 |
Stream44 Studio Open Development Project |
Preview release for community feedback. Get in touch on discord. |
Designed by Hand AI assisted Code |
Ontology-driven structural modeling tool for declaratively building multi-dimensional models with layered reactive functional data processing resulting in realtime interactive visualizations.
bun.sh is required.
After cloning run:
bun install
# Start the workbench
bun run dev
# Open browser
open http://localhost:3000
# Run tests (optional)
bun run test
Point your AI at docs/AI-MODEL-BUILDING.md and make some changes to the examples.
These visualizations are generated from execution traces logged when running examples/03-QuadrantCodepath-SimplePasswordLogin/.
See docs/Roadmap.md for planned visualization libraries.
Semantic models are an indispensable tool for model-driven development and structuring context for AI.
Semantic modeling becomes a complex task when combining multiple models horizontally and vertically leading to slow progress in practical applications of comprehensive models.
Knowledge models require better tooling and toolchain integration to evolve into a well-understood and leveraged technology.
Much focus is on storing the data of a system in a graph.
We are focused on storing the structure of a system in a graph to provide the data-processing indepent blueprint for a distributed system.
Framespace Genesis uses a novel approach to create complex layered models by taking a code-first approach that feels familiar and is efficient to work with by hand and AI.
Models are constructed as code components with declarative mappings to other components and executed by creating and resolving a promise chain across component method invocations. This is made possible by the encapsulate library.
A code-first approach to building semantic entities allows for hoisting functional processing nodes into a graph and for the construction of dynamic graphs with very litte tooling. When the tools disappear and the abstraction is clear new possibilities arise.
-
Explore model development using the encapsulate approach and layered functional processing in a structured graph to discover sclable graph processing patterns.
-
Explore the creation of interactive visual interfaces driven exclusively by models.
Visual models running in the workbench are constructed by calling model APIs implemented in various lower layers.
Exposes model APIs for querying data. In the process model layers are linked into reactive data processing graphs.
Space models are modular & self-contained graph models with strict boundaries.
They are infinite data substrates constructed from simple repeating primitives.
- Capsular
- A model that encodes component-based implementation architectures into capsule spines.
- The schema is dictated by https://github.com/Stream44/encapsulate
- Implementation details: L4-space-models/Capsular/README.md
Semantic models define dimensions in a space by structuring primitives into linked objects with specific properties.
- Capsular / Capsule Spine
- Conveniently query Spine Instance Trees, Capsule Source Trees and Membrane Events in Capsular spaces.
View models structure one or more semantic models into derived layout visual canvases.
See L8-view-models/README.md for details.
Domain models layer view models into topic specific observer oriented perspectives.
The framespace workbench UI SPA implementation that connects to the Model Server.
Repository DID: did:repo:e7b46f0978c2cc02461b480b99a6589a2b6fa888
| Inception Mark |  |
Current Mark |  |
Trust established using Stream44/t44-BlockchainCommons.com |
(c) 2026 Christoph.diy • Code: LGPL • Text: CC BY-SA 4.0 • Created with Stream44.Studio