Key idea & worked example
A reusable result keeps its identity, source and validation scope visible.
A passing check belongs to specific contents and a stated profile.See the worked example →On this page
Two identities, one useful connection
A physical subject and a digital representation need separate identities. The generic teaching graph uses MusicalInstrument and DigitalRepresentation from the released modules, canonicalLabel for language-tagged names, instrumentType for the pipe-organ vocabulary term, and representationOf to connect the model to the instrument. It makes no historical assertionAssertionA statement that can be identified, attributed, qualified and reviewed. Its presence in a source or dataset does not by itself establish truth.Work through the Cuntz example → about Cuntz.
- SubjectLearning organA MusicalInstrument with its own IRI.The model points to the subject
- LinkrepresentationOfAn explicit relationship between two things.Its authored status qualifies the representation
- RepresentationIllustrative modelA DigitalRepresentation with its own IRI.
Your instrument graph
Edit the labels or representation status. Introduce a self-reference to see an actual SHACL error; repair it and export the checked graph.
An instrument and a digital representation, using the MODAVIS 0.1.0 exchange profile. This is a generic practice example.
Technical files · inspect or edit
Direct edits can introduce real validation errors. Rebind / repair updates the supported example’s checksums or representation link.
A MusicalInstrument with its own IRI.
The model points to the instrument.
Representation status: authored.
The drawing explains the supported form. Direct edits to technical files are checked by the reference validator; inspect those files for their exact relationships.
Edit the example, then run validation. Every edit clears the previous result.
Loading the verified example…
Validation runs locally in a worker after you click. The first check loads roughly 20 MB of Python and pinned tools. Inputs are kept in this page’s memory. Downloads are practice outputs; repository publication and additional application profiles need their own checks.
Workflow and reference commands
- Edit the source-evidence note or generic instrument graph in the starter kit.
- For a VAO, update the realization byte count and digest, group total, and carrier manifest byte count and SHA-256.
- Run the pinned validator. Fix errors before exporting. A successful check applies to the declared profile and current files.
Download the verified starter kit ↓
python verify-graph.py --release MODAVIS-0.1.0/source --data learning-instrument.ttl
The native starter contains the unchanged released VAO fixture. The browser starts with a separately identified and attributed learning derivative.
Understand the validation boundary
The browser executes RDFLib and pySHACL against the exact MODAVIS 0.1.0 source release. It loads all released ontology and vocabulary modules and the exchange shapes, excludes the publication shape module, uses no inference and does not fetch remote imports. The report shows a failing focus node, property and message when the shape supplies them.
Reuse the network without copying the example’s assumptions
- Use the immutable release and the modules needed for your domain. Preserve the source, version and licence of the terms and example you adapt.
- Replace the placeholder IRIs with persistent identifiers for your own entities. Keep representations, subjects and file realizations distinct.
- Use evidence and assertions modules when a claim requires attribution, uncertainty or review. Do not treat an authored model as a calibrated measurement.
- Run the exchange checks for exchanged data and add the publication profile when your publishing workflow requires it. Record tool versions, inference settings and exact input files.
Make a change you can check
A reusable result keeps its identity, source and validation scope visible.
- SubjectLearning organA MusicalInstrument with its own IRI.The model points to the subject
- LinkrepresentationOfAn explicit relationship between two things.Its authored status qualifies the representation
- RepresentationIllustrative modelA DigitalRepresentation with its own IRI.
Work through the example 3 STEPS
Use the generic practice example to make one deliberate edit and see what changes.
Edit a label or note
The supported form creates a coherent practice record. Direct technical edits can introduce errors.
Check the current contents
The released validator evaluates the exact current input and returns its actual report.
Export the passing result
An edit clears the previous pass. Export the current checked graph or packed carrier with its report.
- MODAVIS Ontology Network 0.1.0 ↗
Released ontology, vocabulary and exchange SHACL modules. Teaching graph adapted from the verified starter; no remote imports or inference. Publication profile is excluded.
This is an illustrative workflow, not a historical claim or a repository certification.
The useful distinction. A passing check belongs to specific contents and a stated profile.