19-08-2018, 10:15 AM
I'm working on a argument for parsing the script which uses exactly this evidence (there are several "single glyph sequences" throughout the text). It's not groundbreaking, but it does put parsing on a firm basis and shows which glyphs are problematic.

