Correlation/coherence between allographs of different graphemes

[simonastoyanova]

Two levels of this sort of comparison across graphemes, one already possible in Archetype and the annotation tool. Going by the facet component-feature we can see, for example, all instances of apex+serif, whether it appears in A, V, M, Λ or whichever other grapheme.

Deeper level of comparison which isn't yet handled is: does this allograph of A appear often in the same inscriptions as this allograph of P? This would allow the identification of styles (sometimes called scripts, cf. insular) across the alphabet rather than focusing on the variants of a single allograph or component.

Two distinct styles in ISic000093, though they are kept separate in the layout: https://kingsdigitallab.github.io/crossreads/annotator.html?obj=http://sicily.classics.ox.ac.uk/inscription/ISic000093&img=ISic000093.jpg&scr=latin-1
The Fs and the Ps have long and curved setifs on the downstroke, to the point of being consufind (F or E). How often does that happen and it is also co-occuring with the curvy sefir on the R right bar or the N left stroke?

This will become even more relevant for later texts where some uncial and cursive forms start to creep in from other genres/materials into stone inscriptions. Lunate epsilon, sigma, uncial omega etc.