Blue Mountain Transcriptions

This repository contains TEI-encoded transcriptions of the avant-garde magazines digitized by the Blue Mountain Project.

The transcriptions are organized by Blue Mountain Identifier (bmtnid); the registry mapping these identifiers to titles can be found HERE.

The transcriptions are machine-generated derivatives of METS/ALTO/MODS data produced by docWorks. The code that generates the derivatives can be found in the Blue Mountain repository.

Structure of Derivatives

The transcriptions have three major parts:

1. Descriptive metadata. Detailed item-level metadata, including the title, creator(s), and major language of each item of editorial content (articles, illustrations). Most of the creators have been authorized against VIAF. The descriptive relatedItem elements are linked to full-text descriptions (see below).

2. Coordinated OCR. Most issues include a facsimile element that contains a translation of the ALTO data for each page generated during docWorks processing. The facsimile comprises surface elements containing graphic elements linking to source page images and zone elements for TextBlocks, TextLines, and Strings (individual words).

3. Structured content. The body of each issue contains a hierarchy of of div elements whose content is likewise derived automatically from the ALTO files. Line-breaks are marked and linked to specific portions of the facsimile to enable coordination and highlighting on page images.

The Pilot Branch

Seven titles were the subject of a pilot project to correct and enhance Blue Mountain's data for an experimental API, Blue Mountain Springs. The article-level metadata was hand-corrected and structural flaws were repaired. The source documents were measured and described; when possible, additional publication information was added.

Licensing

All files in this repository are available for research use.