TaraSpace
Back to roadmap

Tara Framework · October 2026 target

A file system for spatial applications.

Build canvases locally from files, data, analytics, and semantic relationships—then ship the compiled application to Tara.

The product boundary

tframe is the local framework. tara is the hosted platform. The relationship is similar to a framework compiler and its deployment provider.

Source

Node files, canvas pages, datasets, analytics, and ontology

Compiler

Validation, stable identity, references, layout, and production manifests

Runtime

Fast local canvas rendering with updates as files change

Tara

Deployment, storage, collaboration, sharing, and hosted execution

Project anatomy

Many nodes. Many canvases. One project.

Like file-based pages on the web, every canvas file defines an addressable spatial page. Node, data, analytics, and ontology files can be reused across those pages.

research-workspace/
  canvas/
    overview.canvas.ts
    analytics.canvas.ts
  nodes/
    brief.note.ts
    revenue.chart.ts
  data/
    sales.csv
  analytics/
    revenue.query.ts
  ontology/
    relationships.ts
  tara.config.ts
tframe devtframe checktframe buildtara deploy

Delivery plan

Build the compiler path first, then deepen the runtime.

Each phase should leave a usable product. Dates remain targets until the behavior is tested with real projects.

  1. 01

    Prototype

    File-based canvas

    Developers define nodes and compose a canvas from ordinary TypeScript files. Stable IDs make builds repeatable and keep relationships intact.

    • tframe init, dev, check, and build
    • Node and canvas file conventions
    • Local preview and deployable canvas output

  2. 02

    Next

    Many canvas pages

    A project can contain multiple spatial pages, share nodes between them, and navigate through a generated application manifest.

    • File-based canvas routing
    • Shared nodes and layouts
    • Development navigation between canvases

  3. 03

    Planned

    Data and analytics

    Datasets and analytical definitions become first-class project files. Their results compile into charts, maps, tables, and computed nodes.

    • Local and hosted dataset references
    • Typed analytical queries
    • Reproducible charts, maps, and computed results

  4. 04

    Planned

    Ontology and deployment

    The semantic layer links entities across canvases. Tara deployment turns the compiled project into a hosted, collaborative spatial application.

    • Typed entities and relationships
    • Semantic validation and graph layouts
    • Preview and production deployments through Tara

The prototype is already taking shape.

Follow the public roadmap as local development, multi-canvas projects, and deployment mature.

View full roadmap