Building the Digital Collection
Creation of Digital Images
The digital images were created onsite by Systems Integration Group. They were captured using a Toyo 4-by-5-inch studio camera with a Phase One Photophase Plus digital camera back, or a UMAX Mirage IIse flatbed scanner.
Most of the images in An American Time Capsule are grayscale. If an item had multiple colors printed on it, or the color of the paper was significant, it was scanned in color.
Each master TIFF-format image, or, archival TIFF file, was produced by scanning the item at a resolution of 300 dots-per-inch.
Following scanning, the TIFF files were used as a basis for the creation of JPEG (reference) images. They were processed (rotating, cropping, contrast stretching, and color balancing) on a SuperMac 900s using Photoshop 5.0. The files were then compressed using proprietary compression software, and saved in JPEG format.
Image Alchemy was used by the Rare Book Digitization Team to create thumbnail-size GIF files to display with the bibliographic information and a larger version for use in the Page Image Viewer.
After completing the scanning, the archival TIFF file, along with the JPEG and GIF files, were stored on the Library's RS6000 World Wide Web server.
Creating Digitized Text
The accompanying text transcriptions were converted at an accuracy rate of 99.95 and encoded with Standard Generalized Markup Language (SGML) according to the American Memory DTD. Due to the nature of this collection, some special rules were applied for markup:
The text was transformed with an OmniMark program to HTML 3.2 for indexing and viewing with web browsers. A related database was designed to make possible the page-turning feature, the links from the full text to the page images, and to provide words and phrases to be indexed for retrieval.
In the transcriptions, capitalization and spelling reflect that of the original document. Where the text was illegible or untranscribable, a bracketed statement of omitted text is inserted, with a number followed by the letter 'w' to indicate the number of omitted words; e.g., "{Omitted text, 2w}" means two words have been omitted. If the illegible or untranscribable text is less than one word, a question mark replaces each omitted character; e.g., "Boston harbor c???d for twelve years" means three letters in the third word have been omitted. Where a good guess could be made, the word and a question mark have been placed in brackets, e.g., "[malversation?]".