-
Notifications
You must be signed in to change notification settings - Fork 2
Description
verifiable-issuers-verifiers/index.html
Lines 997 to 1006 in 2cdd528
| The value of the `outputValidation` | |
| <a data-cite="VC-DATA-MODEL#property">property</a> MUST be one or more data | |
| schemas that provide [=verifiers=] with enough information to determine whether | |
| the provided data conforms to the provided schema(s). Each validator MUST | |
| specify its `type` (for example, `JsonSchema`) and an `id` | |
| <a data-cite="VC-DATA-MODEL#property">property</a> that MUST be a [=URL=] | |
| identifying the schema file. The specific type definition determines the precise | |
| contents of each data schema. If multiple schemas are present, validity is | |
| determined according to the processing rules outlined by each associated `type` | |
| property. |
I think the description of this RDF attribute (outputValidation) (among others) needs to make plain whether one value can only contain one data schema, such that if more than one data schema is to be provided, the RDF attribute must occur more than once; i.e., a value that contains commas does not contain multiple data schemas, it contains a single data schema which includes commas in its identifier.
I flag this because attributes/properties like schema:keywords were originally (and may still be) underspecified, and now (for instance) "Multiple textual entries in a keywords list are typically delimited by commas, or by repeating the property" has multiple valid interpretations, so a user might intend a single value of "red, white, and blue" to be a single keyword, but software can instead treat it as three keywords ("red", "white", "and blue"), changing weighting scores (among a great many complications).