|DITA authoring concepts|
When information in a specialised information type needs to be converted into the information type from which it was specialised, the process is known as generalisation.
The opposite of specialisation is generalisation.
When topics are interchanged between authors or organisations, there may be a need to move content in specialised information types back to standard, base DITA information types. Likewise, when a document containing specialised information types is being processed to a delivery format through a publishing tool that is unaware of the specialisation, the content must be transformed to a non-specialised, general DITA structure.
When processing a specialised element that it is not aware of, a processor (that is specialisation-aware) will treat the element as though it is the ancestor element from which it "evolved".
For example, let's assume we have a specialised element called part_desc (based on the propdesc element) in a parts_list topic (based on the reference topic). When we transform that topic to HTML using a publishing tool, the processor, unaware of both parts_list and part_desc will just handle those elements as though they are reference and propdesc respectively.