Toward a Taxonomy of Inventory Systems for Virtual Reality Games

Andrey Krekhov; Ann-Marie M\"uller; Jens Kr\"uger; Sebastian Cmentowski

REVIEW 4 major objections 5 minor 21 references

Toward a Taxonomy of Inventory Systems for Virtual Reality Games

T0 review · 4 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read VR game inventories can be designed from four building blocks, and this paper names them.

desk verdict A modest but real first taxonomy of VR inventory design; the abstract overstates its completeness, but the paper is honest about being preliminary and worth refereeing. read the letter →

arxiv 1908.03591 v2 pith:7JYSXX5F submitted 2019-08-09 cs.HC

Sebastian Cmentowski , Andrey Krekhov , Ann-Marie Müller , Jens Krüger This is my paper · ORCID

classification cs.HC

keywords virtualrealityvideogameinventorytaxonomyuserinterfaceinteractiondesignqualitativeevaluation

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper is trying to establish that the in-game inventory is a tractable design problem in virtual reality, and that its full range can be laid out in a taxonomy of four building blocks: the interface that shows items, the representation of each item, the arrangement of items in storage, and the player's interactions with the inventory. The authors build three inventories from this taxonomy (a flat grid overlay, a virtual shelf of drawers, and a magnetic work surface) and collect qualitative feedback from eight players. The feedback points toward natural, physical interaction and movable, personalizable storage as promising directions, while exposing problems such as occlusion and missing stackability. If the taxonomy holds up, it gives VR game designers and researchers a shared vocabulary for comparing and constructing inventory systems.

What carries the argument

The carrying mechanism is the taxonomy itself, an analytic instrument that splits any VR inventory into four named dimensions and lists concrete options under each. Its work is to turn a diffuse interface problem into a set of discrete design decisions, so that any two inventories can be compared by their positions in the space and new combinations can be generated deliberately. The three prototypes are not the contribution by themselves; they serve as proof-of-concept walks through the taxonomy and as probes that surface the issues designers should watch for.

What would settle it

Compile a catalogue of inventory systems from a broad sample of VR games and check whether every consequential design property can be expressed as a combination of the four building blocks; if any property, such as real-time item decay or shared player ownership, does not fit, the taxonomy's completeness claim fails.

Watch

Extended reading notes

Core claim

The central claim is that inventories in VR games, often underused compared with traditional games, can be decomposed into four atomic building blocks. Inventory Interface determines whether storage appears as an overlay, a virtual object, or a physical proxy, and whether it is static, movable, thematic, or abstract. Item Representation determines whether items are shown realistically or abstractly and at preserved, miniaturized, or normalized scale. Item Arrangement determines layout (linear, grid, ring, slots, or unrestricted), capacity (unlimited, fixed, or dynamic), ordering (unstructured, sortable, or sorted), and possible improvements such as categories, hierarchies, or stacking. Interaction determines how the inventory is opened (button or gesture), how items are added (automatic or manual), and how items are manipulated (pointer-based raycasting or physical action). The paper presents the three prototypes as distinct points in this space and reports that players preferred the most physical and flexible one, the magnetic surface, while still noting its limits in scalability and structure.

Load-bearing premise

The taxonomy is assumed to cover the whole design space of VR inventories, but the paper does not prove that four building blocks are exhaustive, and it evaluates only three instances, so an unlisted dimension such as shared ownership or system-driven organization would make it incomplete.

Editorial extensions

If this is right

VR game designers gain a shared vocabulary for inventory decisions, replacing ad hoc descriptions with four dimensions and their options.
The taxonomy can reveal unexplored combinations, such as realistic three-dimensional items paired with a structured grid, that current VR games rarely attempt.
The study's small-sample preference for physical interaction and original-scale items suggests that natural, body-based inventory handling may be more engaging than raycast menus.
Occlusion and the popularity of movable inventories indicate that spatial placement of the inventory is a design variable in its own right, not a cosmetic detail.
The qualitative results give concrete targets for iteration, such as adding stackability to the drawers and letting players restore a preferred inventory position.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

An implicit consequence is that the same four-block decomposition could describe other persistent VR interfaces, such as crafting menus or quest logs, since the categories are generic; the paper itself does not claim this.
A testable extension is a hybrid inventory that combines the magnetic surface's physical placement with an optional grid-snap or sorting mode; such a design could resolve the tension between freedom and structure found in the study.
Because the taxonomy lacks an explicit dimension for ownership or multi-user access, social inventories, which one participant proposed for the magnetic surface, would likely need a fifth building block if the taxonomy grows in that direction.
The strongest check on completeness would be a design exercise in which independent designers describe existing VR inventory systems using only these four building blocks and see whether any consequential decision is left over.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Colleague,

Quick take: this is a small extended abstract that does something genuinely useful — it puts a first organized taxonomy of VR inventory design on the table, then tests three concrete instances of it. The taxonomy is descriptive and not deeply validated, but it is a real contribution to a sparsely covered topic. Treat the claims as starting points, not results.

What is new: prior work like Wegner et al. (2017) compared two concrete inventory designs but did not propose a cross-cutting framework. Here the authors decompose inventories into four building blocks (interface, item representation, arrangement, interaction) with sub-dimensions, which gives designers a shared vocabulary. The three prototypes (Flat Grid, Virtual Drawers, Magnetic Surface) are well chosen to sample different parts of that space, and the constraints they share (realistic style, button activation, no sorting/stacking) make comparisons cleaner. The qualitative study is small (n=8) but honestly reported as early; the think-aloud quotes are informative and the ranking results are plausible.

Soft spots: the biggest one is exactly what the stress-test note says — the taxonomy's completeness is asserted rather than demonstrated. The abstract says 'covering the broad design space,' but the paper later calls the taxonomy 'preliminary' and says future work will 'refine and extend' it. That tension should be removed: either soften the abstract or add a paragraph in the body explaining what coverage would mean and how one would test it. The study itself cannot validate coverage: eight participants, three instances, one scenario, no measure of whether the four blocks are exhaustive. The authors seem aware of this, which is good, but the abstract's wording overstates. Also, 'huge potential' in the abstract is a bit much for an eight-person early study; a calmer phrase would fit the actual evidence. The occlusion problem is a nice finding, and the suggestion about shared multiplayer storage hints at a missing dimension (social inventories) that the taxonomy does not currently include.

Citation pattern is fine. The related work is appropriately scoped, and the self-citations are to the authors' own VR interaction work that is directly relevant. Nothing here is circular — the prototypes are built from the taxonomy, but the evaluation is about the prototypes, not a test of the taxonomy.

Who it's for: VR game designers and HCI researchers working on game interfaces. It deserves a serious referee for a workshop or extended abstract track; it is not a full research paper yet. Revision should focus on aligning the abstract with the body, and perhaps adding a short section on how the taxonomy's coverage might be assessed.

Verdict: conditional accept with revision.

Referee Report

4 major / 5 minor

Summary. This paper proposes a preliminary taxonomy of inventory systems for VR games, decomposing the design space into four building blocks: Inventory Interface, Item Representation, Item Arrangement, and Interaction, each with several subdimensions. The authors use the taxonomy to design three distinct inventory prototypes (Flat Grid, Virtual Drawers, and Magnetic Surface) and report a small qualitative study with eight participants in a virtual construction scenario. The paper is explicitly framed as an early step, with the stated goal of generating design implications and future research questions rather than validating a complete theory.

Significance. If the proposed taxonomy proves robust, it would provide a useful shared vocabulary for VR inventory design and research, connecting established game design concepts with VR-specific interaction techniques. The paper has concrete strengths: the taxonomy is grounded in existing games and VR interaction literature, the three prototypes are clearly derived from the taxonomy under shared design constraints, and the qualitative feedback, though small-scale, yields plausible and actionable design insights. The authors are also appropriately cautious in the conclusion, calling the taxonomy preliminary and planning future refinement. The main reservation is that the abstract's claim of 'covering the broad design space' exceeds what the current evidence supports; the paper would be strengthened by aligning its claims with its explicitly preliminary status.

major comments (4)

[Building Blocks of Inventories for VR (taxonomy box)] The four building blocks and their subdimensions are presented descriptively without a derivation from a corpus of games, prior classifications, or a systematic coding procedure, and no argument establishes that the listed dimensions are exhaustive or at a consistent level of granularity. Because the abstract's phrase 'covering the broad design space' depends on this exhaustiveness, the taxonomy as presented under-supports the central claim. I recommend either adding a systematic derivation and a completeness argument, or softening the claim to 'a first step / preliminary taxonomy covering identified aspects.'
[Evaluation and Discussion] The qualitative study cannot validate the taxonomy's coverage: it involves eight participants, one construction scenario, and three prototypes, and the paper does not describe any systematic qualitative analysis method, such as thematic coding, transcription procedures, or inter-rater reliability. Consequently, the conclusion's statement that the feedback is 'backing our taxonomy approach' overstates the evidence. The authors should present the study explicitly as formative and exploratory, and either add methodological detail about the analysis or restrict the claims to design implications.
[Conclusion and Future Work] The paper itself calls the taxonomy 'preliminary' and states that future work will 'refine and extend' it, which directly conflicts with the abstract's stronger claim that the taxonomy covers the broad design space of VR inventories. The participant suggestion of shared multiplayer storage also illustrates a potentially consequential dimension—social or shared inventories—that is not present in the current taxonomy. The authors should either explicitly scope the taxonomy to single-player personal inventories or add and discuss such dimensions; as written, a designer using the taxonomy could systematically omit relevant designs.
[Inventory Designs] The three prototypes instantiate only a small fraction of the taxonomy's combinations: there is no real-proxy interface, no automatic item addition, no sorting or categories, no stackability, and limited variation in capacity and layout. The statement that the designs were chosen to cover 'the most interesting aspects of the taxonomy' is subjective and unsupported. The authors should clarify the selection criteria for the prototypes and discuss how the evaluation can inform taxonomy dimensions that were not instantiated, or they should limit the generalization claims accordingly.

minor comments (5)

[Motivation] The phrase 'can provide an enjoying experience' should be 'can provide an enjoyable experience.'
[Item Arrangement] The phrase 'free object placement bares the risk' should be 'bears the risk'; the same correction applies elsewhere if the typo recurs.
[Evaluation and Discussion] The paper mentions an audio-recorded think-aloud process but gives no details on how the recordings were analyzed; a sentence on transcription, coding, or even an informal affinity-diagram approach would improve reproducibility.
[Related Work and References] The claim of a 'first taxonomy' should be supported by a short related-work discussion of prior inventory classifications or comparative studies, going beyond the single citation to Wegner et al.; this would help readers assess novelty.
[Taxonomy presentation] The taxonomy box would be easier to use if each subdimension included a brief definition or example; currently terms such as 'improvements' and 'ordering' are ambiguous without reading the prose.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the taxonomy is a descriptive design-space construction, and the qualitative study is not presented as validation of the taxonomy itself.

full rationale

The paper makes no formal derivation that could reduce to its inputs. Its central contribution is a preliminary taxonomy of VR inventory building blocks (interface, item representation, item arrangement, interaction), presented descriptively with examples from existing games. The three prototypes (Flat Grid, Virtual Drawers, Magnetic Surface) are deliberately constructed to instantiate different regions of the taxonomy, and the qualitative study gathers player feedback on the prototypes; the authors explicitly treat the taxonomy as preliminary and plan to 'refine and extend the existing taxonomy to incorporate additional player and developer input.' No parameter is fitted and then renamed a prediction. The only self-citations are to the authors' prior work on navigation and controller design, cited as examples of prior VR research topics; these are not load-bearing for the taxonomy. The claim that the taxonomy 'covers the broad design space' is an assertion of completeness, but an unsupported completeness claim is a correctness/evidentiary limitation, not circularity, because the taxonomy is not derived from the completeness assumption nor from the study results.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The central contribution is a classification scheme, so the load-bearing input is the assumed completeness of the taxonomy's categories and the assumption that a small qualitative study can ground design implications. No numerical parameters are fitted.

assumptions (3)

domain assumption The success of an inventory implementation depends on comprehensibility, interactivity, contextual embedding, and personalization.
Stated in 'Design Considerations for Inventories', these four factors are asserted as the crucial factors, with citations to game design commentary and VR interaction work, but not derived or empirically validated.
ad hoc to paper The four building blocks (Inventory Interface, Item Representation, Item Arrangement, Interaction) are sufficient to describe the inventory design space.
This is the structural completeness assumption of the taxonomy; it is introduced by the paper without evidence of coverage.
domain assumption Qualitative feedback from eight participants can yield valid first insights for design implications.
Accepted practice in early HCI user studies, but the sample is small and no systematic analysis is reported.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Toward a Taxonomy of Inventory Systems for Virtual Reality Games." pith.science (2026). https://pith.science/paper/7JYSXX5F

@misc{pith2026190803591,
  author       = {Pith},
  title        = {Pith review of: Toward a Taxonomy of Inventory Systems for Virtual Reality Games},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7JYSXX5F}},
  note         = {Machine review of arXiv:1908.03591}
}

read the original abstract

Virtual reality (VR) games are gradually becoming more elaborated and feature-rich, but fail to reach the complexity of traditional digital games. One common feature that is used to extend and organize complex gameplay is the in-game inventory, which allows players to obtain and carry new tools and items throughout their journey. However, VR imposes additional requirements and challenges that impede the implementation of this important feature and hinder games to unleash their full potential. Our current work focuses on the design space of inventories in VR games. We introduce this sparsely researched topic by constructing a first taxonomy of the underlying design considerations and building blocks. Furthermore, we present three different inventories that were designed using our taxonomy and evaluate them in an early qualitative study. The results underline the importance of our research and reveal promising insights that show the huge potential for VR games.

Figures

Figures reproduced from arXiv: 1908.03591 by the authors.

**Figure 2.** The linear quick bar for frequently used items in the game Minecraft [12] [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗

**Figure 3.** The inventory of the game Green Hell [5] is structured by the player alone. However, many VR games still fail to reach the complexity of traditional digital games. One example of typically overlooked features is the in-game inventory [17]. It allows players to acquire new tools and goods throughout their journey and plays an important role in character development. In this work, we establish a starting point for t… view at source ↗

Figures from the paper (7 more)

**Figure 4.** Figure 4: Items are added to the Flat Grid using raycast aiming [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]

**Figure 5.** Figure 5: The Virtual Drawers contain equally-scaled items [PITH_FULL_IMAGE:figures/full_fig_p004_5.png]

**Figure 6.** Figure 6: Players can move the inventory using a handle. A common problem of many games is a cluttered inventory providing poor comprehensibility and overview. Reasons are either massive storage sizes or extreme information densities. The easiest solution is a limitation of the …

**Figure 7.** Figure 7: The Magnetic Surface allows for fine-graded positioning [PITH_FULL_IMAGE:figures/full_fig_p005_7.png]

**Figure 8.** Figure 8: Items stick to the Magnetic Surface until removed [PITH_FULL_IMAGE:figures/full_fig_p005_8.png]

**Figure 9.** Figure 9: The construction scenario used as testbed game. Flat Grid The first inventory design is a flat two-dimensional overlay that is placed at a static position 1.5m in front of the player’s head upon activation. The stored items are arranged using a regular grid of three …

**Figure 10.** Figure 10: Players used various items to test the inventories. "The abstract nature of the flat grid somehow contradicts my expectations towards an immersive VRgame." (P1) "I would love to empty the virtual drawers by turning them upside down." (P2) "I miss the stackability of…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 21 canonical work pages

[1]

Blizzard Entertainment. 2004. World of Warcraft. Game [PC]. (23 November 2004). Vivendi, Paris, France

work page 2004
[2]

Blizzard Entertainment. 2012. Diablo 3. Game [PC]. (15 May 2012). Blizzard Entertainment, Irvine, California, United States

work page 2012
[3]

Cloudhead Games Ltd. 2016. The Gallery - Episode 1: Call of the Starseed. Game [SteamVR]. (5 April 2016). Cloudhead Games Ltd., Vancouver Island, Canada

work page 2016
[4]

Sebastian Cmentowski, Andrey Krekhov, and Jens Krüger. 2019. Outstanding: A Perspective-Switching Technique for Covering Large Distances in VR Games. In Extended Abstracts of the 2019 CHI Conference on Human Factors in Computing Systems. ACM, ACM, New Y ork, NY , USA, LBW1612

work page 2019
[5]

Creepy Jar. 2018. Green Hell. Game [PC]. (29 August 2018). Creepy Jar, Warsaw, Poland

work page 2018
[6]

Doug Dyment. 1968. Hamurabi. Game [PC]. (1968)

work page 1968
[7]

Epic Games and People Can Fly. 2017. Fortnite. Game [PC]. (25 July 2017). Epic Games and Gearbox Publishing, Raleigh, North Carolina, United States

work page 2017
[8]

Juho Hamari and Vili Lehdonvirta. 2010. Game design as marketing: How game mechanics create demand for virtual goods. International Journal of Business Science & Applied Management 5, 1 (2010), 14–29

work page 2010

Show all 21 references

[9]

Andrey Krekhov, Sebastian Cmentowski, Katharina Emmerich, Maic Masuch, and Jens Krüger. 2018. GulliVR: A walking-oriented technique for navigation in virtual reality games based on virtual body resizing. In Proceedings of the 2018 Annual Symposium on Computer-Human Interaction...

2018
[10]

Andrey Krekhov, Katharina Emmerich, Philipp Bergmann, Sebastian Cmentowski, and Jens Krüger
[11]

Mark R Mine, Frederick P Brooks Jr, and Carlo H Sequin. 1997. Moving objects in space: exploiting proprioception in virtual-environment interaction.. In SIGGRAPH, Vol. 97. ACM Press/Addison-Wesley Publishing Co., New Y ork, NY , USA, 19–26

1997
[12]

Mojang. 2009. Minecraft. Game [PC]. (17 May 2009). Mojang, Stockholm, Sweden

2009
[13]

Don Rawitsch, Bill Heinemann, and Paul Dillenberger
[14]

Rocksteady Studios. 2016. Batman: Arkham VR. Game [PS4 VR]. (13 October 2016). Warner Bros. Interactive Entertainment, Burbank, California, United States

2016
[15]

Paul Sztajer. 2010. Mechanical Breakdown: The Inventory. (August 2010). Retrieved July 4, 2019 from https://kotaku.com/ mechanical-breakdown-the-inventory-5612149

2010
[16]

Valve. 2014. SteamVR. Software [PC]. (January 2014). Valve, Bellevue, Washington State, United States

2014
[17]

Konstantin Wegner, Sven Seele, Helmut Buhler, Sebastian Misztal, Rainer Herpers, and Jonas Schild
[18]

André Zenner and Antonio Krüger. 2019. Drag: on: A Virtual Reality Controller Providing Haptic Feedback Based on Drag and Weight Shift. In Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems. ACM, ACM, New Y ork, NY , USA, 211

2019
[20]

In Extended Abstracts Publication of the Annual Symposium on Computer-Human Interaction in Play

Comparison of two inventory design concepts in a collaborative virtual reality serious game. In Extended Abstracts Publication of the Annual Symposium on Computer-Human Interaction in Play. ACM, ACM, New Y ork, NY , USA, 323–329
[1971]

Game [PC]

The Oregon Trail. Game [PC]. (3 December 1971). Minnesota Educational Computing Consortium and The Learning Company, Minnesota

1971
[2017]

In Proceedings of the Annual Symposium on Computer-Human Interaction in Play

Self-transforming controllers for virtual reality ﬁrst person shooters. In Proceedings of the Annual Symposium on Computer-Human Interaction in Play. ACM, ACM, New Y ork, NY , USA, 517–529

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Blizzard Entertainment. 2004. World of Warcraft. Game [PC]. (23 November 2004). Vivendi, Paris, France

work page 2004

[2] [2]

Blizzard Entertainment. 2012. Diablo 3. Game [PC]. (15 May 2012). Blizzard Entertainment, Irvine, California, United States

work page 2012

[3] [3]

Cloudhead Games Ltd. 2016. The Gallery - Episode 1: Call of the Starseed. Game [SteamVR]. (5 April 2016). Cloudhead Games Ltd., Vancouver Island, Canada

work page 2016

[4] [4]

Sebastian Cmentowski, Andrey Krekhov, and Jens Krüger. 2019. Outstanding: A Perspective-Switching Technique for Covering Large Distances in VR Games. In Extended Abstracts of the 2019 CHI Conference on Human Factors in Computing Systems. ACM, ACM, New Y ork, NY , USA, LBW1612

work page 2019

[5] [5]

Creepy Jar. 2018. Green Hell. Game [PC]. (29 August 2018). Creepy Jar, Warsaw, Poland

work page 2018

[6] [6]

Doug Dyment. 1968. Hamurabi. Game [PC]. (1968)

work page 1968

[7] [7]

Epic Games and People Can Fly. 2017. Fortnite. Game [PC]. (25 July 2017). Epic Games and Gearbox Publishing, Raleigh, North Carolina, United States

work page 2017

[8] [8]

Juho Hamari and Vili Lehdonvirta. 2010. Game design as marketing: How game mechanics create demand for virtual goods. International Journal of Business Science & Applied Management 5, 1 (2010), 14–29

work page 2010