Gavin Vaughan and the Ghost Tours of World of Warcraft
Gavin Vaughan is a PhD student in computer science at Durham, working in what he calls speculative computer science. The full title is Designing a Thanatossensitivity Metaverse, which is a death-sensitive Metaverse. His focus is not another product roadmap, but the social machinery that forms around technology when the subject is death, grief, and what happens to data when a person is no longer here.
Vaughan traces his research interest to an article about so called death bots, AI systems built from a person’s data that simulate conversation after death. He is clear about the appeal when a person is alive and the uncertainty when a person is not.
“A lot of my PhD research has looked into what happens with data when we die.”
His work sits at the intersection of digital twins, digital inheritance, and the rituals that may emerge as more social life migrates into platforms and immersive spaces. Vaughan is interested in web3 and what he calls web4 technologies, not as branding, but as an opportunity to reassess ownership, control and consent. The question, as he frames it, is how virtual society should deal with death as a normal condition rather than an edge case. “We need to start thinking about this.”
It is a subject with uncomfortable statistics. Vaughan notes that social platforms offer limited options when a user dies. Accounts can be memorialised, and in some cases families can request closure, but the underlying control sits with the platform. He also references a forecast that, by mid-century, there may be more dead accounts on Facebook than live ones. The detail matters less than the direction of travel. A platform built for presence becomes a warehouse of absence, without the design language to handle what that means.
Rather than staying with social media, Vaughan takes the argument into a different world, one that has been running parallel cultural experiments for decades. World of Warcraft, launched in the mid 2000s, has a long history of in game vigils and memorials for players who have died. In 2007, he notes, a funeral was held inside the game for a player, with more than 100 people attending. The setting was a church within the game world, even though the player was not Christian. Vaughan is interested in what that reveals. Digital ritual borrows familiar architecture and repurposes it, even when the original religious meaning is not shared.
The game also offers a harder lesson. A later memorial service was attacked by the opposing faction, because the rules of Warcraft still apply. The event raises the question that sits beneath Vaughan’s wider research. How should social norms operate inside spaces designed for conflict. Vaughan describes it as an ethical problem rather than a scandal.
“There’s a lot of these ethical questions around how social norms apply in a game.”
That tension has shaped the project he is now developing: ghost tours inside the Warcraft universe. The concept is borrowed from physical tours, which Vaughan attended in York. He is not interested in ghost hunting. He is interested in how tours use place, narrative and history to create meaning. “Ghost tours are not about seeing ghosts.”
He began asking what a digital equivalent might look like. A tour of Facebook memorial pages felt too sprawling and too intrusive. A massively multiplayer online game offered something else: a geography, a set of routes, and a catalogue of people and stories embedded as characters. “Wouldn’t it be great to invite people who play the game, and people who do not play the game, to come visit these places.”
The raw material is extensive. Vaughan believes there are dozens, possibly more than 100, memorialised characters in Warcraft, created by the developers over nearly 20 years. Some are celebrity tributes. Vaughan references a character created to memorialise Robin Williams, because the actor played the game. Others are less famous but not less significant, including characters associated with Make A Wish, and younger players who died from illness. The memorialised characters are usually non playable, which prevents them being attacked and offers a form of protected status inside a combat world. “They’re classed as non playable characters.”
Vaughan’s proposed tours would stitch these scattered sites into a coherent route. The work is closer to a historian’s job than a gamer’s. Locations are spread across a vast map. Names are often known only through forums, old posts, and archived discussion. The game does not present a label that says a character is a memorial. Vaughan describes the research as technical and time consuming. “It would be very sporadic around the internet.”
The format aims to balance accessibility and authenticity. Vaughan plans to livestream the tour for those who do not want to download the game, with separate sessions for the two factions, Horde and Alliance, because players cannot easily cross the boundary. The target duration is around two hours, long enough to travel and explain, short enough to remain watchable. He hopes to run the first tours by the end of March.
The approach carries an ethos, which Vaughan expresses in a single distinction. Many ghost tours trade on trauma, death and sensationalism. His intention is to focus on the person and the story rather than the manner of death. “My approach is going to be more like celebrating their life.”
This ethos is not sentimental. It is a practical design requirement when grief is being handled in public digital space. Vaughan is alert to the ways technology can harm as easily as it can preserve. He describes a case where a deceased woman’s Facebook account was hacked and began messaging people, causing distress. The family could not reach Facebook and eventually had to report the account as spam to get it shut down. For Vaughan, this is not an edge case. It is a predictable failure mode when platforms do not treat death as a core part of account lifecycle.
He extends the point to the question of digital inheritance. If family members cannot control or inherit a person’s accounts, data and digital assets, they are left with a growing burden of unmanaged material. Vaughan uses blunt language about what gets passed down.
“The digital junk that everyone inherits needs to be filtered down.”
His proposed hierarchy is simple. Most of what exists does not need to be kept. Some of it will be important, such as photos, videos, voice notes, and selected messages. From that reduced archive, there may be a smaller set of digital heirlooms that are worth preserving as a legacy. The intention is not only emotional clarity but also environmental responsibility, because storage is physical, and data centres must be powered. Vaughan suggests families should be supported to do this work before death, not after. He references Swedish death cleaning as a cultural analogy, applied to digital life.
The Warcraft ghost tour is therefore not a novelty project. It is a small, tangible experiment in what grief sensitive design might look like inside a virtual environment. Vaughan’s PhD title reflects that ambition. “My PhD title is phantom sensitivity, which is death sensitive.”
In Vaughan’s framing, the gaming context also supplies useful language for governance. In multiplayer games, griefers are people who enter a space to disrupt and destroy. The term is unrelated to bereavement, but it overlaps in an unsettling way with the vocabulary of grief. Vaughan finds that collision instructive. It shows how communities invent concepts for harm, and how those concepts might need to change when the harm is emotional rather than merely disruptive.
The longer arc of his work is towards systems that build off boarding into the same logic as onboarding. He argues that technology companies have spent decades optimising user growth, while paying far less attention to dignified exit. “How do we start getting companies to think about off boarding processes.”
That framing gives the Warcraft tour its place in a broader research programme. It is a guided walk through a landscape of memorials, but it is also an early rehearsal for a future where platforms and virtual worlds will need policies, rituals, and interfaces that do not collapse under the weight of dead data. It tests what people want from remembrance when the setting is interactive, social, and governed by rules designed for play.
Vaughan’s instinct is to keep the language accessible, even when the ideas are complex. Academic papers may land with a small audience. A livestream tour may reach people who would never pick up a journal article, yet still recognise the problem immediately. As he puts it, the aim is to make people think differently, while there is still time to design the next iteration of the web with death in mind.
