Source-linked AI summary
Virtual World, Defined from a Technological Perspective, and Applied to Video Games, Mixed Reality and the Metaverse
Kim J. L. Nevelsteen
TL;DR
Virtual-world definitions lack consensus and may not fit contemporary technologies, making architecture design problematic. The paper uses grounded theory to derive a technology-applicable definition, then reports an ontology, property-based classifications, and pseudo-persistence categories. Its scope excludes culturally defined game-world persistence and treats immersion as qualitative rather than determinant.
Problem
Virtual-world definitions lack consensus and may not apply to contemporary architectures and technologies, complicating the design of supporting architectures.
Method
The paper samples technologies using grounded theory to derive properties and a definition directly applicable to technology.
Results
The paper produces a detailed definition, validates it against existing definitions, and uses an ontology and property breakdown to classify related technologies.
Takeaways & Limitations
Pseudo-persistence categorizes technologies that mimic persistence, while the ontology clarifies relationships among virtual-world terms and acronyms.
Takeaways & Limitations
The paper excludes culturally defined game-world persistence and treats immersion as qualitative rather than a determinant property.
Abstract
from arXiv · showhide
There is no generally accepted definition for a virtual world, with many complimentary terms and acronyms having emerged implying a virtual world. Advances in systems architecture techniques such as, host migration of instances, mobile ad-hoc networking, and distributed computing, bring in to question whether those architectures can actually support a virtual world. Without a concrete definition, controversy ensues and it is problematic to design an architecture for a virtual world. Several researchers provided a definition but aspects of each definition are still problematic and simply can not be applied to contemporary technologies. The approach of this article is to sample technologies using grounded theory, and obtain a definition for a `virtual world' that is directly applicable to technology. The obtained definition is compared with related work and used to classify advanced technologies, such as: a pseudo-persistent video game, a MANet, virtual and mixed reality, and the Metaverse. The results of this article include: a break down of which properties set apart the various technologies; a definition that is validated by comparing it with other definitions; an ontology showing the relation of the different complimentary terms and acronyms; and, the usage of pseudo-persistence to categories those technologies which only mimic persistence.
1 Introduction
The paper addresses the lack of an applicable, generally accepted definition of a virtual world amid competing terminology and changing network architectures. It uses grounded theory to derive a definition, validate it, and classify contemporary technologies.
- Research problem: Existing virtual-world definitions lack consensus and cannot be applied reliably to contemporary technologies.This complicates architecture design and can produce classifications so broad that the term loses usefulness.
- Research problem: A concrete definition would provide a base for qualifying more specific concepts such as networked virtual environments and massive multiplayer environments.The paper presents qualification as a way to relate broader and specialized usages.
- Approach: The study uses grounded theory to analyze sampled technologies and identify properties that determine whether they constitute a virtual world.This approach is presented instead of deriving the definition solely from a literature review.
- Approach: Discriminant sampling tests the resulting theory on technologies excluded from the initial study, including pseudo-persistent games, MANets, virtual reality, mixed reality, and the Metaverse.The additional samples are selected to assess whether the theory holds for advanced contemporary technologies.
- Contributions: The article produces a detailed definition, an ontology of related terms and acronyms, and a property-based classification of technologies.It also accounts for distributed peer-to-peer, cloud-based, and MANet networking, and uses pseudo-persistence for technologies that mimic persistence.
2 Related Work
Related work offers multiple definitions of virtual worlds, but the paper identifies unresolved problems involving locality, sharedness, worldliness, immersion, and the boundary between games and social spaces.
- Existing definitions: Dionisio, Burns III, and Gilbert define virtual worlds as persistent online computer-generated environments for real-time interaction among remote users, but this excludes equivalent single-player local implementations.The paper treats the assumed remote location as a problematic restriction.
- Existing definitions: Bartle’s revised definition implies shared time and space but does not capture the worldliness of a virtual world.The paper adopts Bartle’s concepts of avatars and persistence while adding his properties to the study.
- Existing definitions: Bell and Robbins-Bell emphasize synchronous, persistent networks of avatar-represented people, but do not specify whether participants share time or space.The paper also notes the same local-versus-remote problem identified in another definition.
- Existing definitions: Girvan describes a persistent, simulated, immersive environment with avatars and communication tools, while Spence distinguishes persistent synthetic spaces from games.This paper rejects Spence’s game-versus-social-space distinction because social play can be part of gaming.
- Existing definitions: Singhal and Zyda identify shared space, presence, and time, communication, and sharing as net-VE features, but the paper considers these features too vague for technology classification.The paper specifically questions their application to instance dungeons and MANets.
3 Methodology and Scope
The paper defines its scope around computer-simulated environments and uses grounded theory to derive technology-applicable properties. It treats emerging technologies as potential tests of, and inputs to, the evolving theory.
- Methodology: Grounded theory forms the definition from criteria found through technology sampling rather than limiting analysis to prior definitions or literature-derived criteria.Sample technologies are analyzed for properties related to virtual worlds.
- Methodology: The method can incorporate new technologies by adding properties and updating the definition when an emerging technology challenges the theory.This is identified as the novelty of using grounded theory for technology classification.
- Scope and semantics: The paper uses “virtual” in the computer-related sense of a world simulated by software, not an imaginary world formed in users’ minds.This distinction establishes the paper’s semantic scope.
- Scope and semantics: A virtual world models or simulates an allegory of the physical world, and “world” is not restricted to a planetary world.The paper uses EVE Online’s science-fiction universe as an example containing multiple planetary worlds.
- Scope and semantics: For this article, a virtual environment is minimally wholly synthetic, computer-simulated, and spatially containing.A virtual environment must be sufficiently modeled for the computing device to render it.
- Scope and semantics: Computing systems may contain nested simulations and multiple data spaces, whose loaded states can potentially constitute virtual environments.The paper describes simulation layers from processor instructions through software engines and virtual environments.
- Scope and architecture: Architecture discussions must balance virtual-world properties against requirements such as mobility, decentralization, scalability, and limited computing resources.Resource constraints affect how much world space can be loaded on a client or server.
4 Grounded Theory
The grounded-theory study defines and clarifies properties used to classify virtual-world technologies. It distinguishes determinant properties from subjective or non-determinant characteristics, including temporal sharing, spatiality, size, places, agents, persistence, immersion, and transportation.
- Grounded-theory procedure: The study defines properties for classifying sampled technologies, with final classifications recorded in comparative tables.The section headings correspond to classification columns in the tables and appendix.
- Temporality: Virtual time (VT) is a simulated temporal abstraction that can be altered independently of system time (ST), enabling persistence and pausing of loaded virtual states.ST records computer execution and real-world events, while VT governs the loaded virtual environment.
- Temporality: Agents may share multiple time abstractions, but the 1T property requires simultaneous sharing of real-world time and exactly one virtual-time timeline.Clients can share VT even when their system times are unsynchronized.
- Spatiality: Shared spatiality is not restricted to one room, terrain, or physical place; a virtual world may contain multiple terrains and physics.The definition also excludes real-place telepresence when describing a pure virtual world.
- Non-determinant properties: Size, socially determined place, immersion, presence, and transportation do not determine whether a virtual environment is a virtual world.Virtual environments may be subjectively sized, become places through social meaning, and vary in transportation or perceptual qualities.
- Persistence: Persistence can be simulated as pseudo-persistence, while game-world persistence concerns events that do not reset or stop because of real-world interruptions and remains outside this article’s scope.Pseudo-persistence can make a world appear continuously available through scheduled play times around downtime.
5 Analysis and Result
The analysis assembles determinant properties into an operational definition of a virtual world, while reserving other properties for distinguishing among virtual worlds.
- The study aims to delineate what is and is not a virtual world rather than measure how worldly a technology feels.
- The definition is assembled from properties identified by examining technologies, including shared temporality, real-time interaction, spatiality, shards, agents, interaction, non-pausability, persistence, and avatars.
- A virtual world permits many agents, including humans represented by virtual selves or software agents, to interact and act or react within a shared real-time, non-pausable spatiotemporal environment.
- Its data spaces may be multiple, but together they must constitute one shared persistent shard.
- Properties excluded from classification, such as visual representation, size, indoor or outdoor setting, centricity, mediation, and communication architecture, can distinguish different virtual worlds.
6 Verification
Verification compares the definition with prior work and uses it to classify contemporary technologies and organize related virtual-environment terms. The analysis highlights persistence, agent support, shard structure, and temporal or spatial mapping as key distinctions.
- Verification: Discriminant sampling tests the definition on technologies outside the initial study, including pseudo-persistent games, MANets, virtual reality, mixed reality, and the Metaverse.
- Ontology of VE: The ontology places persistent net-VEs within virtual worlds, while multiplayer and networking properties identify related online, massive-multiplayer, game, and role-playing subsets.
- MUD and IRC: A simulation of a world can fail the human-agent criterion when no humans remain, whereas MUD and WoW retain software-agent perception of the world with zero players.
- MUD and IRC: IRC differs from MUD by lacking a shared spatiotemporal virtual environment, which also prevents the associated software-agent and virtual-interaction properties.
- Pseudo-persistence: Diablo and LoL are classified as pseudo-persistent worlds, showing how instances can blur the boundary between video games and virtual worlds.
- Transhumance: Transhumance forms one shard while its common data space remains connected; network partitions can instead create separate shards, and its physical mapping classifies it as mixed reality.
- Mixed Reality: Mixed reality lies between the virtual and physical when virtual properties are mapped to physical counterparts, using a continuum between those poles.
- Metaverse: The Internet fails the virtual-world criteria for shared temporality, real-time interaction, one shard, and persistence, making it closer to a world of worlds than one world.
7 Conclusion
The article provides a detailed, technology-applicable definition of a virtual world and supports updating it as new technologies challenge the theory.
- The article’s primary result is a detailed definition of a virtual world with all underlying terms defined.
- The definition is applied directly to technology to classify technologies and identify properties that distinguish them.
- Properties that do not determine a virtual world are retained to distinguish among different virtual worlds.
- The grounded-theory approach allows the definition to be updated when emerging technologies challenge the current theory.
- Future work could define technologies that have a virtual world by relaxing the nZ criterion or using centricity or presence.
Appendix
The sampling began with representative virtual worlds and expanded through contrasting technologies and systems to separate digital, virtual, and virtual-world properties.
- MUD and World of Warcraft were sampled as representative first-generation and contemporary massive multiplayer virtual worlds.
- A calculator and graphics calculator were added to separate a digital world from a virtual one.
- Skype and IRC were included to compare a virtual telephone variant with chat.
- Zork was added to differentiate MUD from its predecessor.
- Ultima Online, BigWorld Technology, and EVE Online were added to differentiate virtual worlds, including systems using shards and regions.
A Note on Scripting
The scripting discussion limits classification to technologies as considered in the study because sufficiently rich scripting can extend or repurpose their intended capabilities.
- Sufficiently rich scripting languages might make the limits of a technology’s capabilities impossible to determine.
- Scripting can repurpose a technology beyond its intended use, such as enhancing a web server to serve as a game server.
- The study considers technologies without treating scripting’s possible extensions as unlimited reclassification evidence.
A.1 Virtual (Simulated) Environment (VE)
Nearly all technologies in Table 1 implement some form of virtual environment, while the calculator does not because its abstraction is insufficiently high.
- All technologies listed in Table 1 except the calculator implement some form of virtual environment.
- The calculator simulates calculation on a digital display, but its abstraction is not high enough to implement a virtual environment.
A.2 Shared Temporality (1T)
Technologies share different combinations of real-world, hardware, simulated, and virtual time. These combinations distinguish asynchronous systems, synchronized cloud applications, and systems supporting shared virtual time.
- Email chess can share real-world and simulated time while omitting virtual time, because turn-taking replaces simultaneous virtual activity.
- GDocs synchronizes timed events through a cloud platform, sharing real-world and simulated time across clients.
- Skype shares real-world time and records simulated objective times for synchronization, but does not support virtual time.
- IRC preserves event order without implementing virtual time; its chat protocol is therefore not dependent on shared temporal progression.
- Civ5 Hotseat players share real-world, hardware, and simulated time, but alternate access to virtual time rather than sharing it simultaneously.
- The Internet supports many simulated and virtual times, so not all agents share one virtual timeline.
A.3 Real-time (Rt)
The real-time property distinguishes technologies requiring simultaneous activity from turn-based or asynchronous systems. Email chess, Civ5, and the Internet fail this prerequisite.
- Email chess, Civ5, and the Internet fail the real-time prerequisite, whereas other listed technologies may support ongoing interaction.
A.4 Shared Spatiality (1S)
Shared spatiality requires an allegory of physical space, not merely a visual display or communication channel. The surveyed technologies implement this property in graphical, textual, or game-world forms.
- Email chess players can share a spatial representation of the board, supporting shared spatiality in a single shared space.
- Calculators and GDocs lack an allegory of physical space, although CAD drawings can model physical space.
- Skype video shows participants moving through physical space rather than modeling a virtual space.
- Zork and MUD implement shared spatiality textually through room descriptions, player movement, and topology without visualizing space.
- Civ5, Doom, LoL, multi-room MUDs, and a single WoW shard support a shared allegory of physical space.
- Facebook and static Stunt.io pages do not implement shared spatiality because they do not provide an allegory of physical space.
A.5 one Shard (1Sh)
The one-shard property concerns how a technology’s state and agents are distributed across data spaces, servers, instances, and users. The examples show that persistence, agency, interaction, and continuity vary independently across architectures.
- Email chess synchronizes game state across computers, while calculators, CAD drawings, and Zork use one data space.
- LoL separates players into game instances after they gather in a shared lobby, then synchronizes results with player data.
- WoW divides its world into realms, zones, and instance dungeons, with each realm served as a one-shard data space.
- MUDs can remain virtual worlds despite less movement space than modern games, while Doom and WoW use bounded or teleported regions.
- Doom uses bounded levels and sky boxes, MUD uses narrative rooms, and WoW limits movement through edges or harmful areas.
- Listed technologies support at least one human agent, while multiplayer capacity ranges from two or more users to larger simultaneous groups.
- Many listed technologies lack virtual agents, while game worlds can include entities with agency or automated opponents.
- Without simulated agency, tools such as Facebook may permit human-to-human interaction but not interaction between humans and the world.