This needs to be fixed both in the digital and exported apparatuses.

I did not understand XML namespaces when this was first made. Currently the input file is being unnecessarily modified I stead of handling namespaces better

This could be done in the collation editor, but then readings would be regularized. Since the open-cbgm can filter based on "type" attributes, it is better to present the actual readings in the apparatus but retain the ability to choose whether to ignore or include common spelling variations and nonsense readings when tabulating coherence.

My project only concerns four of Paul's letters (Rom-Gal), so I did not add functionality to identify different book references.