An adaptive interpretation platform for museums, galleries and schools.
Placard maps that ecology. By structuring creators, historical events, materials and ideas into a connected, cited, traversable graph, we hold the full intellectual complexity of a collection - then condense it, at the wall, into a quiet adaptive label that gives every visitor the interpretive layer written for them.
Stylised impressions of Claude Monet's Rouen Cathedral series (1892-94). Spatial placeholders, not reproductions.
As algorithmic image synthesis makes the physical production of pictures effortless, the critical locus of artistic value is migrating. Three positions dominate the argument, and they now coexist on the same gallery floor.
Creativity as an un-individuated, statistical distribution of style, with the generative machine as a collaborative agent.
Value retreats entirely into conceptualisation - the cognitive labour of meaning-making over manual execution.
Artistic integrity located in lived human experience - the subjective, ethical act of choosing what matters as the primary gesture.
After decades spent on the "death of the author", the arrival of AI makes conscious human intent unavoidable once more.
These conflicting methods coexist, demanding an exhibition practice that supports physical craft, digital augmentation and conceptual minimalism at once.
The real revelation of generative neural networks is not that machines can paint, but that human creation has always been deeply networked - an ongoing, systemic dialogue of influence, reference and resistance. Historically the machine's training set rendered influence legible only to algorithms, while the human web of inspiration stayed invisible, scattered across archives, letters and studio notebooks.
Placard builds the human counter-narrative - a visible, authored, consented network owned by the artist.
The graph does not reduce what an artwork means; it documents a provenance of attention - an empirical record of what the artist saw, read and negotiated. And because experimental aesthetics shows that interpretive text directly shapes how a viewer feels and processes a work, this network extends to the gallery floor. Choosing a Lens becomes a curatorial act in its own right, and repositions the visitor as an active participant in making meaning.
Placard holds works, creators, materials and socio-political events as semantic entities in a single governed knowledge graph. Every connection is attributed; every interpretive thread is citable.
One historical trajectory through the Living Network - from Hokusai's serial views of Mount Fuji, through the logic of the series, to the market infrastructure of Durand-Ruel.
Stand in front of a Monet canvas, pull the seriality thread, and trace its structural logic back to Hokusai's Thirty-six Views of Mount Fuji. Follow the economic signal forward to Paul Durand-Ruel and the market mechanics that made series painting a viable strategy in the nineteenth century. Every transition is a validated historical connection, not an algorithmic recommendation.
Researchers, curators and visitors navigate the same complex historical and economic trajectories, at the depth each needs.
To prevent vendor lock-in, the graph is built on the Linked Art data model - an implementation-focused serialisation of the ISO-standard CIDOC Conceptual Reference Model (CRM), expressed in JSON-LD. Your data stays portable, open and standards-compliant.
Behind every masterpiece lies a vast network of historical, scientific and economic context - and exhibition design routinely forces the whole web through one physical bottleneck. The standard 50-to-60-word printed wall label is the narrowest possible aperture into a work's history.
We design for the empirical reality the visitor studies describe. At the Metropolitan Museum of Art, Smith and Smith timed a mean of 27.2 seconds with a work, a median of 17, and of that only about seven seconds on the label - the entire audience a printed card of fifty words will ever get. Because print is static, those fifty words must speak in one compromised voice, at one reading level, in one language, at one mounting height, to the scholar, the tourist, the schoolchild and the visually impaired visitor at once.
Placard is the physical interface of the knowledge graph at the wall, and it keeps the delicate art-label-art rhythm of looking. At rest the tablet shows only the minimalist Tombstone (creator, title, date, medium) - indistinguishable from good print. The deeper layers stay opt-in, revealing themselves only when a visitor asks.
Monet's Rouen Cathedral canvases show why interpretation has to be multi-sided. A social historian, an optical physicist and a market economist can each write a factual, compelling account of the same square of paint. No printed placard can hold that density.
Placard holds these narratives as named Lenses. Every text is commissioned from human experts, fully attributed, and governed by your institution's editorial guidelines.
Each reading here was written and signed by a person (i.e. commissioned, not generated). You commission the Lenses, you name the authors, and you see which ones visitors actually open.
Exhibition didactics have always evolved with the technology of the display space. We read the history as a mandate: the label has been reinvented before, and the physical wall is now saturated.
Princess Ennigaldi-Nanna's public museum at Ur labelled its Mesopotamian antiquities on clay cylinders inscribed in three languages. The very first museum label was explicitly multilingual.
Private collections and cabinets of curiosities carried no public labels at all. Viewers were assumed to hold the classical and theological education needed to decode the work; the silence of the walls was a deliberate social barrier.
Gustav Waagen, director of Berlin's Gemaeldegalerie, introduced the first modern didactic labels to bypass expensive catalogues and open the collection to an uninitiated public.
The mid-century white cube stripped the gallery to project neutrality. But the push for multivocality - the decolonial re-labelling at the Rijksmuseum and the Pitt Rivers - places demands a 50-word paper substrate can no longer carry. The wall has run out of space; Placard is the medium catching up with the sector's ethical commitments.
For two decades institutions have made sincere commitments to inclusive, progressive, decolonised interpretation. Those commitments collide with the physical wall, which stays restricted to one label, in one authoritative voice, at one reading height.
The obstacle is the medium itself. A printed card cannot adapt its length, language or typography, so every progressive inclusion becomes a spatial exclusion, negotiated line by line in curatorial meetings.
Placard ends that rationing of authority. Moving from a static substrate to an adaptive, multi-sided digital layer, the community elder's account, the source-nation perspective and the curator's art-historical analysis stand side by side - each attributed, none displacing the others.
Access is also a compliance matter. The Equality Act 2010 requires public venues to make reasonable adjustments for disabled visitors. Static print cannot adjust; an adaptive interface is, by definition, the adjustment.
In visitor studies, dwell time is the most trusted, empirically validated success metric. From the tracking studies of Robinson and Melton in the 1920s to modern eye-tracking, the research is consistent: time spent paying attention is the prerequisite for learning.
Yet gallery interpretation has never had a real feedback loop. A curator writes a label, it is printed and mounted, and the institution stays blind to whether it was read, understood or ignored.
Placard introduces a precise, passive, fully anonymous feedback loop. By tracking reading depth reached, Lenses opened and the time spent looking after reading, we replace curatorial guesswork with hard engagement analytics.
Because the data schema uses the Linked Art standard, those analytics are clean, standardised and portable. Your institution gains a rigorous, publishable evidence base for trustees, donors and public funding bodies - the first large-scale record of how interpretation is actually read.
A single secure knowledge graph serves four distinct surfaces. Each audience meets the Living Network through the interface built for it, and most visitors will only ever meet the label - which is the point.
Reads as clean static print at rest, opening only on demand. Custom Lenses, translation layers, human-read audio description and typographic accessibility controls.
Unfolds the tombstone into an interactive, multi-dimensional map. Pull conceptual threads through an artist's whole series and outward into political, economic and theological histories.
Built on Visual Thinking Strategies (Housen and Yenawine). Inquiry-based looking over passive consumption, prompting children with open, sequenced questions - "what is going on in this picture?", "what do you see that makes you say that?"
A creator-directed portal where living artists log their influences, material experiments and daily references in real time - an authentic, timestamped, artist-owned record, and a consented counter-narrative to the unconsented scraping of generative models.
Author and publish plural, multivocal didactics with full editorial governance, and measure their impact against standardised visitor analytics.
Frame serial art, installations and retrospectives as unified conceptual statements, giving collectors deep provenance and artist-signed commentary.
Build and own an immutable, timestamped record of your influences, materials and creative lineage, secure from commercial web scraping.
Query and cite structured threads across multiple collections, and analyse the first large-scale empirical dataset of public reading behaviour.
Deploy standardised, inquiry-based visual-literacy frameworks (VTS) in front of the artwork, bridging the gallery and the classroom.
Back a single scalable semantic knowledge graph serving multiple B2B surfaces in a high-barrier sector with virtually no direct incumbent.