clarify that `@none @id` are blank nodes, swap Examples 104 and 105

[VladimirAlexiev]

• Example 104 https://w3c.github.io/json-ld-syntax/#example-indexing-data-in-json-ld-by-node-identifiers-with-set-representation shows @index: @id with @set representation but doesn't explain its utility, since all such arrays are necessarily singletons
• Example 105 https://w3c.github.io/json-ld-syntax/#example-indexing-data-in-json-ld-by-node-identifiers-using-none shows that an @id index being @none is mapped to a blank node. Then it hit me: there can be multiple blank nodes, and they will all be represented as @none, so now we need an array.

I suggest the following editorial changes:

• Edit Example 105

  • The special index @none is used for indexing node objects which do not have an @id, which is useful to maintain a normalized representation.

  • The special index @none is used for indexing node objects which do not have an @id, i.e. are blank nodes. This is useful to maintain a normalized representation. In cases where multiple blank nodes may be present, it is useful to combine @id with @set, see the next example

• Swap Examples 104 and 105
• Edit the old Example 104:

  • When compacting, this ensures that a JSON-LD processor will use the array form for all values of node identifiers.

  • When compacting, this ensures that a JSON-LD processor will use the array form for all values of node identifiers. This ensures a normalized representation in case of blank nodes, all of which will be mapped to @none and therefore need an array to hold their values.

  • Edit its code to the following (last 2 blocks changed):

    {
    "@context": {
    "@version": 1.1,
    "schema": "http://schema.org/",
    "name": "schema:name",
    "body": "schema:articleBody",
    "words": "schema:wordCount",
    "post": {
    "@id": "schema:blogPost",
    "@container": ["@id", "@set"]
    }
    },
    "@id": "http://example.com/",
    "@type": "schema:Blog",
    "name": "World Financial News",
    "post": {
    "http://example.com/posts/1/en": [{
    "body": "World commodities were up today with heavy trading of crude oil...",
    "words": 1539
    }],
    "http://example.com/posts/1/de": [{
    "body": "Die Werte an Warenbörsen stiegen im Sog eines starken Handels von Rohöl...",
    "words": 1204
    }],
    "@none": [{
    "body": "First object without an @id",
     "words": 201
    }, 
    {
    "body": "Second object without an @id",
    "words": 202
    }]
    }
    }

[VladimirAlexiev]

A similar edit could be done to Examples 107 and 108 https://w3c.github.io/json-ld-syntax/#example-indexing-data-in-json-ld-by-type-with-set-representation. In such case the code in Example 108 https://w3c.github.io/json-ld-syntax/#example-indexing-data-in-json-ld-by-type-using-none could be changed to eg

    "none": [{
      "@id": "https://greggkellogg.net/",
      "name": "Gregg Kellogg"
    }, {
      "@id": "http://champin.net/",
      "name": "Pierre-Antoine Champin"
    }]

[VladimirAlexiev]

In the same train of thought: Example 122 https://w3c.github.io/json-ld-syntax/#example-referencing-named-graphs-using-an-id-map-with-none shows precisely two anonymous (blank node) graphs.

So I think the note above it is misleading because no graphs are merged (check in the Expanded and TriX tabs):

Note, however, that if multiple graphs are represented without an @id, they will be merged on expansion. To prevent this, use @none judiciously, and consider giving graphs their own distinct identifier.

Suggest to change it to:

But using anonymous graphs is not recommended because it could lead to confusion. So use @none judiciously, and consider giving graphs their own distinct identifier.