/ The Wrong Archives
My project, The Wrong Archives, is structured in two distinct but interrelated stages. The first stage focuses on the infrastructural architecture of the archive, a system that is conceptually and operationally grounded in the interfacial stack model developed in the previous section. This stage concerns the automated, back-end processes that assemble the archive. The second stage, presented here under the headings "Browsing the Archive" and "Artwork Interface," examines the front-end design and user experience. This stage applies the observational and analytical insights gained from studying platforms like the Rhizome ArtBase to create a navigable, legible interface for the end-user.
The infrastructural core of The Wrong Archives is built using the dynamic page and database functionalities of the Wix platform, implementing a Content Management System (CMS) workflow. This technical choice is a direct application and operational test of the interfacial stack model developed theoretically in the previous chapter. The model, an adaptation of Benjamin Bratton’s concept, reframes the digital environment as a vertical assemblage of interconnected layers. For this archival project, Bratton’s encompassing "Interface" layer is refined into a granular, four-layer stack (User to Hardware, Hardware to Software, Software to Software, Software to User) to precisely dissect the technical curation of the archive. This framework is essential because it moves analysis beyond the graphical front-end to expose the machine-to-machine diplomacy and automated logic that constitute the archive's operational core.
The practical problem of scale justifies this stacked approach. The Wrong Biennale 2022–2023 comprises 148 distinct curatorial spaces; manually coding a unique page for each would be prohibitively time-consuming. The solution is architected within the stack: by constructing a relational database in Wix's backend (an act initiated at Layers 1 & 2), I establish a structured dataset. Wix's software then processes this data through its internal APIs (Layer 3) to automatically generate consistent, formatted pages, transforming a curatorial problem of volume into one of systematic design. This automation embodies the "programmatic curation" made possible by Layer 3, where the curator's labour shifts from manual assembly to authoring the rules of a generative system.
The CMS dataset formalises the metadata I deem essential for both preservation and analysis. Each record functions as a structured profile for a pavilion, containing fields for: the curator's name; a hyperlink to the original external website, tagged dynamically as "dead" or "alive" to visually display the condition of link rot; a descriptive text of the curatorial space; and a dual-tagging system. This tagging system is a critical analytical layer, allowing filtration by both the temporal status of links and the style of the curatorial space itself. From this centralised database, the Wix platform generates dynamic output in two primary forms.
First, it compiles a curated, linear index on the archive's main landing page (Figure 45). This list automatically presents each entry's core data. Crucially, it preserves the hyperlinks to the original external curator websites. This is a deliberate design choice to resist the archival impulse toward enclosure; by facilitating exit from my archive, the interface reinforces the rhizomatic structure of the web itself, acknowledging that the archive is a node within a wider, often fragile, network. Second, the system automatically generates a unique, internal page for each of the 148 pavilions (Figure 46). Access to these pages is mediated through a dedicated button, a distinct UI element chosen over standard hyperlink text to semantically differentiate internal navigation (deeper exploration within the archive) from external departure (out to the live web). These uniform artist pages host the primary preservation artefact, the screen recording, alongside the full set of descriptive metadata.
This conceptual model of the interfacial stack translates into a concrete technical presentation. The process is initiated at Layer 1: User to Hardware, which encompasses my physical interaction with the machine: typing data, clicking, and commanding the local computer’s resources to run the Wix editor. This foundational layer represents the initial, material act of curation and data entry. Its constraints are felt physically, as generating numerous dynamic pages demands significant processing power, occasionally causing system lag, a direct reminder of the hardware layer's physical limits. These inputs are then mediated by Layer 2: Hardware to Software, which governs the essential electronic exchange enabling the archive to exist online. This layer translates my local inputs into signals for Wix’s servers, managing the fundamental client-server request/response cycle. My curatorial agency here involves configuring this connection and overseeing the data transfer between local and cloud environments.
The core archival engine, however, operates at Layer 3: Software to Software. This is where the archive is assembled through automated, programmatic curation. Wix’s CMS APIs query the relational database, apply templates, and generate all 148 unique pages. My curatorial agency at this stage resides not in manual assembly, but in designing the database log and the rules for presentation. This software-to-software dialogue executes the curation autonomously. This layer’s function is further exemplified by the archive’s handling of screen recordings. The videos are hosted on YouTube, a decision that engages another external stack. When a Wix-generated page calls the YouTube API to embed a video, it initiates a software-to-software dialogue between platforms. This seamless integration is a classic operation of Layer 3 diplomacy, leveraging Bratton’s planetary-scale ‘Cloud’ layer for scalable preservation while revealing the archive’s distributed dependence on external systems.
Finally, Layer 4: Software to User binds the curated output of all underlying operations into the final, public-facing interface. This layer presents the minimalist list index, the individual pavilion pages with metadata, tags, and embedded screen recordings. It is the “visual instrument” (Bratton, 2015, p. 224), 5 the legible, navigable interface for user experience and research that renders the complex activity of the deeper stack comprehensible.
The Wrong Archives does not only use an interface; it is constituted through the active orchestration of an interfacial stack. The project demonstrates how Bratton’s macro-scale framework can be adapted to model the micro-scale practices of independent digital curation. From the physical data entry (Layer 1) to the automated page generation (Layer 3) and final visual presentation (Layer 4), the archive exists only through the integrated function of all strata. This analysis validates the stack as an essential tool for understanding digital curation not as surface-level design, but as the strategic configuration of interconnected technical layers.
Browsing the Archive
For the main landing page of The Wrong Archives, I opt for a minimalist design where each entry is presented in a linear list format. This stylistic choice is informed by my observation of websites that utilise simple lists to present information, a practice I study throughout this research period. For instance, Artistsspace employs a straightforward list table for its ‘texts’ section, consisting of the article title followed by the author’s name and publication date. This approach resonates with Laurel Schwulst’s design philosophy, who states that such a website “aims to be surfable, like the original web—before search engines. hypertext, or the link element, is default blue and underlined. Links are oxygen, helping you go from place to place. Hopefully the website conveys unknowable depths” (Schwulst, 2019). 3 I fully agree with this perspective, as the functionality of my website is predicated on hyperlinks, which animate the digital environment and establish its essential rhizomatic structure through multiple points of entry. A similar principle is evident in Rhizome’s own website, not in the specific ArtBase search, but in the overarching site search they term the “index”, an elaborated system of searching via keywords. Their index presents results with the date (placing the most recent at the top), followed by a cover image, title, and finally tags. In an interview, Mark Beasley, the lead developer of Rhizome, who also works on Artistsspace, elaborates on this design. He explains that the index is a conscious move away from a traditional blogroll format to make the platform’s extensive history more accessible.
. The goal is to create a powerful yet quick interface that facilitates deep exploration by prominently featuring tags and allowing users to chain queries together, effectively nudging them into niche areas of the archive (Beasley, 2024). 5 This concept of an ‘index’ is further reflected on Beasley’s personal website, where he lists his projects in a chronological list, explicitly labelling it an index.
Guided by these precedents, I develop my own interface for browsing The Wrong Archives. Each entry begins with the project’s name, followed by the names of the involved curators. The curators’ names are hyperlinked, leading to their personal websites, thus creating an immediate external connection. An ‘Exhibition View’ button then directs the user to a dedicated artist page, which is discussed in the subsequent subchapter. The final component of this list index is a set of active, searchable tags.
The first tag category addresses the temporality of the net and the concept that websites, though often considered immaterial, are living organisms that can fragment and cease functioning over time. As this project is a collection of video recordings from 2021–2022, some documented projects no longer exist online, accessible only through my screen recordings, while others remain live. To highlight this fragility and the imperative for preservation, the primary search tag is ‘Link Rot’, which allows users to filter projects based on their current status: active or rotten.
The second tag system enables searching based on properties I carefully assign to each project. I fully acknowledge Rhizome’s assertion that tagging net art is a complex and treacherous task due to the form's endless variety (Rhizome, n.d.). Therefore, the tags for The Wrong Archives are not intended to define the artworks themselves but to classify the curatorial practices and the ‘how’ of their exhibition in digital space. Conceptually, these tags are divided into three categories mirroring the structure of this thesis: Layout, Structure, and External. The Layout category consists of three tags #Digital, #Hybrid, and #Physical, showing the environment in which the project is presented. As The Wrong Biennale accepts both physical and digital formats, this categorisation helps trace the manifestations of hybridity in contemporary net art practice. The Structure category analyses the use of digital space itself, with three tags: #Static, #Interactive, and #Virtual Gallery. #Static refers to websites that present artwork in a straightforward, linear manner using standard scrolling and clicking. #Interactive describes sites specifically coded to break conventional design, inviting unexpected engagement with their elements. For example, the project This is the Wrong Place to Die (2021–2022) requires users to navigate through a non-linear, portal-based interface. Finally, #Virtual Gallery identifies environments built as navigable three-dimensional spaces, often with a game-like quality. The final category, External, tracks the use of established third-party platforms for presenting content. The tags here are self-explanatory: #Social (encompassing platforms like Instagram or TikTok), #Streaming (primarily through Wrong TV), and #Plug-in (referring to small code snippets that function as standalone applications or browser extensions for the exhibition). This taxonomy allows the archive to be navigated not as a collection of artworks, but as a study of net art exhibition practice in the digital age.
Artwork Interface
The artwork interface within The Wrong Archives is primarily constituted by the screen recordings I produce during The Wrong Biennale of 2021-2023. These recordings vary significantly in length, a direct reflection of my own curatorial engagement and the time I invest in exploring each pavilion. This variation is not arbitrary but serves as a telling data point, implicitly documenting my level of interest and the temporal depth of my interaction with each digital space. The very nature of this method renders the archive highly subjective, an explicit record of my personal navigational preferences and curatorial gaze. Typically ranging from one to five minutes, the duration often corresponds to the interactive complexity of the website in question, with the most extended recordings dedicated to virtual gallery environments that demand spatial navigation. This quantitative aspect of the archive offers a fertile ground for further analysis, potentially revealing which modes of net art exhibition prove most engaging from a user-experience perspective.
A significant limitation of this screen-recording approach, which I readily acknowledge, is its omission of sound. As a medium, audio plays a crucial role in several exhibition spaces, often employed to construct immersive ambient environments. Its absence here undoubtedly flattens the experience of those specific pavilions. This experiential shortfall underscores a broader challenge in digital preservation: while screen recording is a straightforward and accessible technique, it is often an incomplete solution, incapable of capturing the full sensory and interactive dimensions of net art without compromise. This realisation leads me to agree with the more sophisticated approaches pioneered by institutions like Rhizome. Their development of emulation environments, which can host a server to re-enact the original conditions of an artwork, or tools like Webrecorder, which allow for a fully interactive recording of a webpage, presents more holistic models (Rhizome, 2025). My project, in its deliberate simplicity, thus serves to illuminate the practical constraints faced by independent curators who may lack extensive resources, thereby highlighting the profound complexity of the field and the pressing need to develop sustainable, large-scale strategies for preserving net art.
The metadata displayed on my artist interface page is arranged in a manner analogous to the structure of Rhizome's ArtBase. It features an expandable menu, the first element of which is ‘Users’, linking to pages dedicated to the curators responsible for creating each pavilion (Figure 49). Notably, I make a conscious decision not to include links to the individual artists exhibiting within these pavilions. This is a deliberate curatorial choice, coming from the archive’s primary focus on investigating curatorial practice and the design of exhibition environments, rather than on preserving discrete artistic works. The term ‘User’ is specifically chosen for its inclusivity, a concept I elaborate on in the subchapter ‘User/Curator/Artist’; here, it encompasses both the viewers who engage with the website and the curators who architect the spaces for exhibition. The subsequent section is ‘Tags’. This system of keywords facilitates the filtering process within my archive, enabling users to comprehend and navigate the predominant stylistic and structural approaches of each curatorial space.
The final component is the ‘Description’, which diverges from the conventional practice of briefly summarising the conceptual premise of the project. Instead, it describes the exhibition environment itself. This approach is inspired by Laurel Schwulst’s work for Artistsspace, where she champions the practice of exposing image descriptions, or ‘alt text’, as a meaningful layer of interpretation. Schwulst reframes these descriptions not merely as an accessibility tool for screen readers but as an additional "entrance" into an image, offering perspective and insight into the interpreter's gaze (Schwulst, 2019). 4 In a similar vein, I seek to repurpose the descriptive field to document the materiality of the digital space. My descriptions are intentionally concise, adhering to the aesthetic of exposed metadata. They detail the overall layout, visual aesthetics, and functional mechanics of the pavilion, how artworks are displayed and how the environment operates. This descriptive strategy operates in tandem with the assigned tags, providing a prose counterpart to the taxonomic data and collectively framing the archive as a study of curatorial spatial design in the digital realm.
This chapter traces a trajectory from the theoretical modelling of the digital environment as an interfacial stack to the practical instantiation of that model in The Wrong Archives. By analysing the archive's infrastructure through the lens of Bratton’s adapted layers, from the hardware limitations encountered during data entry to the software-to-software automation that generates its pages, the project demonstrates that independent digital curation is fundamentally an act of orchestrating interfacial relationships. The subsequent design of the browsing interface and artwork pages further applies this logic, translating backend data structures into a legible, navigable front-end experience informed by pioneering net art platforms. Ultimately, the archive serves a dual purpose: it is both a preservation tool for a specific event and a methodological case study. It makes visible the often-invisible labour and layered communications that constitute digital independent curatorial practice, arguing that to curate for the net is to actively engage with and configure the very stack that renders it possible.