Skip to content

Commit 610bfe9

Browse files
JeroenDeDauwclaude
andcommitted
Map ECHOLOT's statement-level requirements to the existing model
The grant's "qualification of data statements" / "statement-level access" wording is Wikibase vocabulary, and cross-reading it against NeoWiki keeps suggesting model work that the reification design already covers. Record the position in the ECHOLOT planning doc, pointing at qualifiers-and-references.md, and carry the two real residuals as open questions: write granularity (new Medium Priority question) and editing/display ergonomics for reified values (extends the existing multi-Subject editor question). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
1 parent 606a960 commit 610bfe9

1 file changed

Lines changed: 18 additions & 0 deletions

File tree

docs/planning/ECHOLOT.md

Lines changed: 18 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -2,6 +2,20 @@
22

33
If you are not familiar with the NeoWiki terminology yet, see [the glossary](../glossary.md).
44

5+
## Statement-Level Requirements
6+
7+
The grant phrases several requirements in Wikibase vocabulary: "qualification of data statements" (T3.1),
8+
"statement-level access", and "Wikibase data model support" (both T3.2). In Wikibase, qualification, references, and
9+
rank are machinery on the Statement itself. NeoWiki Statements are deliberately flat, and the model already
10+
expresses the same information: the value is reified into a linked Subject typed by its own Schema, or, when the link
11+
between two Subjects is what needs qualifying, the Relation carries the properties. See
12+
[Qualifiers and References](../qualifiers-and-references.md), which maps each Wikibase construct to its NeoWiki
13+
expression, including in RDF; that mapping is also the translation a Wikibase-model import (T3.2) will apply. No
14+
statement machinery is needed. Statement-level read access is the Subject payload itself, and Relations are
15+
individually addressable by their stable IDs. What remains open is not the model: inline editing and display of
16+
reified values (the multi-Subject editor question below) and write granularity (the statement-level writes question
17+
below).
18+
519
## Open Questions
620

721
### High Priority
@@ -32,6 +46,9 @@ If you are not familiar with the NeoWiki terminology yet, see [the glossary](../
3246
* Is our [Graph Model](../api/graph-model.md) OK? In particular, is it OK to have non-Subject data in there, like the connected
3347
MediaWiki pages? (80% likely, briefly covered in Vienna: can filter out these values when querying)
3448
* How important is multilinguality for ECHOLOT? Do we need to provide anything beyond our current data model to support that?
49+
* Statement-level writes: the Subject write API replaces the whole Subject with no base-revision check, so concurrent
50+
writers lose each other's updates (last write wins). Do WP4 enrichment flows need a partial-update operation or
51+
defined conflict semantics? (Becomes concrete once a reconciliation service writes alongside human editors.)
3552

3653
### Low Priority
3754

@@ -42,3 +59,4 @@ If you are not familiar with the NeoWiki terminology yet, see [the glossary](../
4259
* Is multi-Subject support in the editor essential?
4360
Example: Person has a "Name" property. Name is a Subject with its own PersonName schema. The "Edit Person" form would show the
4461
PersonName fields and create or update both the Person Subject and linked PersonName Subject.
62+
Display raises the same question: showing a linked Subject's values on the linking Subject's page currently requires Lua.

0 commit comments

Comments
 (0)