Bohairic corpora desiderata

[ctschroeder]

In order to begin NLP tools for a Bohairic corpus, we would need:

• minimum 5000 tokens of publicly releasable (CC-BY) Bohairic text in unicode
• manually tokenized according to our standards (possible to try existing Sahidic tokenizer, but there would be a lot of errors)
• manually pos-tagged in Gitdox
• lemma list or lexical resource of some kind? (check with Hany Takla and Eitan Grossman)