X-Clacks Overhead: Making sense of death in the digital age

A man is not dead while his name is still spoken

Sir Terry Pratchett, in Going Postal


In the context of Digital Civics, we often come across the notion of “wicked problems” [14]. Wicked problems is a way of defining those parts of life that are too complex to be easily broken into parts and solved away. The concepts that are hard to define, hard to quantify, and which are laden with societal inter-dependencies. The changing face of death and bereavement in the digital age is, in my opinion, a wicked problem.

Over the course of our introduction to HCI over the last several months, we have delved into some broad discussions around the implications of technology and the role of HCI in understanding and addressing them. In some cases, the introduction of new technology has resulted in short term improvements in convenience or efficiency, but much longer term implications for the way that we live our lives. For this final blog post, I would like to focus on the way that some of these HCI topics can be related to death and dying, which is a crucial and inevitable part of human culture. I will consider how ubiquitous computing and social computing have altered our behaviour with the world and with each other, and our massive generation and consumption of data has altered and augmented our existence and identities. I will discuss the deeply personal, experiential and somewhat taboo nature of death and loss, and how this, like the rest of society, may change with the introduction of technology. Finally I will discuss death as a design space, and the positioning of Digital Civics in promoting civic discussion, co-design of social innovation and development of citizen-based interventions around bereavement, palliative care and digital legacy.

Ubiquitous computing

In discussions of ubiquitous computing, the works and predictions of the late Mark Weiser [13] are foundational concepts around which the research subject of Ubicomp are formed. His insight approaching the end of the 20th century was of seminal importance, largely addressing the concept that technology, as it becomes more sophisticated, should aim to fade into the back of our awareness. We should be able to use technology, or be aided by it, without explicit awareness that we are doing anything unnatural or complex. He expresses this very eloquently using the example of Sal, a fictional character who goes about a typical day making use of different forms of technology as an integrated part of her normal life. Many of the technologies used by Sal now exist in real life and are used just as seamlessly. Many more now-existing technologies did not feature in the story, partially because it was intended as an example and not an exhaustive prediction of the future, but also because forecasting the way that people will appropriate and make use of new forms of technology is a hopeless task – it is itself a wicked problem.

In an earlier blog post, I discussed a paper by Gregory Abowd [1], in which the argument is made that ubicomp as Weiser predicted it has been realised and research needs to move on to the next frontier. At the time of writing, I was very cautious – perhaps overly so – to draw conclusions of my own in this respect. With time, however, I have begun to regard Abowd’s eagerness to move on to whatever comes next with unease. It seems now to be a dismissal of the implications of what ubicomp has created. In the space of a few short years, humanity has moved swiftly on from the old ways – the product of aeons of evolution and gradual societal growth – to a new, interconnected world which has been developed by a select few and for which there is no system of removing or rolling back changes that may ultimately turn out to be damaging.

Social computing

One of the ways that ubiquitous computing has manifested itself is through social computing. The development of mechanisms of contact that allow instant and global communication has reinvented and redefined social interaction for the digital age. In my second blog post, I discuss some of the complexities of social computing from the viewpoint of an external researcher. One feature of social networks and content aggregation websites is that it is possible for interactions to take place between two or more individuals without each contributor having particularly rich detail of who or where the other is. As a result, the opportunities for social interaction are vastly more numerous, yet the depth of those relationships and actual human connectedness may be lessened.

The implications of major social computing sites such as Facebook, Twitter and Reddit are broad and highly complex – another wicked problem. The way that we use such platforms to have serious discussions and share opinions on controversial subjects is fundamentally different from the way that we use speech. For example, one research study looked at everyday socio-political discourse around “poverty porn” reality television programme Benefits Street [2]. It highlights some of the characteristics of the use of Twitter to discuss this highly controversial topic, including the different nature of tweets that were sent during broadcast (primarily abusive towards the subjects of the programme) to those between broadcasts (more likely to defend the subjects or debate some of the wider political issues). The authors look in particular at the discourse produced during “second screening”, i.e. the act of consuming televisual media and using social media simultaneously as a way of producing and accessing commentary on it. Another paper by Starbird and Palen [10] examines the use of retweeting during the 2011 Egyptian political uprisings, concluding that the use of social media to participate in such events in a reduced-risk fashion – labelled by some as slacktivism – may in fact be a productive way for people to contribute to social movements.

The use of social media and other digital technologies to create identities for ourselves that complement, extend or displace our ‘true’ identities presents another problematic aspect of modern computing: how much does our data define us, and how much control over it do we have?


The fact that owners of online applications and platforms have an interest in our data should come as a surprise to no one. While it is possible that you have ventured onto the internet for the sole purpose of reading my blog post, it is much more likely that you are one of the 90% of UK adults who have accessed the internet recently, and you will therefore have had the experience of endless pop-up dialog boxes, emails and compulsory-to-accept privacy statements outlining that:

  1. Your privacy is important to us
  2. We will use your private data to target you with advertising

We are all unique and intricate beings, but what we do and say through ubiquitous and social computing goes a long way in defining us for others. Most of the things we do via modern technology leave a digital footprint, the scope of which has been described quite reasonably as “utterly mindboggling” [11]. It includes not only data that is provided actively, i.e. willingly and knowingly, but also passively as data exhaust, left behind through cookies or mobile network usage [4]. Digital footprints would be sufficiently mindboggling even if technology were to stand still, yet the reality is that new technologies do and will continue to emerge that perpetually change the extent to which our data can be said to represent us. To muddy the waters further, our digital actions and interactions do not necessarily exist on a fixed, linear timeline in the way that we are used to. If you were to seek information about an individual on the internet, you are likely to find a collection of sources presented in no particular order, all viewed together in the present despite representing different pieces or stages in that individual’s chronological life story. For example, Haimson et al [5] consider the perspective of transgender people, who often take active steps to separate themselves from online identities that were constructed prior to their transition, and which they do not view as properly representative of themselves in the present. The authors conclude that there is a need for designing for forgetting, as well as designing for remembering. This idea, while being particularly true for individuals with a clear life transition of some kind, may equally be true of any person whose thoughts, characteristics and opinions have developed over time.

Death, bereavement & digital legacy

In my third blog post, I argued that those of us who participate in the digital world are creating a much more rich, transparent and pervasive personal history through our online actions than those who came before us. In doing so, we are exposing ourselves in depth to not only other individuals sharing that digital world, but also generations to come. When we eventually die, much of our identity will continue to exist, and however much of our personalities we allow to show through our digital activities will remain visible to those who have an interest in seeking it out. This presents some interesting questions about how valuable such data is as a resource, what are the rights of the deceased in relation to data about them, how to make plans for our own digital legacy while we are still alive, and how can digital technology be applied to assist people in coping with their own end-of-life transition and with the passing of loved ones.

An option to set your Facebook legacy contact

Sandi Varnado, in a 2014 review relating specifically to the laws of the State of Louisiana [11], does an admirable job of breaking down the digital footprint as it relates to death into smaller, more digestible chunks. In this review, digital legacy is broken down into items of sentimental value and items of financial value, each characterised as accessible through email as a kind of master key. Examples of sentimental items include photographs, videos, documents, books, music and other such collections, as well as social media accounts and video game characters. Financial items can include not only services like Paypal or online banking, but also blogs, social media followers, web domains, intellectual property, mailing lists, and, increasingly often, video game characters and items. These are examples that are relevant now – or, at least, in 2014 – and which form the beginning of a list that will very likely grow over time to include types of possessions and data that we are presently incapable of predicting.

Death is a sensitive subject

The topics of death and loss are quite often avoided in everyday life. We are all aware of our own mortality, and in general conversation it is mostly considered socially unacceptable, even taboo, to bring it up. We feel uncomfortable talking to others about grief and loss, and often fail to do so in the absence of knowing the ‘right’ thing to say. Given the way that we skirt around the subject, you might be inclined to forget that our death is one of the most significant events in each of our lives. This avoidance of the topic can result in difficulties, both emotional and legal, when trying to carry out the will of the deceased. It is already difficult, even with plans in place and when dealing with issues that as relatively cut and dry as physical or financial possessions. When we extend this dilemma to include components of our digital legacy, the difficulty is magnified.

If you were to die today, is there anyone who would have comprehensive knowledge of your online activities or possessions? Do you have financial assets that you control and access via the internet, with plans in place for these to be passed to friends or family?  Have you left instructions that would allow people to access your photos, videos or cloud-stored documents? If a loved one were to seek access to a password-protected online account of yours after your death, would you see this as an invasion of your privacy or a liberation of your digital legacy? What do you want to happen to your social media accounts – should they be closed down or memorialised? Do you have online friends or communities who you would like notified of your passing? These are all questions that must be asked, and yet all too often we wait until it is too late to get the answer.

Death as a design space

The questions and discussions around dying in the digital age as I have presented them above largely relate to the practical. In doing so, I have so far taken an approach similar to that laid out by Oulasvirta and Hornbæk in HCI Research as Problem Solving [8]. In their 2016 paper, they propose a model to break down HCI research into empirical problems (relating to real-world phenomena in human computing), conceptual problems (relating to theories, principles and models) and constructional problems (relating to understanding the construction of interactive artefacts). However, while the latter of these three groups does include design, this problem-solving approach is limited in its consideration of problems that exist or that will come to exist in the future. There is of course a place for this kind of pragmatic approach (and I think it should be the main approach), but there is also a place for design around death and technology which has its sights set on sense-making, rather than problem solving.

The title of this blog post, “X-Clacks Overhead”, refers to a famous digital tribute to author Sir Terry Pratchett, who died in 2015. In his 2005 novel, Going Postal [9], Pratchett had included a plot based around a system of semaphore telegraph towers called The Clacks, which are a satirical take on the internet. In the novel, a message was sent along the “overhead” (or the metadata) of The Clacks containing the words “GNU John Dearhart” – G being an instruction to pass on the message to the next tower, N meaning to not log the message, and U meaning to return the message when it reaches the end. Sent as a memorial to the character by his father, John Dearhart’s name would resultantly be passed around the Clacks for as long as the system was in operation. Since his death, the message “GNU Terry Pratchett” has been silently included in the header of all manner of internet data packets (see here for a list of public-facing websites broadcasting the signal https://xclacksoverhead.org/listing/the-signal). This is a touching memorial to Sir Terry, but it also represents a way of remembering a person that only makes sense in the digital age. It is not strictly functional, nor does it solve a problem, but it has inherent meaning in the context of the passing of a beloved writer.

GNU Terry Pratchett, ASP.NET version

In their book Speculative Everything, Anthony Dunne and Fiona Raby [3] discuss the place of speculative design, which aims to trigger creative thought, discussion and critique of possible futures, rather than remaining within the probable or strictly plausible design spaces. In doing so, they argue that we can ask the question “what if?”, and allow consideration of what the implications are of various technological or social developments, regardless of their likelihood of actually taking place. They also promote the design of objects or artefacts that do not have value as marketable products, or which are at odds with consumer demand, such as their Statistical Clock, an electronic device that speaks aloud the number of fatalities from transport accidents as and when they are reported in the news in order to remind the owner of the frailty of life. No person is likely to want such a device, but its creation sparks a dialogue around what is and is not acceptable, and what, to an extent, we already have access to via internet news media.

The Statistical Clock, by Dunne & Raby [3]

Innovations in how digital technology is or should be used in the context of death and bereavement are unnecessarily limited if our expectations are built entirely upon how existing technology can be implemented in traditional ways. I would argue that speculative design confronting new ways that technology could be used in this area has a valuable place alongside affirmative design practices that seek solutions for problems that already exist and have been identified.

Digital Civics

So, how does this all tie in to the principles of Digital Civics? One of the core viewpoints of Digital Civics and the Open Lab is that research is done with people, rather than to people. The topic of death is one that relates to all of us, and the way that we go on to implement technology to manage it will depend on the actions of the wider population as much as expert designers. In his book Design, when Everybody Designs [6], Ezio Manzini discusses how modern information technology has increased the quantity of individuals who are likely to produce solutions in the face of complex problems. He talks of social innovation – new societal ideas and models – as a way to address intractable (see: wicked) problems in society, and the balance of expert design and diffuse design in driving innovation. He characterises co-design processes as taking place within designing networks built up experts and non-experts designing together.

The Digital Civics research agenda reflects this style of coproduction [7,12]. But, how do we go about establishing such a designing network around the topic of death? I believe that one aspect of conducting research to that effect is to encourage citizen engagement and discussion around the topic. With civic discussion and coproduction between citizens and expert designers, we can begin to establish a picture of what, in an ideal world, digital death and legacy might look like. With such a foundation in place, people may be drawn together to innovate solutions for the many constituent parts of this problem – in all its wickedness.



[1]         Gregory D. Abowd. 2012. What next, ubicomp?: celebrating an intellectual disappearing act. In Proceedings of the 2012 ACM Conference on Ubiquitous Computing – UbiComp ’12, 31. DOI:https://doi.org/10.1145/2370216.2370222

[2]         Phil Brooker, John Vines, Selina Sutton, Julie Barnett, Tom Feltwell, and Shaun Lawson. 2015. Debating Poverty Porn on Twitter: Social Media as a Place for Everyday Socio-Political Talk. In Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems – CHI ’15, 3177–3186. DOI:https://doi.org/10.1145/2702123.2702291

[3]         Anthony Dunne and Fiona Raby. 2013. Speculative everything : design, fiction, and social dreaming.

[4]         Fabien Girardin, Francesco Calabrese, Filippo Dal Fiore, Carlo Ratti, and Josep Blat. 2008. Digital Footprinting: Uncovering Tourists with User-Generated Content. IEEE Pervasive Computing 7, 4 (October 2008), 36–43. DOI:https://doi.org/10.1109/MPRV.2008.71

[5]         Oliver L. Haimson, Jed R. Brubaker, Lynn Dombrowski, and Gillian R. Hayes. 2016. Digital Footprints and Changing Networks During Online Identity Transitions. In Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems – CHI ’16, 2895–2907. DOI:https://doi.org/10.1145/2858036.2858136

[6]         Ezio Manzini. 2015. Design, when everybody designs: an introduction to design for social innovation. The MIT Press, Cambridge, Massachusetts.

[7]         Patrick Olivier and Peter Wright. 2015. Digital civics: taking a local turn. interactions 22, 4 (June 2015), 61–63. DOI:https://doi.org/10.1145/2776885

[8]         Antti Oulasvirta and Kasper Hornbæk. 2016. HCI Research as Problem-Solving. In Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems – CHI ’16, 4956–4967. DOI:https://doi.org/10.1145/2858036.2858283

[9]         Terry Pratchett. 2005. Going postal. Bloomsbury, London.

[10]       Kate Starbird and Leysia Palen. 2012. (How) Will the Revolution be Retweeted? Information Diffusion and the 2011 Egyptian Uprising. (2012), 10.

[11]       Sandi S. Varnado. 2014. Your Digital Footprint Left behind at Death: An Illustration of Technology Leaving the Law Behind. La. L. Rev. (2014), 719.

[12]       Vasillis Vlachokyriakos, Clara Crivellaro, Christopher A. Le Dantec, Eric Gordon, Pete Wright, and Patrick Olivier. 2016. Digital Civics: Citizen Empowerment With and Through Technology. In Proceedings of the 2016 CHI Conference Extended Abstracts on Human Factors in Computing Systems – CHI EA ’16, 1096–1099. DOI:https://doi.org/10.1145/2851581.2886436

[13]       Mark Weiser. 1991. The Computer for the 21st Century. Scientific American. 265, 3 (1991), 94–104.

[14]       Peter Wright and Patrick Olivier. Digital Civics and the Cities Challenge. Open Lab, Newcastle University.


Author: Jack Holt

Leave a Reply

Your email address will not be published. Required fields are marked *