spGrp should self-nest

[lb42]

The element spGrp is primarily useful for the encoding of musical numbers especially when more than one speaker/singer participates. In a sense, it is divLike but it does not tesselate the text. It permits interruption of the  act-scene-speech hierarchy, like a nested text or quote.   However, unlike div, ispGrp cannot self-nest.

I propose two changes : first, the content model of spGrp should be expanded to permit spGrp as an explicit alternate along with sp and the members of  model.stageLike and model.global. Second the attribute @org should be added, with values "sequential" and "parallel", indicating whether the  components of the spGrp are to be performed in sequence or in parallel. This seems a more natural and simpler approach to the problem of representing simultaneous singing of different words by different performers than the appropriation of the @sync attribute (which is intended for speech transcription, rather than performance).

Here's an example from a 19c light opera (Rophino Lacy's version of Cinderella) , but there are plenty of others

[lb42]

P.S. A better way would be to add spGrp to the att.divLike class and expand the associated valList for @org