaps these are the kind of questions we should ask of these technologies, as they begin to directly affect the spatial as well as the social, cultural, political and economic, and perhaps architects and planners, working alongside others, should be challenged to pose or address these questions too. [NB. This is what my team at Arup does.]

In Eindhoven we see two very different interpretations of how a city might begin to generate answers to these questions, in two very different districts: Strijp-S and Woensel Noord. Perhaps both might be woven together to create an Eindhoven which takes advantage of such technologies, yet with that broader public mission, a richer ethical dimension, with a broader idea of the city as a public good?

Loose-fit Strijp-S

In contrast to some high profile tabula rasa approaches to smart cities elsewhere, European cities like Eindhoven are rich accretions of layered history, going back centuries, if not millennia. Mostly, they are already built. Is there any room for such new spatial and infrastructural invention in these more sophisticated, more complex conditions?

An urban retrofit such as Eindhoven’s Strijp-S development provides an example of how yesterday’s fabric might provide the ‘loam’ for today’s and tomorrow’s European ‘network urbanism’. A functioning tech cluster, it is potentially far more than that.

In comparison to other local tech clusters—or ‘urban innovation districts’—the relative creative diversity of Strijp-S will be its strength. The development includes exemplary social housing as well as housing on the open market; a range of businesses, from tech startups to local food producers; circular economy showcases and design studios; giant skate parks and training colleges; high-end restaurants to discreet, somewhat grungy bars with murals of Lemmy and Ludvig Van.

Various scenes from Strijp-S, during Dutch Design Week 2015
This huge skate park will be a test of Strijp-S’s development programme: will it survive, as land value increases?

The openness of its built condition provides numerous opportunities for adoption by locals, from malleable incubator spaces through to high-grade office space for sports marketing companies via shipping container villages, from pristine roof gardens to empty car parks affording temporary occupation by Dutch Design Week, and perhaps most interestingly of all, the reversal of the traditional design and development processes demonstrated by the Space-S project, a form of scaled-up baugruppen model being developed on-site.

Inside Trudo’s fine social housing
Good incubator spaces at Strijp-S. Very good doorways enable openness or privacy.
Incubator spaces carved out of old Phillips buildings
Ludwig versus Lemmy in a bar at Strijp-S
Converted industrial buildings at Strijp-S
More Strijp-S; keeping the overhead pipes from the former industrial use is a masterstroke (though apparently not part of the original plan.)

The path dependency that the history of Phillips provides intrinsically imbues the place with a sense of invention and creation that more manufactured ‘innovation districts’ generally cannot. The stories contained in those walls, or embodied in the slowly-greening elevated pipes, generate the necessary milieu almost effortlessly. Yet it still takes careful stewardship and active curation to ensure that the diversity is retained, and can continue to enable the sensation of thriving urbanity that one can viscerally feel at Strijp-S on a good day. At other times, the place still feels ‘baggy’, somewhat empty, over-scaled in comparison with the actual activity on-site (as with the other retrofitted Philips HQ in the city, now reimagined, a little unimaginatively, as ‘High Tech Campus’.)

High Tech Campus suffers a little in comparison to Strijp-S, although it can be seen as a counterpoint, and part of the ambitious Brainport Eindhoven regional plan for the area, simply one of five or six different innovation campuses, each with a different focus and environment.

But places take years to write new stories about themselves.

The ‘smart strategy’ for Strijp-S is carefully framed in this context, and shows genuine promise. The question now, as with most things Strijp-S, is whether such promise can be truly realised.

Networked systems can act as a kind of connective tissue laced in-between existing infrastructures, which plays exactly to the concept of urban retrofit seen elsewhere here. These systems can also contain positive dynamics for Strijp-S, such as a foregrounding of urban experience, of human-centred services, of distributed, cellular organisation and shared ownership of infrastructure, of closed-loop off-grid systems, of optimising existing infrastructures, resources and fabric over building anew. The data generated by such fine-tuning, when radically enabled via open sensor technologies, could provide insights into when and how to build new infrastructure too, potentially enabling forms of lighter, cleaner, more agile, intensive, productive, adaptive, participative structures, systems and organisations.

David Weinberger famously argued that the Internet’s core design principle is “small pieces, loosely joined”, and this feels like a match with the iterative, ‘loose-fit’ sensibility of much of Strijp-S’s development. (Which, incidentally, feels quite the opposite of the inert, static masterplanning models that sit in Strijp-S’s marketing office.)

The reality of Strijp-S, despite its more diverse moments and the genuinely bottom-up baugruppen-esque components such as Space-S, is that it is a ‘top-down’ developer-led development. Yet the breadth of Strijp-S’s smart strategy, whilst not quite encapsulating all of the above, appears to recognise the real possibility in networked systems. However, it is clearly playing ‘catch-up’ to the rest of the development. It has come afterwards, and must now be woven into a development already well-progressed. Yet this ‘catch-up’ also provides an opportunity for learning from its first steps, and networked urbanism can move far more quickly than bricks and mortar ever could, should speed become an issue.

The proof will be in the pudding, as they say — or rather, the prototypes and subsequent products that enable new forms of living, working, playing, learning and decision-making at Strijp-S. Will Strijp-S genuinely engage with the possibility of such systems? And can Eindhoven see Strijp-S as a platform on which to test services that could be deployed more broadly across the region — or even generate practical ideas that could work across Europe?

The key question here is how to deploy a smart strategy — which is implicitly pinned up on structures largely generated by Silicon Valley — within a Northern European context with a strong history of social democracy, and of civic urbanity.

Bottom-up Woensel Noord

An entirely alternative approach, for a very different place, emerged out of the workshops organised under Het Nieuwe Instituut’s State of Eindhoven project. These focused on the Woensel Noord district of Eindhoven, and in particular its essentially residential Woenselse Heidel and De Tempel neighourhoods. The housing was built largely after the post-war period, as a form of family housing for workers at local industrial concerns. It is frequently referred to by locals as “breadwinner housing” i.e. a standardised mid-20th century model of terraces of houses for a notional family of four, where a father works for Phillips or DAF (or equivalent), the mother stays at home and looks after the 2.2 children: small garden in the front, larger at the back, shared green space inbetween the blocks, community services (bars, medical centres etc.) dotted throughout. Nothing bad in itself at all, yet a little too car-dominated, as with the prevailing urban planning at the time – even in the Netherlands – with an emphasis on private space over public, perhaps, and clearly built around a very clear idea of how 20th century nuclear families, working within an industrial context, might live, work and play.

Photos of housing types from our bike rides around Woensel Noord

Though these neighbourhoods predate the Vinex programme of housing that shaped much of the Netherlands in more recent decades, the urban fabric in Woensel Noord perhaps doesn’t feel that dissimilar to the outcomes for Vinex described in Vinex Atlas compendium:

Although hopes were pinned on bustling, vibrant urban areas, most got little further than monotonous housing estates.

And yet our investigations in Woensel Noord—essentially, a form of light-touch field research, led by Het Nieuwe Instituut, involving observations, interviews, workshops and various site visits—rapidly revealed that today’s population is anything but “monotonous”, and of quite different composition to the ‘breadwinner’ model that the housing was built for. Woensel Noord now is a very diverse place, simultaneously featuring gentrification and food banks, freshly arrived economic migrants, second and third generation migrants, and still some representatives of families from the first generation of ‘breadwinners’ that the place was built for. As a set of neighbourhoods, there were clearly many movements and activities that would appear informal in nature, compared to the communities of the mid- to late- twentieth century. Yet the building stock was unchanged, formally inert, still frozen in the mould designed for those original tenants.

Visits to, and discussions at, community centres, food banks at a local church, clothing exchanges, and local clubs

This tension was immediately fascinating, and through discussion nurtured germ of an idea we ended up calling Cloud Atlas.

The local municipality, Gemeente Eindhoven, was taking part in a national urban retrofit programme, based around a platform called Woonconnect, essentially focused on introducing sustainable elements into existing urban fabric, initially in Eckart Vaartbroek neighbourhood, and slated to be introduced across other areas of Woensel Noord. As far as we could tell, it focused solely on the built ‘elements’ of individual houses, co-opting Koolhaas’s language earlier, rather than the community of users of such elements i.e. local people, or streets, public spaces, neighbourhoods, services and the other elements of cities that bind places together. Perhaps we could usefully augment this programme with some detail about Woensel Noord? (It’s often better to ‘hitch your wagon’ to an existing programme, rather than split attention and energy in trying to create another one.)

In doing so, would we be able to demonstrate new ways of working with data, foregrounding its limitations in order to decry the ‘control room’ model of smart cities, with its false pretences to completeness and accuracy, in favour of something that recognised the essentially ephemeral and ambiguous nature of data? And yet with the goal of making something actually more interesting and perhaps useful as a result. This would necessarily involve counterpointing with qualitative research—finding different kinds of data—as well as taking advantage of the various local ‘literacy-building’ interventions led by our colleagues on the advisory group (particularly Anab Jain and local design firm Beam it Up working with local schoolchildren, local ‘salons’ by Tsjalling Swierstra, and Albert Jan Kruiter, who led a discussion around the numerous ‘data deserts’ that exist, augmented by insights from the many local stories collected by HNI with Kennisland’s Chris Sigaloff.)

We put these local elements together—the retrofit opportunity, our observations and insights about that community, and its interplay with the existing buildings and spaces—with an interest in design practices that enable user-centred, research-based, data-enabled fine-tuning of place. We wondered how we could better understand the diversity of community in Woensel Noord in order to better retrofit the architecture and infrastructure.

We noted that the official statistics about the area were somewhat static and unrepresentative of what we were seeing on the streets—almost as if the official data was as inert and anachronistic as the buildings themselves, both of these structures locked together in a bureaucratic embrace. (The Dutch have a tradition of being impeccable record-keepers, but of a certain type, characterised by the core urban analysis platform in use by municipalities, Buurtmonitor.)

As Albert Jan Kruiter put it:

The existing systems are in fact unsuited for the transformations that are taking place in the social domain: they do not fit the logic of the neighbourhood.

So, faced with arguably overly-simplistic census data that no longer seem to capture the essence, or “logic”, of the place, if it ever did, a different form of urban data would be required. Perhaps informal data would better capture the more ‘informal’ living patterns? (note: using the term ‘informal’ in the sense that it is informal related to the existing definitions of what formal was. These are subjective terms, clearly.) By informal data here, we mean data captured from sharing economy or social media platforms, using that as a proxy for activity, for fluid networked interaction. If one could gather data about, say, the journey of a power drill moving through the neighbourhood via the resource-sharing platform Peerby, would that tell us something about embedded social relationships in Woensel Noord and surrounds? Could we also grab ride-sharing data, Airbnb data, food delivery app data, any geo-located social media activity, such as Instagram pics, Twitter tweets or Facebook check-ins, volunteer groups or sports clubs using Facebook to coordinate attendance or events, and so on? Can we grab big chunks of cellphone data from operators, in order to understand high-level movement patterns? Could we work with IKEA to understand how some of the local community is using their canteen as a meeting place rather than a store? Can we locate all the vacant spaces in a neighbourhood, no matter how transient the vacancy? Can we grab environmental data for real-time sensors and correlate with health data from local doctors’ surgeries? Could we see the vapour trails of platforms like Nudge and Nextdoor, or locals like 040goedbezig or Gebiedonline?

Learning from Mumsnet as well as from Buurtmonitor.

Could we stir these points and vectors together into a big minestrone of data, in order to conjure up a richer sense of movement, of human interaction and of networking in the area? Might this kind of informal data neatly counterpoint the qualitative insights from ethnographic research, user research and other ways of engaging directly with people? Might that informal data be more in tune with the ‘qual’ data, in fact—and simply put, more useful—than the simplistic formal ‘quant’ data usually gathered by institutions? Yet might all three types of data—formal, informal and qualitative—combine to be genuinely insightful?

Very scruffy initial sketches from my notebook, indicating a very messy cloud of data over the planned Woensel Noord below (right), and the interplay between capturing data about the formalised welfare society versus the informal participatory society, inspired by OS City’s early visualisations (left)

Clearly, such an approach is immediately fraught with concerns: how to gather such data without impinging on privacy, most obviously and problematically. Dutch cities, including Eindhoven, are already under scrutiny for this , generally quite rightly. Yet if we can make the civic value of the data-gathering clear — and the act of it tangible and approachable — indicating what for and why, perhaps we might still work with such data.

It may still open liability issues for those involved in data-gathering. There are clearly technical and intellectual property issues involved in ‘scraping’ data from public-facing websites, or even issues as to gathering data from platforms like Uber, Airbnb or Facebook in the first place, given that they generally show little willingness to share detailed analytics. Equally, these platforms are no more wholly representative than the ‘top-down’ record-collecting of Buurtmonitor et al — but they do create a very different comparative counterpoint.

As a design research exercise, these concerns were flagged early, but crucially they were not allowed to derail the process of developing ideas. Embedding and documenting these concerns at the core of the idea was fundamental, yet not enough to prevent the initial development of a concept. Equally, the workshop would directly address issues of ownership of such recombined data, and its platform—in short, ideas centred on the data and its ownership being as local as possible, with the idea of small data rather than big data to the fore, as well as transformed practice around transparency, seamfulness, clear statements of public value, and the idea that the data might fade, have a way of being forgotten, or a ‘use by’ date, or be thrown away or become opaque once it leaves the streets from which it was generated.

In order to develop concepts, we sketched out the idea of a platform, dubbed Cloud Atlas (the name borrowed from David Mitchell’s book), implying some mapping of this messy, cloudy, uneven and opaque cloud-based data, a kind of mixing chamber that would enable stakeholders to collate and contrast numerous types of data, pitching the ad-hoc patterns of Uber drivers against the more regular, timetabled public transport patterns, say. Or noting the formal land ownership/tenure patterns plotted against actual space-use patterns via Airbnb. We envisaged Eindhoven as this far more complex terrain of data as a result, synaptic flashes of temporary interactions, fibrous nerve centres of consolidated connection.

To some extent, this was informed by another aspect of The State of Eindhoven: the mapping carried out by OSCity between July and October 2015, using informal sources like Facebook, Twitter and other platforms. That early work suggested the value in mapping the formal ‘welfare state’ against the so-called ‘participatory society’, with the former comprising public services such as police, healthcare, and education, and the latter comprising city gardens, cultural venues, sports clubs, non-profit foundations, community centres, and so on. (Of course, ‘participatory society’ is a loaded term too, redolent unfortunately of the largely discredited ‘Big Society’ rhetoric from the UK, as picked apart in Anab Jain’s essay.)

We wanted to build on that idea, and complicate it usefully, by continuing to push further into the ephemeral, arguably ‘bottom-up’ activity. And if that activity was generated and captured by increasingly powerful Californian corporates, and so not so ‘bottom-up’ in a sense, we wondered whether we could co-opt that data to generate local value, at least in terms of understanding some of the patterns of activity in Eindhoven’s society.

Understanding and describing this mixing chamber in high level was enough to pursue the idea into design-driven workshops, in order to figure out what to do with a new kind of understanding, where one possible.

Crucially, the goal of the workshops was to go beyond data. Data in itself does nothing. Its value is often overplayed, through careless statements such as “data is the new oil” (as if oil was a value proposition worth emulating in the 21st century.) Only when it is transformed into something—a service, an intervention, a policy—does data become genuinely productive. This means a shift of emphasis as regards data: from record-keeping, to service-creating. This is quite fundamental, as Albert Jan Kruiter points out:

When it comes to control and accountability, the government collects and processes a huge amount of data. But when it comes to service provision and assistance, it is less forthcoming. In this respect, the municipality of Eindhoven can begin at neighbourhood level.

The workshops were design-led in essence, then, focusing on synthesis over analysis, effectively. A diverse group of participants, albeit without local citizens in this instance, comprised local municipality, local housing operators, academics, and various local ‘cultural intermediaries’, as well as Het Neiuwe Instituut and Arup project teams. Each group presented positions on local context through a variety of different lenses—opportunities, issues, development schemes, ongoing research, and so on—before splitting into mixed groups, to develop potential projects, based on the notion that a Cloud Atlas platform could exist.

Linda Vlassenrood of Het Nieuwe Instituut, and me, setting the scene for Cloud Atlas on day one
Some of the Cloud Atlas workshop participants

Numerous ideas emerged, suggesting that a) Cloud Atlas could be a generative platform worth pursuing, yet that b) the focus on creating local interventions locally, as opposed to merely building databases, or outsourcing to pre-existing solutions, was found to be fundamentally important, and useful. This was partly an exercise in getting these particular stakeholders used to the idea of ‘making’, in order to prepare the ground for real making, subsequently.

Making these systems locally is not merely important in terms of increasing the likelihood of producing more grounded, appropriate and self-determined solutions (and thus likely to be more used, engaged with, as well as more focused in the first place), but also in terms of a key message: can we co-opt the dynamics of ‘Big Tech’, without inheriting the broadly unrepresentative ideologies and cultures they are produced within, and create solutions that take advantage of their techniques but are honed for European cities and communities?

Rather than go through all the ideas generated, one in particular demonstrated this sense of possibility, twisting some of the capabilities of Airbnb-like systems to local needs.

‘Roomsel Noord’ emerged from a story from the field research, based around elderly residents, often women, living alone in the family home in Woensel Noord. We heard about a resident, “Edith” (not her real name), who was typical of this situation: she could now barely afford to pay the rent on the house she’d been living in for years, yet did not want to have to leave the home she had brought up her family in. The house itself, still stuck in ‘breadwinner mode’ due to the intransigence of built fabric, cannot flex to her current condition—living alone, and with decreasing mobility, and increasing loneliness, which we know directly affects mental and physical health, with terrible outcomes for people, as well as increased costs for welfare state.

Edith’s social condition had changed, and was doing so increasingly fundamentally. Yet her physical environment had not.

The other social changes in the neighbourhood, however, meant that other potential uses existed for the now-superfluous space in Edith’s house. Data appeared to show increasing amounts of flexible working-from-home or not in an office, and an increase in the demand for flexible spaces accordingly, usually catered for by co-working spaces or serviced offices in the commercial centres of Eindhoven.

What if we could begin to use some of the spare space in the neighbourhood, yet consolidating the position of people who already live there, rather than the usual gentrifying process of expunging them? Could Edith remain in her house, yet offer up some of its temporarily vacant spaces, entirely on her terms? What kind of retrofit would that imply? Would that, in turn, begin to create, or simply reveal, a more vibrant, diverse and mixed-use Woensel Noord?

To cut a story of a long workshop short, Roomsel Noord was envisaged as a platform and service, built locally and operated locally—ideally by the Gemeente, or Woonbedrijf, or as a development of Woonconnect—that would enable Edith to use her house’s spare capacity to make money to help cover her rent, yet without her giving up the spaces she needs. It would imply a physical retrofit of the ground floor of these houses, which could be easily converted into temporary workspaces or community spaces, yet securing the rooms Edith needs upstairs. These spaces can also be used by Edith whenever she wants, of course; for instance, when her extended family comes to visit at weekends. It was a form of Airbnb, in a sense, but built on Edith’s own terms. Roomsel Noord would be a digital platform, accompanied by a programme of physical retrofits addressing the existing building stock, enabling Edith and others to use their space more flexibly, to their benefit, whilst retaining true ownership of it.

My quick sketch of Roomsel Noord: co-working on Wednesday, grandchildrens’ playroom on Saturday

Edith (again, a fictional user, but based on insights from the Woenbedrijf participants) could enable her family home to continue to respond to her personal needs, and those of her family, whilst also sharing aspects of it with her neighbourhood. The value generated in that sharing is financial — enabling her to stay in her home — but also social; by dissolving the walls between houses, and between houses and the street, the community is able to connect in new ways.

Sketching Roomsel Noord, with Arup’s Laurens Tait and Het Nieuw Instituut’s Klaas Kuitenbrouwer

So a street of individual houses, albeit in a terrace but with little connection anymore, becomes a more fluid consideration, with an increasingly shared front-of-house, and a lateral, horizontal relationship across the ground plane, and yet with atomic private spaces still retained upstairs and out the back.

Again, this is something managed on Edith’s own terms, as with platforms like Airbnb yet with true local ownership of decision and platform. She need only open up the ground floor for rent when she wants to—but those in the workshop from Woonbedrijf, whose staff have intimate knowledge of the community from their everyday interactions around maintenance, community services, rent and, informally, wellbeing, felt that this was viable enough to be pursued as an idea. It could even mean greater social connectivity, with folks like Edith coming into contact with different communities, and vice versa. That small data need only be owned, and relevant, super-locally. Again, complex, but perhaps less complex than we might think, if approached bottom-up rather than top-down, or if started with the sense that this only needs to work for a few streets in Woensel Noord, at least initially, as opposed to scaling to the entire Western world i.e. avoiding the typical tech start-up dynamic. (The European Commission-funded DECODE project is developing useful toolkits around data sovereignty and super-local ownership that could underpin such a platform.)

What those ‘ground floor activities’ are would of course need further research, and Cloud Atlas could be a way of beginning to assess the likely demand there, by learning about these increasingly informal and fluid working practices by scraping public adverts or announcements around workspace or event-spaces or community spaces. This is Cloud Atlas in demand-assessment mode; not giving definitive answers, but conveying the sense that something might ‘be there’, and warrant further investigation and engagement on the ground, face to face. Cloud Atlas generates leads, as per a detective story, rather than conclusions.

Gemeente Eindhoven’s Henri Koolen, with Arup’s Claire Fram

Clearly the physical retrofit would be an intriguing architectural problem. We quickly explored how a ground floor might be converted, through constructing different secure doors and entrances to various parts of the house, and how different forms of joinery and fittings might enable a variety of living/working conditions, rather than fixed, single-use spaces. This, whilst complex, is obviously part of a ‘trend’ in architectural design, led either by the impact of Airbnb, or the various co-living movements like Space-S or baugruppen in Berlin, or further, as discussed by Niklas Maak in his book Living Complex. This is likely to be a useful extension of practice for programmes like Woonconnect.

Security was discussed, as well as the emotional resonance of converting a family home to a business/community space, even if temporarily. It was suggested that certain areas, and building types, are more obvious starting points than others. Signage announcing the presence of the system, and available space, would need to be subtle, mobile and in character — and part physical, part digital. Some analysis of the number of homes, number of employed residents, and percentage of 65+ residents, was conducted.

In particular, we focused on the delicate balance to be struck by not modifying spaces so much that they no longer ‘feel like home’, or add unmanageable complexity. Issues of security, privacy and identity would be fundamental here—but then they are anyway, with any smart city project; it’s just that most projects don’t realise that. The fact that the Roomsel Noord idea immediately foregrounds these psychological and sociological complexities means that it is a promising project, in a sense: it feels like a ‘high stakes’ project because its subject matter matters. Strong, ethnographically-led, design research, informed by architecture, would help with this, as opposed to traditional architectural practice.

This is quite unlike those classical ‘go to’ smart city projects, such as smart streetlighting or smart meters, which simply do not matter in the same way, yet clearly could also have issues of privacy and identity attached. Somehow that is the worst of both worlds, enabling their careless roll-out, yet largely to little interest from the community, and largely benefitting people elsewhere.

Equally, the Cloud Atlas platform could help, as a kind of analytical ‘back end’ gathering informal data to be interrogated and developed by what we called an Understanding Service; essentially the same team carrying out the ethnographic research above. Thus we have quantitative and qualitative, worked together as a ‘stew’, providing on the one hand, insight into what services could be built, and on the other, a database back-end to build those services around. We envisaged these services to be designed, built and operated by something we dubbed the Interventions Service. While these names may need a bit of work, they were both key to the workshop outcomes, which were as much about shifting the discussion within the municipality and associated key stakeholders as it was about actually creating a platform. (In this sense, it was a similar exercise to the Brickstarter project, which Bryan Boyer and I led in Helsinki: developing a design probe in order to advance a discussion, both internally with a municipality, and externally, within an emerging civic tech movement.)

My quick sketches of where an Interventions Service might start, using our ‘Use Cases’ template
My quick sketches beginning to suggest various informally shared activities in Woensel Noord

This began to give the workshop participants a sense of Cloud Atlas—exploring how rich data might be able to better understand contemporary social dynamics, in turn in order to better understand how to transform the physical environment, via digital and other means—and crucially, began to develop a sketch of what one of those user-facing services might look like, as well as a more holistic view of the organisational ‘dark matter’ required to counterpoint and challenge the databased view of the world, and thus provide leads for engaging with the ‘matter’ of the place.

As is our wont, the Arup team sketched out a model for this kind of approach, indicating that hardware (the streets, or spaces, or buses, or gardens etc.) have associated actions, which can be understood via datapoints, from which we can generate insights — and with this flow, we can in turn use new actions to ultimately change the hardware of a place. This is a responsive, feedback loop back and forth, as a kind of relatively complex system.

Model emerging from the Cloud Atlas workshop, by Arup’s Laurens Tait

As a ‘design probe’, the workshop, and the idea of Cloud Atlas, attempted to flush out what kind of insights and applications might be focused on a place — in this case, Woensel Noord — as well as understanding what processes, cultures and skillsets would be required to develop, work with, and exploit such platforms and services. And that germ of an idea, Roomsel Noord, was a retrofit service, as an extension of Woonconnect, aimed at enabling use of shared and private space within Woensel Noord, connecting independent and flexible workers with ‘empty nesters’, for the benefit of both.

As well as the core flows around such a platform, and some of the organisational support requirements, we also sketched out other local services that might benefit from this ‘informal data engine’, such as a local autonomous shuttle, owned and run by the community. Given that most costs in transit are labour, the labour cost disappearing makes it viable for new forms of public transport, if conceived as such. So an on-demand shuttle, running in the gaps in-between tram and bus, could be something we can imagine working at a community scale, ferrying folk like Edith, as a form of ‘social transport’, to and from shops, station, bingo hall, garden, dancehall and so on. Its routes are generated around ‘non-grid’ patterns of movement, and so provides a counterpoint to that grid of public transport. (This would be in sharp comparison to a non-locally owned, non-public transport non-grid offer, like Uber or Lyft, which is already showing signs of destabilising the core premise and ridership of public transport, reducing cycling and walking, as well as leaching econmic value elsewhere, which would be deleterious reduction in quality of life, and indeed life choices and opportunities, for the likes of Edith and most other citizens, and in fact arguably anti-urban in essence.) Similarly, other forms of infrastructure might be increasingly viable for local ownership—including energy and other infrastructures of everyday life.

Within that we can see roles for private sector, public sector, ‘third sector’ and others. As a rough heuristic based on Tony Judt’s dictum, and paraphrasing slightly—‘don’t privatise railways, don’t nationalise sandwiches’—clearly, autonomous vehicles such as those local shuttles could be made by private companies, just as, in fact, the market leaders in such things happen to be French firms Navya and Easymile. Whereas local cooperative structures could own and operate a shuttle or two, and the local municipality can ensure that its data is gathered, stored and analysed, via delivering simple integrated ticketing with the broader Eindhoven transport networks. A ‘sketch’ of a network urbanism model emerged as one of the ideas in the workshop, primed by our inputs, yet the workshop focused on Roomsel Noord as a more obvious place to start, which would nonetheless begin to build a shared public infrastructure.

Public institutions and public life

Given the way the workshop was framed and facilitated, such a ‘design probe’ immediately puts on the table the question of new forms of institution, and organisation, ranging from cooperative ownership models through to the role of municipality itself. Suffice to say, these could not be resolved in a two-day workshop. Yet it was clear that there was an essential public-ness to all the models discussed, as opposed to the usual rhetoric around ‘start-ups’. The Gemeente could be seen as an ideal location for Cloud Atlas, in a sense, as the only entity with long term responsibility for the city of Eindhoven, as the legitimate arbiter of public data. Equally, one could imagine cooperative forms would work well at a neighbourhood level, around infrastructures like shuttles and energy systems, as well as potential interventions like Roomsel Noord. Either way, these questions require a truly on-the-ground analysis of what is going on in a place like Woensel Noord, as per the programmes that Het Nieuwe Instituut initiated under The State of Eindhoven, and a recognition of the value of what, and who, is already there.

In an important piece for India Times, Gautam Bhan discusses how the much-heralded smart city movement in India has, somewhat predictably, delivered very little; whereas Indian cities themselves have been progressing

Bhan says:

The failure — or at least the postponement — of the grand is also the survival of the ordinary and the everyday; the survival of citizens over cities; of infrastructures of everyday dignity over big, signature, spectacular projects; of incremental change over instantaneous transformation; of the bazaar over the mall, the shared auto over the expressway, survival over smartness.

Bhan writes that we need a different vocabulary and discourse around smart cities, quite different to that of the “grand schemes of programmers, policymakers and the high priests of transformation”, instead recognising the resilience, creativity, ingenuity and everyday innovation, if that is even the right word, of citizens and neighbourhoods themselves. Different words.

Words that start from who we are, that root themselves in the way our cities have been built and run, that don’t look for models anywhere other than our own streets.

Importantly, this is not to say that Bhan is recommending the dissolution of public policy, or even of grand projets themselves; as he says, “No one wins when public policy stutters.” The challenge is how do we rebuild and reconceive public policy and public services, learning from the streets and communities yet retaining a core local ownership, in every sense, and finding a role for private sector innovation nonetheless? In the broader public policy that smart cities are framed within, what lessons should we draw from the failure of privatisation in the UK to address the domain of the public sector, even when massively incentivised to do so? Or how do we face the growing sense that the tech sector won’t actually be able to sustainably scale once it hits the complex terrain of public life and public services?

In both cases, we can see the dawning realisation amongst many that we cannot get the private sector to do public sector jobs. A simple enough thought, yet still counter to the training of a generation of policymakers in place across Europe. Yet the Netherlands still has the possibility for a more balanced approach to this than many, as captured in Martijn de Waal’s deceptively simple diagram, indicating the interplay between three components that a smarter smart city—what he calls a hackable city—might be built around:

Diagram of flows in the Hackable City, by Martijn de Waal et al

Crucially, the regeneration of the institutional layer, and the public sector generally, may be the key element there. And there are signs of that happening, with the growth in ‘urban innovation labs’ within municipal governments and even programmes like The State of Eindhoven itself. Perhaps these are signals that we can deploy contemporary design and development practice, with its focus on human-centred and place-centred design, yet with the outcomes for local environments, communities, cities like Eindhoven?

Yet we still need to make an active choice here, and most smart city policies and strategies within Europe are mystifyingly silent on these matters, perhaps inadvertently reflecting that previous generation’s legacy thinking. Few of the EC’s smart city demonstrator programmes—with the potential exception of the Sharing Cities and Organicity programmes—are directly addressing these broader political, cultural and societal issues.

Yet the stakes are high. To put it bluntly, who would you rather see thrive in Eindhoven: Uber, or Edith?

Uber or Edith? Unicorns or society?

Well outside of Cloud Atlases, Eindhoven, and of urban development generally, the relative success of the UK’s Government Digital Services (GDS) initiative could also exemplify such a European response to these technologies.

It too borrows the productive techniques of Internet-based software — agile software methodologies, user-centred design, iterative prototyping — yet uses them to bind together an effective, meaningful and reliable array of government services. These are much the same toolkits that the products of the neoliberal Californian ideology use to negate the need for government, yet here deployed to reinforce government rather than diminish it. The value of GDS is not simply in the billion pounds per year (a conservative estimate, apparently) that it saves the UK taxpayer by not outsourcing, but in fact through those better services themselves. (Equally, when it did outsource, GDS shifted the balance from a few multinationals based around London to a national network of thousands of UK SMEs.)

GDS was not envisaged as an ‘austerity-era’ cost-saving measure—in fact, they refused to estimate a likely cost saving when asked to do so by the UK Cabinet Office—but as a better digital service for citizens. And it turns out that it was better, in almost every sense of that slippery word, as a result of being delivered by the public sector, as opposed to being outside and outsourced. (This case study, alongside others, is detailed in Helsinki Design Lab’s book Legible Practices.) This gives credence to the idea of Cloud Atlas, and its ‘Understanding and Interventions Services’, being run either by the municipality, Gemeente Eindhoven, or some other locally-owned, publically-owned entity.

Russell Davies, ex-Head of Strategy at GDS, has described how we need to frame a British or broader European response to the hegemonic stand-off of ‘startups versus government’. Optimistically, he sees huge potential for a distinctly home-grown approach, given what he called “the blind spots of Silicon Valley”. These he described as an overriding obsession with venture capital (VC) as the only possible ‘fuel’, and VC’s obsession in turn with ‘unicorns’. Davies also suggested other blind spots include a “super-serving of a San Francisco user”, entirely atypical in a broader global context (possibly even a broader American context), yet whose archetype is “forced via market power into other places”, another form of elite projection. These blind spots further include an assumption that “regulation, government and wider society can just be ignored.” This is simply not true, never mind undesirable.

Instead, the success of GDS more broadly indicates a role for technology in the public sector, which nonetheless absorbs, re-frames and deploys the values and game-plans of the private sector where applicable. This active curation of public and private to forge a hybrid — rather than the lazy assumption that the private sector is necessarily and always the most effective delivery mechanism — could mean that, in Davies’s view, “a more diverse vision of technology will out-compete a white, male rich version of technology.”

(Davies was interviewed for a podcast series hosted by venture capitalists Andreessen Horowitz, interestingly enough. His co-interviewee Martha Lane-Fox, effectively the UK’s ‘digital champion’, made similar points.)

Could we draw from this example of ‘active curation’ of contemporary technology’s dynamics for our European cities? Of adoption, yes, but also adaptation? Of working with diversity rather than monocultures? Of developing local solutions that generate and retain value in local urban economies, rather than feathering the nest of venture capitalists elsewhere?

Given this array of highly-capitalised services that is beginning to re-shape global cities, generally underpinned by ideologies that are quite different to those that have tended to form European urbanity — that “super-served” San Francisco dweller, rather than a citizen of Eindhoven — it behoves us to find a way to similarly forge a new network urbanism.

The promise of places like Strijp-S and Woensel Noord is that they show us a way forward by building on our past and present, and through making the case for diversity. They demonstrate approaches to layering an array of new spaces, activities and cultures over the ‘good bones’ of an inert physical structure; it works in the gaps left by history, and generates new possibilities as a result. They indicate the value of wrestling with the true complexity of public life, public space and public services, in a way that the private sector generally finds too difficult. They demonstrate the value of the super-local, as opposed to the allegedly-global, for the solutions found and invented on our own streets, as Gautam Bhan has it.

While the smart strategy of a place like Strijp-S is nascent, its scope means it is well-placed to add further layers of activity. And whilst retfrofits of places like Woensel Noord, such as Woonconnect, are also nascent, ideas like Cloud Atlas demonstrate how everyday technology might be co-opted and put to use for the benefit of the city as a public good, to generate a wide range of shared values on Eindhoven’s own terms, rather than be complicit in a leaching of local value to elsewhere. This could be highly productive, a genuine creative challenge as well as a core public good.

That emphasis on diversity of approach, and in fusing these top-down and bottom-up dynamics within a super-local public context, can be taken into that ‘bigger than a cellphone, smaller than a building’ urban space described earlier.

For here are people, networks, organisations, structures, vehicles, spaces, surfaces, objects, infrastructure, flora, fauna. And here are technologies. We need new tools, perspectives and processes to understand these shifting layers of activity, in order to shape the urban outcomes we might desire for European cities. It requires direct engagement with technology, recognising its role as a key driver of urbanism, yet framed holistically, via a new synthesis of disciplines, contexts and experiences, an entirely new form of network urbanism.

Could this more diverse, super-local approach blindside the “blind spots” of Silicon Valley, and forge an appropriately European approach to a civic urban technology better suited to places like 21st century Eindhoven?

Perhaps that is the question.


This 2017 essay is the original edit of the recently published article in the collection Becoming a Smart Society (Het Nieuwe Instuut, 2018) available in Dutch and English.

Like what you read? Give Dan Hill a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.