Ingest and display translation alignments from Ugarit

scaife-viewer / sv-mini

front-end for the Scaife "SV Mini" prototype

https://scaife-viewer.org/

MIT License

3 stars 0 forks source link

Open jacobwegner opened 4 years ago

jacobwegner commented 4 years ago

Data sources:

Discussion when we get to grooming:

Moving away from "left/right" to mapping queries to `passageReferences
How to preserve "orders" in AlignmentSet. E.g, for Digital Sira, we actually duplicated each set in full
Pipeline for Ugarit + TEI XML over to ATLAS

jacobwegner commented 4 years ago

Capturing a few things for future discussion with James / Greg:

What do we want to demonstrate with the existing Hafez data set?
Is it useful to "fall back" to blobs?
Do we need "nested" alignment records to handle Hafez or word-level + sentence alignments?

jacobwegner commented 4 years ago

Had some new thoughts today:

We can still show a "three-way alignment" that is an improved parallel mode (chunk by text parts ref
we could additionally prototype a "chunk by ref" alignment for the Farsi / English Hafez
I can dig up a German alignment chunk too