The definition of subtitle adaptation set says it must have role=subtitle. However, the content selection chapter seems to consider a wide range of roles.
I propose we rename this to "text adaptation set" and remove the role=subtitle constraint and leave the role-specific logic for the content selection chapter to specify in detail (which it already does).
We only have video, audio and subtitle. We have no text in our IOP. Text is anyways misleading as we subtitles may have images as well. So I do not see why we should change it.
The definition of subtitle adaptation set says it must have role=subtitle. However, the content selection chapter seems to consider a wide range of roles.
I propose we rename this to "text adaptation set" and remove the role=subtitle constraint and leave the role-specific logic for the content selection chapter to specify in detail (which it already does).