Designing for longevity in the Internet of Things (IoT)
By Jeremy Barribeau. University of Washington. 2018.
Abstract
The Internet of Things (IoT) is expected to grow at an exponential rate in the coming years. The Gartner Hype Cycle positions IoT near the point of peak expectations, revealing a horizon of consequences beyond. It is the intent of this paper to expose to designers and others involved in the development of the future of IoT the importance of addressing these potential consequences by developing strategies toward the improvement of IoT device longevity, so as not to perpetuate the unsustainable rate of obsolescence existent in prior generations of consumer technologies. I present an analysis of potential causes of premature device obsolescence, including technical and psychological causes, and extrapolate potential strategies that may be used to combat them, such as coupling hardware and software lifecycles and ‘ensouling’ devices. The strategies that emerge are intended to catalyze critical discussion regarding their potential efficacy. It is incumbent on design professionals to adapt this amalgam of learnings into their practice going forward, and to continue a critical discourse regarding sustainable practices in the design of technology.
Introduction
The internet of things, or “IoT”, whilst lacking a standardized definition, is an emergent technological phenomenon, generally considered to refer to an interconnected network of ‘smart’ devices. In more specific terms, it is the augmentation of everyday objects with microprocessors and wireless communications abilities (Karimi & Atkinson, 2013). It is a technological evolution on a scale that is “scarcely fathomable” (Fritsch, Shklovski, & Douglas-Jones, 2018, p. 1), and promises “great transformative potential.” (Fritsch et al., 2018, p. 1). It is the intent of this essay to untangle the potential from the hype of IoT, in order to avert a perpetuation of the unsustainable trend of short-lived consumer electronics products. I shall aim to equip the designers of tomorrow’s IoT devices with the necessary perspective to make responsible decisions with regard to device functionality and longevity. I will be analyzing a selection of product examples and sustainable design practices, in order to instill in designers a foundation of “ecological literacy” (Escobar, 2018, p. 158). In positioning existing consumer product examples and theories on sustainability within the context of IoT, I intend to consolidate disparate and abstract notions into a format tailored for designers practicing within this domain, who are rarely the intended recipients of existing critical discourse on the topic. This is especially timely in light of the contemporary imperative for sustainability and the relatively newfound agency that designers now enjoy in the deciding of what gets made and why.
The scale of IoT
The term, the ‘internet of thing’ dates back to the late 90’s (Vitali, Rognoli, & Arquilla, 2016). Despite this, the field is relatively nascent some twenty years later. So too are the applications and use cases that we see in IoT today. However, change is afoot: The pace with which IoT is anticipated to proliferate is extraordinary. There are currently more than 9 billion devices connected to the internet, which is expected to soar to 1 trillion by 2025, creating up to $6.2 trillion annual impact (Costigan & Lindstrom, 2016). Rand Europe estimates that there will be up to 50 billion IoT devices by 2020 (Schindler et al., 2013). This is in contrast to the ‘mere’ 7.3 billion personal computers tablets and smartphones that will exist in 2020 (Rothchild, 2014). If the scale isn’t yet apparent, consider for a moment the number of laptops and smartphones in your home today. Now consider the number of appliances, locks, light bulbs, power outlets and so on that will likely be transformed into IoT devices in the coming years (Karimi & Atkinson, 2013).
The scale of IoT proliferation and the potentially limitless applications of IoT technology should make evident the relevance and urgency of the topic for many design professionals. It is conceivable that the eventual ubiquity of IoT will inevitably create opportunities and overlaps across many design domains. As IoT crests the “peak of inflated expectations” in Gartner’s emerging technology hype cycle (Fig. 1) (Panetta, 2017), we find ourselves at an opportune moment to critically analyze the trajectory of this technology and our expectations of it. TOPP (2016, para 12) points to the glut of “connected gadgets” we see appearing in the market as a symptom of a technocentric attitude, resulting in “solutions looking for a problem”. A reflection on the potential consequences of said trajectory is necessary in order to identify remedial strategies and reach the “plateau of productivity” that IoT must mature into (Fig. 1, Panetta, 2017). It is the proposition of this essay that the arrival of IoT at this plateau and realization of its potential is only made possible by a significant adjustment of industry and designer practice toward improved device longevity.
Related works
It is no surprise that the peaking of IoT as indicated by the Gartner hype cycle (Panetta, 2017) has corresponded to a number of efforts to spur a reflection of ethics within IoT and sustainable design practices in general. These efforts stem from a number of sources, including creative agencies (such as TOPP), academic institutions and foundations (such as Rand, Mozilla and GCSP). There is a plethora of topics of concern within the IoT domain. The majority of foci center on security, privacy and control of information, usually from manufacturer (e.g. Karimi & Atkinson, 2014), legal (e.g. Rothchild, 2014) and public policy perspectives (e.g. Costigan & Lindstrom, 2016). It is worth observing that such publications tend to be limited in distribution and highly targeting toward specialized professions. In other words, they are unlikely to cross the desk of a designer.
Of particular note to designing with longevity in mind, are a handful of exemplary efforts to catalyze critical and reflective practices by IoT stakeholders. One such effort is the Mozilla foundation’s initiation of a ‘Trustmark’ for the internet of things. Mozilla Foundation, the non-profit parent to the company that develops the popular web browser, Firefox is known for its principles-based approach and advocacy for an open internet. It is interesting therefore that they have turned their attention to IoT. Who better than Mozilla to address a domain that some consider will make “the Internet revolution seem small by comparison.” (Costigan & Lindstrom, 2016, p. 9). They have also taken on the challenge of developing a protocol for interoperability, which I will touch upon later. In contrast to the aforementioned literature examples that aim to create government policy and oversight, The Mozilla Trustmark is an example of a self-governance framework. It aims to provide a certification standard that may be awarded to products and companies that opt into the program and fulfill certain evaluation criteria (Bihr, 2018). The criteria are largely in advocacy of the end-user, promoting full transparency, including data practices, security, manufacturer background information, supply-chain ethics, product function awareness and expected lifecycle (Bihr, 2018).
A related and comprehensive effort is that of RIOT, or the Responsible Internet of Things by the ThingsCon community. This interdisciplinary community meets annually at a conference and publishes an annual collection of essays. The two collections of essays released thus far are as diverse as the authors’ various domains, tending to surmise ethical learnings out of their respective projects and observations. Topics include IoT utility in a number of contexts, ethics and governance, industry and practitioner guidelines, and the pitfalls and consequences of technology amongst others.
IoT Design Manifesto is a one-page living manifesto document, outlining ten principles for the ethical design and development of IoT devices and services (Fig. 2, “IoT Design Manifesto 1.0 | Guidelines for responsible design in a connected world”). It is unique in that it was created by an assortment of design consultancies in order to guide their own work, and influence that of others in similar positions of influence. It is neither regulatory nor academic, but rather is indicative of an industry demand for ethical guideline and an example of the type of self-governance initiative and empowerment of the individual designer that we hope to build upon. This is further confirmed by a profusion of similar efforts to create industry guidelines, each by means of the manifesto format. Fritsch et al (2018), in their comparative analysis of 28 such manifestos on the topic, aptly interpret this phenomenon as a collective “response to a moment of crisis” (Fritsch et al., 2018, p. 3). The commonalities in the manifestos are many, and include a tone of urgency and an anticipation of impending epochal transition, to which we designers must respond. The manifestos are a call to action for designers to accept and claim responsibility for their influence in the IoT domain. Interestingly, in locating responsibility with the designer, none of the manifestos infer any value in public policy intervention (Fritsch et al., 2018). Despite an apparent universal argument for transparency and responsibility, there is a lack of a unified strategy in how responsibility can translate into actionable responses.
Method
The study of sustainable interaction design (SID), while incipient by most measures, is certainly mature in comparison to IoT-specific discourse. It is therefore apropos to our focus on design for longevity in IoT. The canon within SID is Blevis’s 2007 paper on the topic: “Sustainable Interaction Design: Invention & Disposal, Renewal & Reuse”. In his text, Blevis equips us not only with a principled theory for sustainability, but also a translation into praxis by way of a guiding rubrik. A significant tenet in Blevis’s argument is “achieving longevity of use” (Blevis, 2007, p. 507), and forms the specific frame of inquiry I shall apply to the broader imperative for a critical reflection on the trajectory of IoT thus far outlined. The intent of Blevis’s (2007) paper may be surmised as providing a springboard for tangential investigations in the SID field, evidenced by the extra research questions proposed to his readers. In order to develop a comprehensive ‘literacy’ of the factors that contribute to device longevity, we will be examining Blevis’s following concepts in relation to IoT device development: Product consequence, promoting quality and longevity, achieving heirloom status, transferal of ownership, remanufacturing for reuse, and decoupling ownership. As longevity pertains to a comparative measure of time, it is fitting that I shall complement Blevis’s (2007) aforementioned SID concepts with that of a chronological structure. That is, that we shall approach the concept of longevity holistically, and not just in context of individual product, situating IoT as both a “backward and forward compatible” (Tsaknaki & Fernaeus, 2016, p. 5979) endeavor. To this end, I shall present an analysis of what might precede an IoT product, how the product might achieve (or fail to achieve) longevity of use, and how one might develop strategies to prolong a device’s longevity and value.
Establishing value with IoT: What came before.
I propose an expanding of the traditionally limited temporal frame used to consider the ecological impact and longevity of a product. In addition to considering the life cycle and use of a proposed product, the designer should also consider what ‘came before’ it. In establishing the value of an IoT product in its earliest phase of concept germination, it may seem straightforward to evaluate the landscape of products or solutions that precede it. Often though, this analysis occurs in a competitive light, rather than in the interest of sustainability or cementing the permanence of the proposed IoT device. In order to identify a baseline upon which to comparatively evaluate a proposed product, we must first consider both the potential for unintended consequences (TOPP, 2016) and what may be “displaced or obsoleted” (Blevis, 2007, p. 503) by the proposed product. The introduction of a new and presumably ‘superior’ product in most cases will result in the obsolescence of the product it replaces. The smart thermostat replaces its ‘dumb’ predecessor, the smart light bulb may result in a quantity of still functional (and likely toxic) light bulbs being disposed of. This begs a number of questions including how a solution may be ‘phased-in’ or transition into place, and what can be done with the products that are obsoleted. One approach is to design products not as isolated solutions, but as complementary to existent products. The Emberlight product (Fig. 3), an adapter that makes traditional bulbs ‘smart’ (Paul, 2017), exemplifies an IoT product that endeavors to fit within an existing context without generating bulk obsolescence. Blevis’s (2007) example of the Garmin GPS product (Fig. 4) is fitting, as it “re-enable[s]” (TOPP, 2016, para. 17) older cars with enhanced functionality. Whilst the GPS product carries a resource cost in its manufacture, it presents a clear ‘net-gain’ in its extension of the lifespan of the older car and also in its reduction of unnecessary mileage caused by inefficient driving routes. As with a smart thermostat, or a lightbulb with an occupancy sensor, the GPS achieves net-value in its capacity to “promote more sustainable behaviors” (Blevis, 2007, p. 503).
Establishing value with IoT: When is smart, really smart?
IoT devices are frequently crowdfunded on platforms such as Kickstarter (Vitali, Arquilla, & Tolino, 2017). On one hand, crowdfunding may benefit the IoT domain by funding projects that might not otherwise attract traditional forms of capital (Vitali et al., 2017), potentially enabling smaller players to enter the high stakes manufacturing game currently crowded with big name brands such as Apple, Google and Amazon (Francis, 2017). However, on the other hand, the glut of connected devices on Kickstarter can also be interpreted as a symptom of the ‘hype phase’ within which we currently find IoT (TOPP, 2016). Certain Kickstarter products present a more complex value proposition whereby the discernment of whether it offers a net-value is made somewhat more challenging. The “Ecogarden” product on Kickstarter (Fig. 6, tinyurl.com/yajhbo46) is one such example. The concept of a fully automated, closed-loop aquaponic ecosystem is attractive at first blush. However, the product is self-contained and has no physical user-interface, as it is controlled exclusively via a smartphone app. This introduces complexity in establishing whether the product offers a net-gain, or not: What is to become of this product should its connectivity fail? In the stewardship of living creatures, is full automation appropriate, considering the life and death consequences failure would result in? In what use-case would someone wish to avail the marketed feature of remote control from “anywhere in the world”? Is an electronic appliance to sustain decorative fish truly an ecologically progressive proposition, as it is indeed marketed?
In the Trojan room coffee pot at Cambridge University (Fig. 7), we find an example that at face value seems to be the forebear of the gratuitous smart devices of today. This now famous coffee pot from 1991 was quite possibly the world’s first IoT device. It had a primitive digital camera trained on it, delivering still images every 20 seconds to a server, so that coffee-craving academics would view the ‘live’ image, rather than suffer the inconvenience of walking all the way to the coffee machine, only to find it empty (Stafford-Fraser, 1995). Whilst the functionality may seem a little absurd, it was developed in response to a real, existent user pain-point. In fact, as a testament to its utility, it was in service for 10 years. This invention responded to the existing routines and emergent behaviors of its context and users, which TOPP (2016) argues are fundamental criteria for the success of any IoT device.
Technical causes of reduced longevity
Consumer technology products have a finite lifespan. They either stop functioning or lose relevance and are replaced by new, theoretically improved products. The first causal category that begs investigation is what I shall describe as ‘technical longevity’. According to Grebler (2017), there are three causes that result in IoT products turning into “paperweights”: Connected services, local software, and physical degradation. I will expand upon the first two, as well as broadening our analysis to include enduring maintenance, asynchronous technologies and qualities, interoperability and psychological perception. Each of these factors may contribute to diminishing or increasing the lifespan of a product. Failure in any one of them will result in product obsolescence or break down and should be considered the primary foci for manufacturers and designers of IoT products; a sort of reimagined Mazlow’s hierarchy outlining the basic IoT needs that must be fulfilled in order to eventually enact higher level strategies toward longevity.
The risks and rewards of external reliance
IoT devices are most often enhanced by the leveraging of external capabilities, whether utilizing internet-enabled functionality, or ‘collaborating’ with other proximal IoT devices. Furthermore, there is improved economy in manufacturing a product with less local data storage and processing power, instead offloading this functionality to a remote server (Newman, 2016). Likewise, no longer requiring an included screen or input UI on every single product affords the designer many opportunities and freedoms. Furthermore, there are obvious ecological benefits to a reduction in redundant hardware components in these cases. The constant connectivity at the core of IoT promises another ecological gain, in that a device can receive software updates that improve its functionality over time, potentially delaying its replacement. However, these core value propositions present somewhat of a double-edged sword. Reliance on external processing, constant connectivity, and third-party communications protocols, devices and operating systems that are sometimes beyond the control of a single IoT manufacturer, introduces risks of failure and impact to the longevity of an IoT device.
Reliance on connectivity and the phenomenon of ‘Abandonwear’
If internet connectivity and the functionality that it entails is core to the value proposition of an IoT device, it then follows that an IoT product divorced from the internet is left with little remaining utility. What use would the Ecogarden be if its founders move onto a new project, “stop[ped] paying the server bills” (Burns, 2016), and left users with a control app that no longer functioned? It is difficult to know exactly how many IoT devices have been rendered useless “abandonware”(Perlow, 2015) as a result of backend service shutdowns. There are certainly too many to list here, though we will analyze a select few in order to identify the leading causes of technical failure. The company behind the $400 Aether Cone smart speaker (Fig. 8) and Rdio, (the music streaming service that it exclusively used) both went out of business in 2015 (Perlow, 2015). After being acquired by Nest (a Google/Alphabet owned company), the IoT smart hub Revolv (Fig. 9) was abandoned along with its backend services. Netgear abandoned its connected camera product Vuezone in a pivot toward their new smart product line Arlo. Between insolvency, acquisition and strategy-pivots, numerous examples abound of manufacturers leaving consumers in the lurch. There appears no evidence of manufacturers readily accepting accountability, nor committing to long-term backend functionality and enduring firmware updates. A guarantee of manufacturer solvency and long-term commitment is fundamental to device longevity when IoT devices are defined by their connection to, and reliance upon a remote internet backend.
Asynchronous technologies
Reliance on internet backends and continued manufacturer commitment is a side effect of the inter-reliance built into the IoT ecosystem. It may presumably be mitigated by means of an adjustment to individual manufacturer strategy. Conversely, beyond the control of a single IoT manufacturer is the reliance of IoT devices on external technologies, protocols and operating systems that facilitate the various IoT ecosystems. The Bluetooth, WiFi, Zigbee and Z-Wave communications standards that are utilized by many IoT devices are continually evolving to incorporate improved technologies and functionality. The Bluetooth in my 2011 car, for instance, is not able to take full advantage of the functionality offered by the improved Bluetooth in my 2017 iPhone. Asynchronous product and technology cycles mean that IoT devices that use smartphones for setup or control may run into serious compatibility issues. Manufacturers of IoT devices will therefore need to commit to continually updating their phone and tablet apps in order to keep up with operating system updates and find ways to continue to support users of legacy phone hardware and operating systems.
Mismatched lifespans
Certain IoT implementations have the potential to create a mismatch between hardware and software lifespans. The integration of electronics and software into a product that is perceived as having a long lifespan, such as jewelry, for instance, may be to the detriment of its inherent material longevity (TOPP, 2016). Blevis (2007, p. 503) argues that “software and hardware are intimately connected to a cycle of mutual obsolescence”. By this, he means that hardware advances prompt the development of new supporting software, and software updates can outpace the abilities of hardware, thereby demanding hardware upgrades. In the case of a ‘smart’ IoT refrigerator with a touchscreen interface (Fig. 10a), we can imagine that the primary functionality of refrigeration is likely to function for ten years at the very minimum. Yet, it is hard to imagine that computational power, screen resolution and touchscreen interaction paradigms will not all advance significantly even in the next two years (Fig. 10b). In such a situation, the product becomes weighed down and devalued by its lowest common denominator. The inevitability of improved technology and new features may cause a consumer to prematurely replace their fridge because of its outdated IoT components and resulting outdated feature-set. Blevis (2007) presents the Leica M8 digital camera (Fig. 11) as a positive example of a product constructed with such quality that we may surmise it will be used and valued for a long time. Whilst there are aspects of this product that are admirable, such as its backward compatibility with existing generations of lenses, it is also an example of an unsuitable coupling of permanence and impermanence. The material quality of the camera can never hope to keep pace with developments in processing power, information storage capacity, screen quality, imaging sensor ability, user interface, digital formats and so on. For this reason, it may end up being ‘retained’ for its impressive material quality, but not necessarily used, nor relevant in the long term. In this way it promotes replacement and does not reflect a real and holistic form of longevity.
The need for interoperability
“IoT is a veritable tower of babel” (Perlow, 2015). Between the various proprietary ecosystems (Perlow, 2015) and battling protocols (Karimi & Atkinson, 2013), the IoT landscape is resembling the VHS/Betamax format wars of the 1980’s. With IoT, consumers are not simply choosing a product to buy, but declaring fealty to a brand or closed ecosystem (Paul, 2017). They risk being left out in the cold if they accidentally back the wrong horse. Consumers, when surveyed, have been found to purchase redundant electronic devices precisely because of “cross-platform compatibility issues” (Blevis & Stolterman, 2017, p. 17). The call for interoperability and the creation of communications standards is a common one (Banafa, 2017; Lindqvist & Neumann, 2017; Newman, 2016; Perlow, 2016). The introduction of standards could open up the IoT domain for smaller players to compete with big name brands, their products able to ‘collaborate’ with others in a universal ecosystem. In fact, consumers currently shy away from offerings from smaller companies for fear of the companies disappearing (Perlow, 2015). This would be mitigated by the promise of interoperability. Pogue (2013) places the ball in the consumer’s court, inciting them to reject manipulation by the technology industry. Consumers, by way of their purchasing decisions, have the power to rebuff the makers of closed proprietary systems (Perlow, 2016).
Psychological causes of reduced longevity
Having analyzed a number of leading technical factors involved in the longevity of IoT devices, I shall now endeavor to uncover factors of a more immaterial nature, as “things have a technical, an economical, and a psychological life span” (Verbeek, 2006, p. 373): There are psychological factors at play that pertain to consumers’ expectations of how long their artifacts should last, as well as actors who exert influence on said expectations. The same person who may feel embarrassed by their four-year-old iPhone, has no qualms driving a four-year-old car, nor do they feel the need to replace a ten-year-old fridge in their home (Pogue, 2013). In a survey conducted by Blevis & Stolterman (2017), respondents were asked how frequently they would choose to replace certain products if money were no object. For cellphones, mp3 players and kitchen appliances, the majority responded that they would ideally replace these items every year, and laptops every two. It is difficult to extract from these findings what proportion of responsibility the consumer bears, versus the industrial marketing engine. Cellphone networks, through efforts such as Verizon’s “New every two” campaign (Pogue, 2013, p. 32), have constructed an arbitrary frequency of replacement for which no technical explanation can be found. Apple’s more recent iPhone upgrade plan has now shortened this replacement period to one year. IoT devices are beginning to unnecessarily fall into this same construct. We are treating IoT devices as “disposable”, and using asset depreciation models appropriated from other narrowly related domains (Perlow, 2015). The industry is rushing product to market, releasing new features as soon as they are invented (Banafa, 2017), while consumers are also perpetuating this trend in their constant demand for new features (Pogue, 2013). It may then fall to designers to assume the voice of reason and intervene in this highly unsustainable, self-perpetuating cycle.
Blevis & Stolterman (2017), in their aforementioned survey, discover a number of factors which contribute to peoples’ perceptions of item value and permanence. Items with visual appeal, tools of creativity, items that reflect identity and items received as gifts were amongst the factors that promote retention and which I will be pursuing a deeper understanding of. I should note, however, that retention of items does not translate into longevity of use and value. Rather this attachment of sentiment is valuable in understanding how we may hope to stimulate connections between people and their artifacts, not just for retention, but a more constructive and holistic interpretation of longevity. Perceptions leading to improved longevity are inferred by consumers not only from an item’s utility, but also the emotional qualities intentionally imprinted upon it by its designer (Wakkary & Tanenbaum, 2009).
Summary of causes
Thus far I have identified an assortment of factors that contribute to diminished device longevity. Each presents an opportunity for intervention toward improved longevity. Beginning with establishing an initial value proposition and evaluating net gain in consideration of preceding products as well as a product’s potential to affect changes in user behavior. And on to an analysis of potential technical causes of failure and decreased value, including arguments for and against reliance on external components and infrastructure, asynchronicity of technologies and artifact permanence, and the need for improved interoperability. And concluding with consumer expectations of product longevity and the somewhat less corporeal aspects that may be harnessed in the promotion of perceived permanence and value. I will now identify four overarching veins of strategy, each with the potential to significantly impact the trajectory of IoT device definition, design and use.
Modularity and Repair
A shift toward modularity and repairability is one such strategy through which we can affect greater product longevity. The future is, of course, uncertain: It is difficult, nay impossible, for the designer to imagine how a product designed today will fair in the world in ten years’ time. Modularity may help us to design in recognition of our uncertain future and to extend a product’s conceivable life by enabling upgradability. Google’s project Ara (Fig. 12), Fairphone (Fig. 13) and Puzzlephone (Fig. 14) are all forays into modular smartphones. Conceptually, they allow for consumers to customize functionality by choosing components that best suit them. The products address a mismatch in technical longevity, in that a battery that lasts only three years can be easily replaced whilst retaining a screen designed to last ten. If a consumer requires extra computing power, or an improved camera, the respective modules alone may be upgraded (Peters, 2016). The old modules may be used by other people (Peters, 2016) in a what Blevis promotes as a “conscientious redistribution of older technologies” (Blevis, 2007, p. 511). This approach may serve to delay complete device replacement. Ideally, products could be designed for complete disassembly and separation, so that their constituent components and materials could be recycled (Jackson & Kang, 2014; Peters, 2016), or even better, reconstituted into new and updated iterations. This frames longevity in a new light, in that if there is less consequence and cost associated with reconfiguring an old product into a new one, we may not need a product to exist in its original form in perpetuity.
Ensoulment through materiality
As we have discussed thus far, product longevity may be conflated with both continued utility and personal attachment. On the latter point, strategies should be developed so that “attachment between products and their users could be stimulated and enhanced” (Verbeek, 2006, p. 373). There are two such strategies that bear consideration, yet at face value are antithetical to one another. One is to employ material quality in order to elicit a perception of permanence, the other to embrace impermanence by designing products to be adapted by users. Both are applicable tactics in aid of what Blevis & Stolterman (2017, p. 2) describe as the “ensoulment” of artifacts. Whilst earlier we found fault with Blevis’s (2007) example of the Leica M8 camera (Fig. 11), its build quality and use of enduring, authentic materials is certainly attractive. Though, perhaps a more suitable example may be found in a less complex and more enduring range of household products. Verbeek (2006), for instance, exemplifies Sven Adolph’s unconventional ceramic heater (Fig. 15). Its conspicuous design, authentic materiality and interactivity promotes it above the station one would expect of such an appliance. TOPP (2016) likewise attributes the continued success of Google’s Nest smart thermostat to its timeless design (Fig. 16). Vitali et al (2017) support the same assertion in their example of Thingk, a crowdfunded kitchen scale and clock made of marble, aluminum and wood primitive geometries (Fig. 17a, 17b). So, it would seem that authentic and permanent materials play a role in the success of domestic appliances that exist in a context where one might already find such permanent materials in use. Through proximal existence to like-materials applied throughout the built environment, the product may conceivably be inscribed with a corresponding perception of permanence. We can infer also, as Blevis & Stolterman (2017) discovered in their survey, that as longevity may be derived from items that carry a capacity for personal expression, that this may also be true of a smart appliance clad in materials more commonly used expressively in interior decoration. This strategy is further advocated by Blevis’s (2007) rubrik, which insists upon the improvement of quality in pursuit of a product attaining “heirloom status” (Blevis, 2007, p. 507).
Design for impermanence: Materiality
Materials have persuasive potential beyond that of their implied or inherent permanence. The physical nature of technological artifacts can convey replaceability, customizability (Wakkary & Tanenbaum, 2009) and reveal delightful qualities through intentional patination and wear markings, for instance (Steven, 2006; Verbeek, 2006) [A sort of graceful aging, if you will (TOPP in Fritsch et al., 2018)]. Contrary to the strategy of using enduring materials, is the proposition that permanence in one state is not the key to enduring value and longevity of use (Tsaknaki & Fernaeus, 2016). One contention regarding the use of ‘permanent’ materials is that they convey a semantic of stagnation and rigidity in time. They are “locked for modification… [and] also difficult to repair” (Tsaknaki & Fernaeus, 2016, p. 5977). The proposition of departing from the use of durable, permanent materials, seems in direct contrast to the seamless, polished consumer electronics products so common today. In fact, this strategy is likely a bitter pill to swallow for designers trained in the modernist aesthetic and its tendency towards minimalism. Products such as the Apple iPad are designed in such a way that they are impersonal, resisting “sentiment and nostalgia …[and] immune to attachment” (Herrman, 2018). Instead, Tsaknaki & Fernaeus (2016) are proponents of a phenomenological honesty that recouples hardware with the ephemerality of technology and software and the individuality of the end-user. “Things are constantly becoming, changing and transforming” (Tsaknaki & Fernaeus, 2016, p. 5978), and should be made to convey a “sense of liveness” (Tsaknaki & Fernaeus, 2016, p. 5976) in order to persist in a constant tumult of socio-cultural change.
Design for impermanence: Participatory design
The use of impermanent materiality can invite adaptation (Fig. 18). There is a psychological value in an artifact that is caused by its capacity to support modification by the unselfconscious designer, in what Enzio Manzini’s describes as “caring for objects” (in Blevis & Stolterman, 2017, p. 5). In caring for objects, the end user takes on the role of caretaker (Tsaknaki & Fernaeus, 2016), developing an affinity for the object. The principle is not dissimilar to the “Ikea effect” (Norton et al in Robbins, 2017, p. 51), whereby personal value is inscribed in Ikea furniture through the very process of its assembly.
The effect is presumably intensified further in the phenomena of ‘Ikea hacking’ (Fig. 19), whereby the ‘consumer’ becomes a co-designer of the artifact through their creative expression. Blevis’s (2007) imperative for designers to guide consumers away from their current fetishization of invention may be paradoxically achieved in embracing quite the opposite: In celebrating invention by empowering the consumer to appropriate and re-invent. It entails a design strategy that considers the product a “living entity” (Gerhard Fischer in Wakkary & Tanenbaum, 2009, p. 3), continuing to evolve “after deployment” (Tsaknaki & Fernaeus, 2016, p. 5978) and beyond the hands of its original designer. In this case, not knowing the future is remedied by not being overly prescriptive, not locking out later appropriation and modification, and recognizing that technologies can evolve beyond their original context and imagined purpose (Wakkary & Tanenbaum, 2009). The temporal boundaries of pre and post-deployment need not preclude participatory design at other opportune moments in the product lifecycle. The Thingk kitchen appliance I brought up previously (Fig. 17a, 17b), is not only an example of authentic materiality, but also of collaboration between the manufacturer and the end-user throughout the design process. Both its materiality and its very definition (including its feature-set) were the results of a successful crowdfunding campaign. Crowdfunding, whilst problematic in its promotion of superfluous novelty, can provide a platform for such participatory design. It can guarantee market demand before manufacture and fund projects that would never otherwise see the light of day (Vitali et al., 2017).
Sustainable business models
An alternative strategy to enabling product adaptation by users and community may be to consider a future for IoT that moves beyond a hardware-based business model. Longevity becomes both incentivized and paradoxically irrelevant (in terms of our current interpretation) if IoT devices are not owned by their end users. Rather an alternate financial model would incentivize manufacturers to design devices that are built to maximize extensibility or to be redistributed and repurposed upon obsolescence within its original context and ‘owner’. IoT will eventually reach a point of hardware saturation and the industry will need to adapt its business model (Villum, 2018). As computing becomes ubiquitous, it may blend “into the fabric of everyday life” (Weiser in Vitali, Arquilla, & Tolino, 2017, p. 595) to a degree that renders its interaction modalities, not its hardware, most prominent. At this particular moment in the future, consumers may perceive IoT not so much as devices, but as invisible infrastructure, likely subscribing to whichever artificial intelligence product is the ‘flavor du jour’ (Newman, 2016). We already see such a trend in the software industry moving from the traditional model of retail licensing to that of subscription licensing (Tsaknaki & Fernaeus, 2016), and it is only a matter of time until hardware follows suit. In the meantime, predictions of the vast quantities and intimacy of data generated by IoT devices is suggestion in itself of a potential revenue model (Karimi & Atkinson, 2013; Lindqvist & Neumann, 2017). If IoT revenue models begin to center on data, versus hardware, then improved longevity may be undermined by the potential consumer and regulatory backlash against this pervasive technology. Such a cataclysmic mass obsolescence needs to be preempted by the promotion of privacy and security policies that favor the consumer above all else. Consistency, transparency and ethical privacy policies, robust security and visibility of ‘invisible’ technology will become necessary topics for future investigation. Consumer utility will need to be consistently balanced against risks to consumers’ wellbeing.
Discussion
I started by building upon Blevis’s (2007) SID foundation, applying it specifically to IoT. After an analysis of technical and psychological causes of diminished device longevity, I outlined four overarching strategies that may be employed after addressing these initial causes. Modularity and reparability, ensoulment through materiality, design for impermanence and innovative business models may each be applied in suitable circumstances to affect a significant pivot in the trajectory of IoT; a departure from superfluous product and needless consumption, toward a sustainable realization of the IoT’s potential for real and lasting utility and desirability.
This paper was catalyzed by Blevis’s 2007 text, and it is in the same vein that I consider this paper to be a springboard for continued discourse on the subject of IoT device longevity and the broader topics surrounding IoT and sustainability. To this end, there are a number of further research questions that have emerged out of my investigation into the causes of insufficient IoT device longevity. I believe there is a role for speculative design and design leadership within the IoT domain. Thus far, IoT has been the purview of industry, its value promulgated by marketeers. There seems to exist a void in the creation of IoT products and services that are conceived in isolation from industry, and its inherent hardware and technology-centric business models. The industry would benefit greatly from human-centered design initiatives and a greater diversity of future visioning.
We should consider, for pragmatism’s sake, an alternative to the self-governed application of ethics that underpins the strategies presented throughout this essay. Whilst the various IoT manifestos elude any proposition of government oversight, it could be argued, that for an issue of such global consequence, regulation is necessary. It is surprising even, that given the potentially grave repercussions of IoT device failure, government involvement is not more common. This becomes especially important when anticipating the possibility of data-based revenue models and the necessity for consumer advocacy this will demand. Zittrain’s (2018) proposition of enforcing accountability and commitment from device manufacturers by way of a financial bond system is intriguing and warrants further investigation. There seems to be an opportunity for serious impact in the application of such accountability schemes within the crowdfunding domain, especially. Similarly, a system to more effectively communicate to consumers a device’s anticipated lifespan, upgradability, backend service reliance, and limitations of interoperability before they make a purchase would be a valuable contribution in the short term. Moreover, a regulatory intervention is essential in order to ensure that manufacturers create and issue ‘dead man’ or open-source firmware and enable local simulation of backend infrastructure upon a product’s end-of-life.
Conclusion
The decisions we make today will result in either a continuation of current, unsustainable practices, or in the forging of an alternative to the rampant consumerism that pervades the technology industry. IoT, by virtue of its anticipated scale, is one domain that will have far-reaching consequences, and therefore obligates those involved in its development to take a stand and implement corrective actions towards improved device longevity and greater sustainability. Through the analysis and recommendations presented in this paper, designers should recognize the gravity of the situation and their potential for impact. They should feel empowered to take action, if not by means of the specific strategies identified herein, then by the conception of others and the continuation of critical discourse in the IoT domain that this essay ultimately aspires to catalyze.
Bibliography
Aether Cone IoT speaker. [Photograph]. (n.d.). Retrieved from https://thegadgetflow.com/portfolio/aether-cone-the-thinking-music-player/
Banafa, A. (2017, March 14). Three Major Challenges Facing IoT — IEEE Internet of Things. Retrieved November 6, 2018, from https://iot.ieee.org/newsletter/march-2017/three-major-challenges-facing-iot.html
Bihr, P. (2018, April 6). Towards a trustmark for IoT. Retrieved November 7, 2018, from https://medium.com/thingscon/over-the-last-few-months-ive-been-working-on-the-core-concept-of-a-trustmark-for-iot-that-still-e1b5b120f1d8
Blevis, E. (2007). Sustainable interaction design: invention & disposal, renewal & reuse. In Proceedings of the SIGCHI conference on Human factors in computing systems — CHI ’07 (p. 503). San Jose, California, USA: ACM Press. https://doi.org/10.1145/1240624.1240705
Blevis, E., & Stolterman, E. (2017). Ensoulment and sustainable interaction design. Proceedings of IASDR, Hongkong., 23.
Burns, M. (2016). Nest demonstrates the risks of being an early adopter by shutting down Revolv | TechCrunch. Retrieved November 24, 2018, from https://techcrunch.com/2016/04/04/nest-demonstrates-the-risks-of-being-an-early-adopter-by-shutting-down-revolv/
Costigan, S. S., & Lindstrom, G. (2016). Policy and the Internet of Things. Connections: The Quarterly Journal, 15(2), 9–18. https://doi.org/10.11610/Connections.15.2.01
EcoGarden: The World’s Smartest Interactive Ecosystem. (n.d.). Retrieved November 23, 2018, from https://www.kickstarter.com/projects/120758580/ecogarden-the-worlds-smartest-interactive-ecosyste
Emberlight iot product. [Digital image]. (n.d.). Retrieved from https://portalzine.de/dev/iot/turn-any-bulb-into-a-smart-light-emberlight/
Escobar, A. (2018). Designs for the Pluriverse. Radical Interdependence, Autonomy, and the Making of Worlds. Duke University Press. https://doi.org/10.1215/9780822371816
Fairphone modular, ethical smartphone. [Photograph]. (n.d.). Retrieved from https://shop.fairphone.com/en/?ref=header
Focus heater by Sven Adolf [Photograph]. (n.d.). Retrieved from http://www.sabrinahauser.com/?p=231
Francis, B. (2017, June 28). Building the Web of Things — Mozilla Hacks — the Web developer blog. Retrieved November 6, 2018, from https://hacks.mozilla.org/2017/06/building-the-web-of-things
Fritsch, E., Shklovski, I., & Douglas-Jones, R. (2018). Calling for a Revolution: An Analysis of IoT Manifestos. In Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems (pp. 302:1–302:13). New York, NY, USA: ACM. https://doi.org/10.1145/3173574.3173876
GPS device in older car. [Photograph]. (n.d.). Retrieved from https://www.lifewire.com/smartphone-or-car-gps-1683388
Herrman, J. (2018, September 28). What I Learned from Watching My iPad’s Slow Death. The New York Times. Retrieved from https://www.nytimes.com/2018/02/06/magazine/what-i-learned-from-watching-my-ipads-slow-death.html
Huggies Tweet Pee IoT diaper product. [Digital image]. (n.d.). Retrieved from https://weputachipinit.tumblr.com/image/145920705690
Ikea hacking, unattributed. [Photograph]. (n.d.). Retrieved from https://www.onecrazyhouse.com/17-ikea-hacks/
IoT Design Manifesto 1.0 | Guidelines for responsible design in a connected world. (n.d.). Retrieved December 9, 2018, from https://www.iotmanifesto.com
Jackson, S. J., & Kang, L. (2014). Breakdown, obsolescence and reuse: HCI and the art of repair. In Proceedings of the 32nd annual ACM conference on Human factors in computing systems — CHI ’14 (pp. 449–458). Toronto, Ontario, Canada: ACM Press. https://doi.org/10.1145/2556288.2557332
Karimi, K., & Atkinson, G. (2013). What the Internet of Things (IoT) Needs to Become a Reality, 16. Retrieved November 23, 2018, from http://www.mouser.fr/pdfdocs/INTOTHNGSWP.PDF
Katz, L. (2013, May 8). TweetPee: Huggies sends a tweet when baby’s wet. Retrieved November 23, 2018, from https://www.cnet.com/news/tweetpee-huggies-sends-a-tweet-when-babys-wet/
Leica M8 digital camera. [Photograph]. (n.d.). Retrieved from https://www.35mmc.com/22/08/2016/leica-m8/
Lindqvist, U., & Neumann, P. G. (2017). The future of the internet of things. Communications of the ACM, 60(2), 26–30. https://doi.org/10.1145/3029589
Nest smart thermostat [Digital image]. (n.d.). Retrieved from https://www.williams-sonoma.com/products/nest-thermostat/
Newman, J. (2016, April 21). “Smart Homes?” Not Until They’re Less Dependent On The Internet. Retrieved November 6, 2018, from https://www.fastcompany.com/3059023/smart-homes-not-until-theyre-less-dependent-on-the-internet
Panetta, K. (2017). Top Trends in the Gartner Hype Cycle for Emerging Technologies, 2017. Retrieved November 19, 2018, from https://www.gartner.com/smarterwithgartner/top-trends-in-the-gartner-hype-cycle-for-emerging-technologies-2017/
Paul, F. (2017, November 21). What happens when an IoT implementation goes bad? Retrieved November 20, 2018, from https://www.networkworld.com/article/3238004/internet-of-things/what-happens-when-an-iot-implementation-goes-bad.html
Perlow, J. (2015, December 15). IoT Abandonware: When your cloud service leaves you stranded. Retrieved November 20, 2018, from https://www.zdnet.com/article/iot-abandonware-when-your-cloud-service-leaves-you-stranded/
Perlow, J. (2016, July 12). All your IoT devices are doomed. Retrieved November 6, 2018, from https://www.zdnet.com/article/all-your-iot-devices-are-doomed/
Peters, A. (2016, October 13). Does A Phone That Lasts Forever Have A Future? Retrieved November 6, 2018, from https://www.fastcompany.com/3063872/does-a-phone-that-lasts-forever-have-a-future
Pogue, D. (2013). Death to the Upgrade. Scientific American, 309(3), 32–32. https://doi.org/10.1038/scientificamerican0913-32
Posey, B. (2017). Unintended consequences of IoT device proliferation. Retrieved November 23, 2018, from http://techgenix.com/iot-device-proliferation/
Project Are modular smartphone by Goggle. [Photograph]. (n.d.). Retrieved from
Puzzlephone modular smartphone. [Digital image]. (n.d.). Retrieved from http://www.puzzlephone.com/
Robbins, H. (2017, June 29). RIOT. The State of Responsible IoT. The path for transparency for IoT technologies. Retrieved from https://www.thingscon.com/the-state-of-responsible-iot-2017
Rothchild, J. A. (2014). Net Gets Physical: What You Need to Know About the Internet of Things. Business Law Today, 1–5.
Samsung smart fridge, 2014. [Photograph]. (n.d.). Retrieved from https://www.forbes.com/sites/michaelkanellos/2016/01/13/hold-the-laughter-why-the-smart-fridge-is-a-great-idea/#65dcf4d97d40
Samsung smart fridge, 2016. [Digital image]. (n.d.). Retrieved from https://www.techhive.com/article/3065441/home-tech/samsungs-over-the-top-family-hub-smart-fridge-is-now-on-sale.html
Schindler, H. R., Cave, J., Robinson, N., Horvath, V., Hackett, P., Gunashekar, S., … RAND Europe. (2013). Europe’s policy options for a dynamic and trustworthy development of the Internet of things. Luxembourg: Publications Office. Retrieved from http://dx.publications.europa.eu/10.2759/22004
Stafford-Fraser, Q. (1995, May). Trojan Room Coffee Pot Biography. Retrieved November 23, 2018, from https://www.cl.cam.ac.uk/coffee/qsf/coffee.html
Thingk IoT products [Digital image and Photograph]. (n.d.). Retrieved from https://www.indiegogo.com/projects/thingk-kitchen-devices-for-internet-of-things
TOPP. (2016, January 12). RESPONSIBLE IOT: 3 ESSENTIAL IOT DESIGN FEATURES. Retrieved November 6, 2018, from https://medium.com/the-conference/responsible-iot-3-essential-iot-design-features-504ce4c62e77
Trojan room coffee pot. [Photograph]. (n.d.). Retrieved from https://petapixel.com/2013/04/03/the-first-webcam-was-invented-to-check-coffee-levels-without-getting-up/
Tsaknaki, V., & Fernaeus, Y. (2016). Expanding on Wabi-Sabi as a Design Resource in HCI. In Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems — CHI ’16 (pp. 5970–5983). Santa Clara, California, USA: ACM Press. https://doi.org/10.1145/2858036.2858459
Verbeek, P.-P. (2006). Materializing Morality: Design Ethics and Technological Mediation. Science, Technology, & Human Values, 31(3), 361–380. https://doi.org/10.1177/0162243905285847
Villum, C. (2018, September 14). RIOT. The State of Responsible IoT 2018. Retrieved from https://www.thingscon.com/responsible-iot-report/
Vitali, I., Arquilla, V., & Tolino, U. (2017a). A Design perspective for IoT products. A case study of the Design of a Smart Product and a Smart Company following a crowdfunding campaign. The Design Journal, 20(sup1), S2592–S2604. https://doi.org/10.1080/14606925.2017.1352770
Vitali, I., Arquilla, V., & Tolino, U. (2017b). A Design perspective for IoT products. A case study of the Design of a Smart Product and a Smart Company following a crowdfunding campaign. The Design Journal, 20(sup1), S2592–S2604. https://doi.org/10.1080/14606925.2017.1352770
Vitali, I., Rognoli, V., & Arquilla, V. (2016). Mapping the IoT: Co-design, Test and Refine a Design Framework for IoT Products. In Proceedings of the 9th Nordic Conference on Human-Computer Interaction — NordiCHI ’16 (pp. 1–3). Gothenburg, Sweden: ACM Press. https://doi.org/10.1145/2971485.2987681
Wakkary, R., & Tanenbaum, K. (2009). A sustainable identity: the creativity of an everyday designer. In Proceedings of the 27th international conference on Human factors in computing systems — CHI 09 (p. 365). Boston, MA, USA: ACM Press. https://doi.org/10.1145/1518701.1518761
