R&D Prototypes

Structured data that gains depth

HyperRelational handles structured data in several formats — proprietary RDMS, CSV, XML and JSON — and provides data and metadata exploration options drawn from geometric visualization properties.

From spreadsheet to cube

A spreadsheet is a flat object: each cell only makes sense from a row header (the primary key) and a column header. Joining two tables with a join keeps everything on the same plane — exactly what limits understanding.

The master/detail arrangement, on the contrary, keeps each table on a different plane. Thus, data organized by the entity-relationship model forms a 3D cube — and HyperRelational also handles metadata to discover new ways of joining information across the entire collection.

HyperRelational: table in grid view and master/detail form displayed side by side

Data and metadata exploration

Complementary ways of visualizing structured data, drawn from geometric properties.

01

Multiple formats

Imports structured data from proprietary RDMS, CSV, XML and JSON, handling them under a single exploration model.

02

Grid view and master/detail

Switches between the grid table and the master/detail form, choosing the ideal depth level for each query.

03

Geometric visualization

Maps the entity-relationship model to a 3D cube, preserving the spatial relations between tables.

04

Metadata exploration

Uses the collection's metadata to discover new ways of joining information across different tables.

From plane to cube

Why join flattens the data — and how master/detail organizes it in depth.

Data cube unfolded into a cross shape, representing tables joined on the same plane
Unfolded cube (join). Joining tables with a join keeps all data on the same plane, like an open cube.
Dataset represented as an assembled 3D cube with colored faces
3D cube (master/detail). The master/detail arrangement keeps each table on a plane — the data forms a cube and the relations are preserved.