PPEA Imaging Protocols September 26, 2005
Image Processing Standards
Before beginning your own imaging project, contact the Archive, since some aspects of the processing standards for our published images may change from project to project, or to suit newly developed display software.
|For projects deemed to be adapted to current posted guidelines, step-by-step documentation is available on the How To Page.
Postings on the status of projects already underway in the University of Virginia 551 Alderman Library office are available on the Imaging Status Page.
Imaging Copy of Record (COR)
Imaging COR pertains not only to the folio images for an edition, but also to any "abbrevs" (images of suspension marks), samples of erasures or any other images to be referenced anywhere in the front matter or text.
It also includes the entity files that accompany such images, to make them displayable electronically, since any change in the one affects the other.
In the early days of the Archive, most jobs were done by a very few people who were in constant and direct contact with one another, or by individuals working remotely who handled most aspects of their work alone. Each of these conditions made it possible to have little regard for workflow management, or the maintenance of Copy of Record, because the chances were slim indeed that any two people would be working on the same file without knowing it, or depending on a file being in a state ready for publication without also being personally responsible for its being in that state.
This remained true of the imaging work on the Archive until very recently, when a number of people working remotely began to depend for their work on possession of the exact copy of the images that was expected to go to press.
As a result, issues of image COR began to arise: What was it? Who "owned" it? How could one know? These questions were addressed in a way that would be the most accessible to anyone using copies of the images pre-publication.
Image COR Basic Procedure
The basic procedure for maintaining COR of images is quite simple. TIFFs as received (or locally produced) are ripped to a reference set of JPEGs, always known as PLEBEs. CDs of these PLEBE images are sent to each editor, and are also kept in 551 Alderman and 404 Bryan Hall at the University of Virginia. These CDs always have metadata inserted into each file that specifies their non-publishable, early-reference-only character. This metadata can be read using the Photoshop File | Image Info function, but it is also stored on the CDs in tabular format, in a .csv file, readable either as plain text, or in most spreadsheet programs.
The PLEBE images are NEVER Copy of Record, since they are never considered to be publishable. They are useful for making the initial transcription, and for proofing it for letter forms and color. They must not, however, be relied upon for any but the most obvious color distinctions such as that between blue and red paraphs. Shifts in text or correction ink color can be invisible even in the TIFFs, and even when they are quite clear in the manuscripts itself, so such fine-grained distinctions as that should be made on the basis of on-site inspections only, with only perhaps a list of spots to check made up from the PLEBEs or later versions of the JPEGs.
Lists of spots to check with ultraviolet because of possible erasures are best left to the Pre-Publication Proof stage, since PLEBE images can be slightly blurry, and can have optimization artifact that can either disguise or simulate the traces of erasure.
Pre-Publication Proof Sets
A first Pre-Publication Proof set of images is then produced, and the metadata is updated to reflect this fact. Pre-Publication Proof sets are named after famous proscribed Romans, in alphabetical order. So far, sets have used the names Antony and Brutus. Negative comment on an Antony set invokes corrections, and the production of a Brutus set. The up to date information on which set is COR of the images is posted on the private web site Imaging Status Page.
If you possess images of an earlier version than is listed on this page, you should use them only for reference, until you obtain the updated/corrected images.
It is naturally prudent to check with the Archive before basing a great deal of work on the images you possess. A glance at the image metadata and a call or an email assures that no mistakes will be made.
It is also prudent to destroy old copies of images immediately when you have received a new set.
Image COR Procedure Outline
The following is a breakdown of the Image COR procedure:
Images purchased by the archive have varied greatly in quality as they have come from the originating libraries. Although every attempt is made to produce a high quality set of images for publication, the results have varied greatly, with some images having been scanned from slides while others have been shot directly from the manuscript with high-end digital cameras at very high resolutions.
Even in the best of all possible worlds, images scanned from slides can reach an effective resolution only as high as the film grain on the slide - around 3000ppi at best. Color digital images shot as TIFFs which used to be done at 600ppi, or twice the resolution of the typical printer of the day, are now shot at 300ppi, well under the resolution of many relatively inexpensive photographic printers now available.
Given this abiding lack of uniformity, the PPEA produces the best images for publication that can be made available for any given project. Adjustments must be made for each project, but overall, the following criteria have withstood the test of time:
For step by step instructions on how to process images for publication, see the How To Page on the private site.
Revised on the following dates: September 26, 2005, July 28, 2005.