Reverie Studio’s Abstraction Approach
Editorial note: This article was originally published in the Knutpunkt 2026 book Seasons of Larp. It has been reprinted from there with the editors’ and authors’ permission. It has not been edited by Nordiclarp.org.
Character design in larp is a rich artistic environment with many interpretations and approaches. Every design team confronts the same challenge: how to create characters that are rich, playable, and emotionally resonant without building systems that struggle with ordinary larp-runner challenges such as participant drop outs, or designing characters that are unsatisfying to play. This article traces Reverie Studio’s evolution on how to deal with these challenges and our current design philosophy used for Fracture (USA, 2023) and to prepare for Neon Void 1984 (USA, 2026), with a participant-driven technological character system: Abstraction. The goal is to create a procedural yet expressive framework where each participant authors their own way into the world while maintaining a cohesive, thematically guided tone.
In our earlier work at Reverie Studios, we wrote fully scripted characters with complex webbed ties, often drawn from the Nordic and American larp traditions. Each sheet could span multiple pages, with descriptive traits, light and dark personality aspects, pre-written questions for reflection, and a network of five or more pre-designed connections to others. It was labour intensive on the writing and design side to build the systems for these characters, and brittle in play if any of the written elements didn’t align for the participant or if there were participant drop outs. If they didn’t like the character as written, we encouraged participants to change whatever elements suited them, so long as they communicated to us or other participants if it would likely affect the overall story or connections. To us, that felt more like a bandage for a design flaw we hadn’t quite addressed.
As we grew our practice through larps like Eskhaton (USA, 2019) and The Haunting of Gray Manor (USA, 2020) we realized that the traditional “full-sheet” model was not only a workload problem but a design problem. It assumed a level of stability that large-scale larp may not deliver. When participants drop out, ties break; when people role-play differently from what was written, narrative scaffolds can collapse. And most importantly, too much prescriptive writing can prevent the participant’s sense of ownership. Our emerging focus became resilience: creating character systems that survive disruption, thrive in participant hands, and scale sustainably for designers.
Our first opportunity to put this into motion started when we collaborated with Juhana Pettersson in 2020 on designing Saturnalia (USA, 2022). We proposed Saturnalia, a city-wide World of Darkness larp in New Orleans’s French Quarter for about 200 participants, and we needed to solve a design challenge to allow those 200 participants to make meaningful characters that were as fully fleshed out as the long form sheets we traditionally produced. The approach we took with Juhana built on the work he did on other projects such as Enlightenment in Blood (Germany, 2017) in which he used a software tool called Larpweaver (created by a Texan company, Incognito Limited). As Pettersson discusses in the article ‘The Death of Hamlet – Deconstructing the Character in Enlightenment in Blood,’ about why he used this system, “the method of larp organizing where each participant is provided with a written character is a lot of work, especially in big larps” (Pettersson, 2022).
Because our design goals differed from those for which Larpweaver was designed, we built our own tool for Saturnalia, incorporating feedback from Juhana, and approaching character writing with a new mindset. The tool worked, and we successfully ran the larp, managing the around 200 participants who all made characters, with the normal problems of drop outs and fills being mitigated without arduous recasting. After Saturnalia, we kept using and improving the tool instead of abandoning it, explored other applications for it, and named it Abstraction.
Abstraction
We use the term “abstraction” for our character creation system to represent multiple layers of problem solving for larp character writing. When we use it, we mean it is accomplishing a reduction in labour through software development, but also in an artistic sense. Abstraction aims to distill character creation on both the writing and participant side. We create an intermediary form between larp traditions where participants make characters using, for example, a point buy system, and more structured larps where the larpwrights deliver fully realized characters. The approach we use should provide customization to participants to tailor a character, while also reducing the workload on writers.
The system uses an application portal approach for larp designers and partici-pants, loading in writing assets in containers that are connected through relationship mapping, and checked out via a shopping cart approach. We can assign limits on the assets when needed to create uniqueness for selection and make selection contextual. For Saturnalia, we had character abilities that appeared as choices depending on the vampire clan that was selected. By design, only a few characters could select a certain attribute (e.g., older vampire). Similar to a “choose your own adventure,” participants follow the decision tree, adding choices that branch based on selection, and eventually arriving at a unique character all their own. When they complete the character, it locks the selected assets so others, if unique or limited, can not select them, and generates a character sheet the participant can save or print.
For Fracture we used a hybrid method, allowing participants to select character ‘seeds’ (such as Overburdened Camp Administration) which had some of our longer form approaches for writing like a short history and pre-written relationships, but allowed participants to add additional layers as they built the character to expand this seed in a direction that is enjoyable for them. In post-mortem we determined that our future larps, such as Neon Void 1984, would break from the pre-written method as it was still labour intensive and participants enjoyed the options they could select that were more custom for their experience based on responses we received in our post-event survey.
For Neon Void 1984, we are adding a layer of complexity to the tool: generating social connections without involvement from the larp designers. In our previous larps, we wrote these relationships for each character to create meaningful points of connection to other characters. The degree to which these written relationships manifested in participant experience depended on:
- the quality of the writing of the relationship (since authors often matched characters out of necessity rather than intentionality, not all written relationships displayed the same level of quality).
- whether the participant wanted to en-gage with the relationship as written (did the participant get any satisfaction from the writer’s interpretation of the character they are portraying).
- how the participant portrayed the charac-ter (a relationship as written differs from how the participant portrays it).
It was also the most labour-intensive aspect of character writing, requiring many revisions. We are looking to address these concerns with character relationships.
Our proposal is to write categories of social connections, similar to other selectable assets in the character creation system, representing typical character relationships. When a participant chooses a relationship type, they express their desire for that kind of relationship, even though they don’t yet know the details, such as who it is with, how it started, how important it is, etc. (participants are free to determine these details collaboratively). Once all characters have been created in the system, the larp designers engage in a process that will sort and match the choices and make assignments that provide the traditional relationships and add a social component that can be explored before the larp or during workshops.
This kind of relationship assignment will require careful algorithm design. The algo-rithm should be equitable, fair, and highly configurable. By equitable, we mean that there should be no preferential treatment of one participant over another. A system where participants choose their matches might overwhelm popular larpers with unwanted attention while excluding new larpers.
By fair, we mean the algorithm should try to assign relationships as closely to what the participants requested as possible. Ranked-choice voting for relationship types (not the individuals in the relationship) is an appealing choice as it should help ensure that everyone gets at least some of what they want. For example, imagine the unlikely scenario in which only one larp participant has requested a romantic relationship. Here, there is no one for the participant to be paired with, so they cannot get that relationship. If this participant selected “romantic” as their top relationship pick and “rival” as their second highest, the algorithm should ensure that it satisfies their second pick, because it could not meet their first.
The algorithm, when complete, will be highly configurable, similar to the rest of Abstraction, so it can adapt to the needs of individual larps. A larp may want to limit the number of a specific type of relationship for narrative reasons (for example, an academic larp where one professor is being blackmailed has a unique tone from a larp where all professors are being blackmailed). Other larps and other types of relationships could be unlimited (for example, every student can have a friend and a rival). Some relationships, like friend circles and love triangles, can involve over two characters, so relationships don’t always have to be between two people. The algorithm should be flexible enough to support such designs, fairly distributing any limited relationship types using the ranked choice system above.
With Neon Void 1984, our intention is to implement the mentioned systems. Participants will select categories that act as sliders or levers: calibration options to fine-tune the character they are making and offer linkage points to other characters or stories of interest. Then, they receive a complete character as a result of the choices they make in the system. This character is custom to the participant and has enough story elements from the choices they make to match the long-form characters we used to write as a studio. They will also have connections to other characters that are in alignment with the play experience they’d like to have.
If drop outs occur, which our experience shows happens approximately 20-25% of the time per event, we believe Abstraction‘s design approach will mitigate the impact on both organisers and participants. An incoming participant that is filling a spot can join the event and get a customized character that is built to their spec, and can onboard through the same systems, including the relationship algorithm running again to make a connection, without a serious loss of fidelity of their experience. When the process was manual in the past, the existing characters and relationships restricted the incoming participant to which characters that were available. From a design perspective, Abstraction‘s systems are preferable, making casting more straightforward, and allowing the incoming participant to design a character as anyone else. We don’t have to design a fully written long-form character at the beginning and worry about a loss of experience for anyone if their relationships don’t show up.
Challenges
While the Abstraction system offers some advantages, it also raises new design challenges, which we raise below. In developing this system, we are not proposing that it replaces any existing approach to character design, or that it is superior. We developed this system out of our own necessity to reduce labour in writing and design of characters, to give participants more options for what their character is for their own experience in the larp, and to deal with some challenges associated with character casting (such as when a participant drops out). In our experience on the design side, it is a successful project, and participant surveys reflect satisfaction with the outputs, but we acknowledge challenges that exist.
Depth versus Modularity
Can modularly assembled characters evoke the same emotional depth as authored ones? Some participants thrive on co-creation, while others miss the curated poetry of a hand-written role. Balancing expressive richness with scalable systems remains a key artistic tension. In the past, our writing efforts have focused on sequential steps for character development that incorporate participant feedback and desires. Using a survey approach, our first pass at a character design asked participants to describe broadly the idealized character. The writers used these inputs to create a draft output, and the participant accepted or changed it. The outcome was a written character semi-built to spec for the participant, which many participants enjoyed (based on post-event surveys), but we felt we could offer a similar level of customization while avoiding the level of labour required to use this approach.
For our purposes on the design side, Abstraction offers a more streamlined approach to this process, and a more customized experience for participants. Whether it is as good as a traditional long-form character as we would have written in the past or not, we do not have any hard evidence beyond participant surveys, which suggest participants enjoy the system. There are other approaches to character design such as collective larps with cocreative design, black-box larps with minimalistic characters, etc. But for the kinds of larps our studio runs (which are larger-scale ‘blockbuster’ style larps in a longer form style) the tool we developed offers the labour savings we desire and a solution to participant desire for a meaningful character. Also, we think the tool can scale down to accommodate other formats, such as fest larps, enabling participants to develop characters quickly in an intermediated format.
Technical Accessibility
Software introduces opportunities, but also maintenance costs. Not all design teams can sustain platform development or infrastructure. Our long-term goal is to make Abstraction open-source or hosted for community use, allowing other designers to build upon our framework without heavy technical lift. Using Abstraction comes with a trivial financial cost. We estimate that if others were to use it, we might charge a nominal fee for use, primarily to cover hosting and maintenance/development. We do not expect the price point to be a barrier; however, the use of the system requires adapting the design and using a system that may be unfamiliar. A potential larp organiser wanting to use Abstraction would have to be educated on the system to load in their assets and similarly educate any participants on how to use it, which are additional steps to take in running the larp. As we have not yet worked with any larp designers other than Juhana on this, we do not have an estimation of how difficult using this system will be and how long it will take for other designers to become familiar with it.
Cultural Adaptation
Different larp cultures have different relationships to authorship. Nordic and American scenes vary in how much agency players expect. Clear onboarding and designer education must accompany Abstraction‘s use to align expectations on the designer and participant sides. Conceptually, the system is a break from some traditions. Since there’s so much variability in character development, the storytelling approach for a larp must align with the character design. The system must allow players to select a unique character seed if the story or plot of a larp is tied to a specific character. The system must also account for the character’s potential modularity in other aspects, such as personality, relationships, or other potential changes that are selected during creation using Abstraction‘s ‘choose your own adventure’ approach.
Algorithmic Ethics
Introducing procedural matchmaking raises ethical concerns around consent, representation, and perceived bias. It will be essential to have mechanisms for participant review or veto, along with transparency about how relationships are assigned. The approach of being equitable, fair, and configurable means solving some potential sorting and matching problems, but may introduce others. We balance this and allow characters to belong to selectable relationship groups, and also state that participants can develop relationships on their own, and that procedural matchmaking is only meant to be an additional tool for easier onboarding into the larp.
Comparisons
We recognise other tools exist, but we designed ours for our specific circumstances and design needs that have grown organically with us as a studio. As we developed as a studio, we stopped looking at characters as a static sheet to be written and made up of thematic components that participants bring to life, and instead as a dynamic engagement engine. We want participants to feel they have the agency to build a character they like within a framework, to save labour on the studio side in writing by focusing on core themes we want to see in a larp – as opposed to working them narratively into the sheets, and making sure that those core themes are present, even if there is no specific character that must be the catalyst for that theme. For Neon Void 1984 we intend to have Soviet infiltrators, but it does not matter which character is the infiltrator, and who ends up picking that trait is not relevant to our design approach. We believe that Abstraction fits those criteria, and furthermore we believe it can be a useful tool for other studios to consider using to solve similar problems.
There are other character creation systems and methodologies available in the larp tradition that achieve some of these goals. We cannot speak to all of them, but there are likely some that fit better with what we are looking to achieve than others. For instance, in ‘The Mixing Desk of Larp’ [The Mixing Desk of Larp, 2019] there is a pedagogical framework for larp design which uses the concept of ‘faders’ like those on a mixing board which is useful as a basis for the theory of what we are looking to achieve with Abstraction, but there is not an accompanying software application to use which would achieve the results we desire as an output. Other approaches for character creation that are lightweight on the design side and offer meaningful participant involvement, such as the co-creation method used for Lumen: The Journey to Hope (Italy, 2025) are inspiring, but for our purposes we are aiming to provide pre-larp creation and on-site calibration, and are not basing our larps on Jungian archetypes – so this is not the ideal fit for how Reverie approaches larps or why we designed Abstraction.
There are some technological tools that fit most of the criteria, and currently offer more features than we do, such as LarpManager (LarpManager, 2025) which is the closest analog to Abstraction. The approach Larp-Manager takes as a system differs from the one we propose for design, however. LarpManager’s features are impressive, including a payment and registration system which Abstraction currently lacks, but it is a system more for selecting a fully designed character, and would work well for larps featuring long-form character sheets or short-form characters paired with analog solutions for achieving similar results to how we designed Abstraction.
We do not propose to replace the methods used by other larp designers, as Abstraction was developed as part of an ongoing exploration of our own needs. Nonetheless, we think that making it available to other larp designers could offer benefits in terms of labour savings and participant satisfaction.
Future Enhancements
Looking ahead, we would like to expand Abstraction to include:
- Analytics and feedback loops: giving designers insight into which combinations lead to stronger partici-pant satisfaction. These tools can help design, but also give participants some agency in the choices that are most interesting to them. Long term, this can give better insight into the satisfaction of participants, which leads to better design.
- Integration with other larp infrastructure: such as registration systems, casting databases, and debrief tools. Currently, many of these systems are ad hoc and exist in many forms that are not connected except through manual labour by organisers. Centralising them will make a larp using Abstraction a more holistic system. It would be ideal to offer a one-stop place for all aspects of a larp from registration and payment to casting, but at this time Abstraction fulfills only part of those needs.
- Enhanced UI and UX elements for both participants and designers: include the ability to import/export assets, allowing for plug and play for designers, and templated character sheet exports so that participants have a nicer looking character at the end of their creation.
- Availability for larp designers: at the time of this article, the tool is still in active development for Neon Void 1984, but we intend for it to be available for use by other larp designers in the future. We are working to build the tool to a point where it is approachable for designers and participants. Larp designers can contact us (contact@reverie.studios) to request early access, but it would be on an alpha-testing basis. When it is in a state we feel comfortable releasing, we plan on making it available for development and use with a community announcement. We hope to have this completed by 2026.
Conclusion
Resilience in larp design is not simply a logistical concern; it is an artistic philosophy. By treating uncertainty as a design parameter rather than a failure state, we can build larps that thrive on emergence and intentional desire. In developing this tool, we aim to enable participants to have more input in the co-creation opportunities. Pettersson states, “As a player, those moments of design-enabled reciprocity are beautiful: Someone makes the larp work for me and I make it work for someone else” (Pettersson, 2021, p.258). The Abstraction system represents one pathway toward that future, where characters are not static blocks of text but dynamic systems capable of developing with their participants. They can fulfill their desires and develop a framework for collaborating and co-creating with other participants as the larp designers in Abstraction empower them to weave together.
We are working towards building a space where larp designers can find their own way to iterate on this model, bringing additional freedom and flexibility to the design, and increasing participant agency as co-creators. We hope that this approach can help make large-scale larp more sustainable and creatively empowering. In the same way that larp itself decentralizes narrative control, resilient character systems decentralise authorship, transforming the designer’s role from writer to architect of possibility.
Bibliography
Nordic Larp (2025) ‘“’The Mixing Desk of Larp.’”’ Available from: https://www.nordiclarp.org/wiki/The_Mixing_Desk_of_Larp [Accessed 9 December 2025].
LarpManager (2025) ‘“’Larp Manager – The free LARP management platform.’”’ Available from: https://larpmanager.com/ [Accessed 9 December 2025].
Lumen: The Journey to Hope (2025) “Characters.” Available from https://lumenlarp.org/index.php/characters/ [Accessed 9 December 2025].
Pettersson, Juhana. ‘The Death of Hamlet – Deconstructing the Character in Enlightenment in Blood.’ Available from: https://www.nordiclarp.org/2018/01/30/the-death-of-hamlet-deconstructing-the-character-in-enlightenment-in-blood/ [Accessed October 20, 2025].
Pettersson, Juhana. (2021) Engines of Desire: Essays on Larp As the Art of Experience. Finland: Pohjoismaisen roolipelaamisen seura.
Ludography
Enlightenment in Blood (2017) Pettersson, J. and Pedersen, B., Germany, Participation Design Agency
Eskhaton (2019) Kopczynski, G., Pickering, K., and Brown, V., USA, Reverie Studios
Fracture (2023) Eaton, C., Kopczynski, G., Pickering, K., and Hansen, T., USA, Reverie Studios
Lumen: The Journey to Hope (2025) Italy, Erasmus+ EDGE Consortium
Neon Void 1984 (2026) Kopczynski G., Pickering, K., Eaton, C., and Hansen, T., USA, Reverie Studios
Saturnalia (2022) Pettersson, J., Kopczynski, G., Pickering, K., Eaton, C., and Hansen, T., USA, Reverie Studios
The Haunting of Gray Manor (2020) Kopczynski, G., Pickering, K., and Brown, V., USA, Reverie Studios
This article is republished from the Knutpunkt 2026 book. Please cite it as:
Kopczynski, G., Pickering, K., and Metzger, M. 2026. ‘Designing Characters for Resilience.’ In Seasons of Larp eds. Erlandsson, A., Svanevik, M., Holkar, M., and Brind, S. Gothenburg. Knutpunkt.
