FAQ

From Open Siddur Project Development Wiki

This is an attempt to gather answers to frequently asked questions (FAQs). It will be expanded as more questions are asked (so we know which ones are asked most frequently!)

Technical Questions

Why not use OCR to speed up acquisition of siddur texts?

Optical character recognition (OCR) works well to identify characters for English texts, and, in fact, we do use OCR on Latin-alphabet texts, such as the Singer Prayer Book and the 1917 JPS. We have not yet found any software (open or closed source) that will recognize Hebrew letters with vowels with sufficient accuracy to make the proofreading effort less work than manual transcription.

Every once in a while, we test the software that is available to us to see if it has matured enough. The open source hOCR (Hebrew OCR) project, led by Kobi Zamir, has advanced the farthest towards this objective. If you're an open source programmer and interested in OCR then please help develop HOCR. It shows a lot of promise!

What do I need to start transcribing Hebrew texts?

You will need to download and install Unicode fonts and a Biblical Hebrew Keyboard Layout. We've prepared documentation for your keyboard setup that explains how to install and configure your keyboard layout, step by step.