Content Translation does not differentiate between different types of references (cite book, cite web, cite IETF etc) - they are all marked as Reference. You can see the difference between Content Translation and Visual Editor for the same reference:
|"Reference" in Content Translation||"Book" in Visual Editor|
More importantly than the label being different, the support for editing these references is also different. Specific types of references have specific dialogs to edit them, which are not used when those references are edited in Content Translation.
For example, on the ContentTranslation clicking on Edit will display the Basic Reference inspector (not template-based dialog):
In VE there will be different, template-based editing dialogs, e.g. if the above reference will be edited on the page (en:Transport Layer Security)
The expected behaviour is for Content Translation to work as Visual Editor does in this regard: showing the specific dialog for editing specific types of references. That would make the experience more consistent and make dialogs more appropriate to the data being edited.
Reported based on the info from T198390#4421534
- T202165: Content translation: Problem with handling references...
- T205420: CXServer fails with HTTP 500: "MT processing error: undefined"
- T205616: CX2: Section alignment broken when translating "Fleming: The Man Who Would Be Bond" from Hebrew
- T205429: CX2: Uncaught TypeError: Cannot read property 'element' of null at VeCeCXTransclusionInlineNode.ve.ce.View.initialize (ve.ce.View.js?9c8a2:109)