Azaroth suggestions

[gkellogg]

• "This specification defines JSON-LD" => "This specification defines JSON-LD 1.1". Fixes #217.
• Remove "if desired" Fixes #218.
• Improve default language definition. Fixes #219.
• it's => its fixes. Fixes #221, fixes #223, and fixes #239.
• ore => or. Fixes #222.
• Simplify simplicity. Fixes #224.
• Improve data model overview. For #225.
• Clarify that literals are not blank nodes. Fixes #226.
• Improve definition of @nest. Fixes #227.
• Update @none definition. Fixes #228.
• sccope => scope. Fixes #229.
• Clarify that @version is set to 1.1, not "json-ld-1.1". Fixes #230.
• Improve requirements of terms. Improve requirements of terms.
• Reference security considerations for remote contexts. Fixes #232.
• Update local dfns to not be links by removing data-cite. Fixes #233.
• Add a missing description of list objects and set objects. Fixes #234.
• Fix wording of nested properties. Fixes #235.
• Add forward references for other things that affect the context in section 4.1. Fixes #236.
• Better describe example 25. Fixes #237.
• Move subsection on using the document base for the default vocabulary until after the section on the document base (4.1.2.1 => 4.1.4). Fixes #238.
• Fix grammar in Scoped Contexts section. Fixes #240.
• Split example 46 into separate examples. Fixes #241.
• Re-defined => Redefined. Fixes #242.
• limits => limit. Fixes #243.
• Fix grammar in Imported Contexts section. Fixes #244.
• effect => affect. Fixes #245.
• Further clarify that type coercion takes place only on strings, not maps. Fixes #247.
• Make examples clearer. Fixes #248.
• Remove old warning in note about redefining IRIs to have a different meaning, which is now disallowed. Fixes #249.
• Use a scoped, rather than an embedded context for default language example. Fixes #250.
• Note that indexing of language tags depends on their not using sub-tags other than the primary language-tag. Fixes #251.
• Remove uninformative note on type coercion from 4.5. Fixes #252.

---

Preview | Diff

[gkellogg]

Actually, there are several more I'll add on to this PR tomorrow.

[gkellogg]

Still a couple of more to go.

[gkellogg]

I'll do the rest in another PR.

[gkellogg]

We can probably improve things more, but yes, the key for an expanded definition my be an IRI, which can’t be a term.

[gkellogg]

Perhaps we should update the context definition to say that keys can be noted keywords and strings treated as terms of IRIs. The values of which may be strings, interpreted as IRIs, or expanded term definitions, or words to that effect.