@@ -13,13 +13,15 @@ Metadata is specified in the form of key-value pairs. Keys may only occur once i
-[`title`](#title)
-[`description`](#description)
-[`keyword`](#keyword)
-[`creator`](#creator)
-[`publisher`](#publisher)
-[`license`](#license)
-[`language`](#language)
-[`creator`](#creator) on top level of a collection
-[`publisher`](#publisher) on top level of a collection
-[`license`](#license) on top level of a collection
-[`language`](#language) on top level of a collection
-[`format`](#format)
-[`learningResourceType`](#learningresourcetype)
Some attributes are only _mandatory on the top level of a collection_, these attributes are inherited down to exercises contained in subfolders, if not overwritten there.
The remaining ones are optional and can be used to provide extra information. It is encouraged to also fill in the optional attributes. For more details, see the description of the individual attributes. For each attribute, an example how it should look in the metadata is provided. The format used in these example is YAML, if you use JSON the syntax is different.
If you have content in a repository which is not easily describable with a single metadata file, [collections](#collections) might be useful.
...
...
@@ -474,7 +476,7 @@ In such cases a single metadata file is not well suited to describe the entire r
