Naming Conventions

Editions

Page Sections

Print editions are divided into volumes as well as alphabetical sections. For OCR processing, we organize the materials around the alphabet and subdivide all entries into smaller groups of ≅225 pages. We refer to these as page sections. Sections are designated by their alphabet letter followed by a 2-digit number for their order, beginning with 01. Thus section s01 is the first 225 pages in the letter "S," section b02 begins with the 226th page of "B," and r03 covers 451-675 of the letter "R."

Entry Sections

After pages are converted to entries, we add the print volume number to the section name, following the letter. Thus s01 becomes s1101, when it occurs in volume 11 of the print edition. Sections are eliminated after initial processing of the files, See entry Folder for more information.

Filenames

We use three phases in the process of creating machine-readable text from historical Encyclopedia Britannica editions.

Page phase

The process begins by creating text from single pages of the source. The OCR is output as docx files, which we transform into TEI using XSLT. Most of this work is done using page files.

Entry phase

We use Pythonto convert the page files into files containing one entry each, entry files. These files use different file-naming conventions than the page files, adding 2-digit sequence to the end that indicates whether it was the first, second, third, etc. entry on the page. We also add the last four digits of the source image filename, to insure a unique ID for each page within the entry and for the entry file itself.

Master phase

We add index terms to entry files, we change their name to master files. This metadata is stored in the <teiHeader> of the master file. All digital editions created by the project are generated from these masters.

* Abbreviations: ed = edition; img = last 4 digits of image name; ltr = letter; pg = page number; sec = section; seq = entry sequence on the page; vol = volume.

Folder Names