The Library of Congress Veterans History Project Home  
Experiencing War: Stories from the Veterans History Project

The items that comprise these collections were donated or collected by veterans, their families, friends, historians, folklorists, all volunteers. The types of materials are many and conditions in which they were created vary. Much of the printed and photographic materials are the originals and the library has taken steps to display them in their truest state, with little or no digital processing. Some items received by the project are photo-duplications or photocopies of originals; others may be copies from copies. These images can vary greatly in their degree of quality. Some digitized images have received post-processing to be made more legible in the digital domain. The equipment in which audio and video recordings have been created ranges from broadcast quality to consumer grade, therefore the sound and image quality will vary from file to file. Minimal pre-or post-processing has been performed on these recordings. The Veterans History Project is displaying the highest quality images that have been made available by the donors.

Sound recordings

The sound recordings in this presentation were transferred from the original analog audiocassettes to digital audiotape (DAT) to produce a master source for digitization. Wave (wav.) files were created from the DAT tapes at a sampling rate of 44,100 Hz per second, 16-bit word length, and a single (mono) channel. The RealAudio and the MPEG 2, Layer 3 (.mp3) files were derived from the Wave files using the .mp3 and RealAudio plug-in of Sonic Foundry's SoundForge software. Some surface noise may be apparent on the recordings, and files may start or end abruptly, as on the original recordings. Minimal adjustments to volume and equalization were made to certain files.

Moving Images

The moving images in this collection were transferred from their original videotape formats to Digital Betacam to produce a master source for digitization. MPEG-2 files were created by the contractor Vidipax in New York, with the Optibase MPEG MovieMaker 200S hardware card, and controlled by Optibase's MPEG Fusion user interface with the total data rate at 3Mbps. The RealMedia-G2 files were derived from the MPEG-2 files through Helix Producer Plus and created for users who have at least a 56K modem. The data rate is 450 kbps and the video size is 352x240. Some background noise may be apparent on the recordings and files may start or end abruptly, video and audio levels may change as on the original recordings. As often is the case with streaming media, image quality will be affected by modem speed, network traffic, processor speed, and other variables.

Photographic Material and Manuscripts

Photographic material and manuscripts were scanned on site at the Library of Congress Information Technology Services (ITS) scan lab using overhead digital cameras with Phase I software, and post-processed using Adobe Photoshop software. Images were scanned in either grayscale or color mode, depending upon the original, at 300 ppi and saved in the uncompressed TIFF file format. An uncompressed master image file was produced for each photograph. For display on the Web, we generate derivative JPEG images of varying sizes for display on the story, page turners, and photo album pages.

Text Transcriptions

Documents are transcribed with minimal changes to the original text in an effort to preserve original appearance, content, and idiosyncrasies of composition. Period language and terminology are also retained. Transcription is literal with regard to the writer's capitalization, punctuation, spelling, and visible corrections (i.e. crossed out words). Spelling errors are indicated with [sic]; however, recurring errors in spelling or grammar within a single document have been marked the first time and and not subsequently. Crossed out words are presented {in braces}. Any special emphasis (i.e. underlining) is presented *between asterisks*. When the writer spells out the word "and" it appears spelled out; when a symbol is used for "and" the editor used the ampersand (&).

Alterations to the original text are consistent throughout and all editorial comments appear in [square brackets]: a single space is placed between sentences, nontextual symbols (circled page numbers, scribbles, etc.) have been changed to plain typographical characters when possible or omitted with a note, obvious slips of the pen have been silently corrected, and malformed letters are presented as the intended letter, based on spelling and context. All superscripts and underscored superscripts have been lowered to the regular line of text and are not underscored. Single word interlineations are indicated with a single caret (^) before the word; multiple word interlineations are presented ^surrounded by carets^.

  Home >> Building the Digital Collection
  The Library of Congress
  March 4, 2004
Veterans History Project Home
Contact Us