Issues and questions
Should the cataloging scheme be reorganized to arrange documents chronologically?
Lorrie: Just to clarify, are you asking if certain genres of digital files such as the registers and minutes should be renumbered (e.g., 1_1_1, 1_1_2, 1_2_3, etc.) so that they are ordered chronologically? If that is the question, then I would say definitely no. The individual identifying numbers correspond to the ID numbers on the physical documents, the latter of which would be far too time-consuming to change. Retaining that ID link between the digital object and physical object will be critical for researchers to locate the latter (or vice versa). Also, since the CSH collection always has the potential to grow as staff come forward with donations, any attempts at chronologically numbering the documents would be aggravating.
That said, on the user end of things, it would be extremely helpful if users had the option of viewing the catalog, or what I would call the inventory, as a chronological list. For example, they would could see a list of the registers as they were accessioned (e.g., 5_1_1, 5_1_2, etc.) or they could see them in roughly chronological order. One thing to keep in mind is that for the registers especially there are some overlapping dates.
Should sub-directories that split documents based solely on numbering be retained? Is there a rationale for the creation of such sub-directories that justifies retaining them?
Lorrie: At least in the case of the photographs, Crowley appears to have arbitrarily (at least it seems so from the outside) grouped the digitized images into folders. So in short, I do not believe there is a good reason to retain the sub-directories, although a preservation metadata note specifying how the images were grouped by Crowley would be useful for documenting the digitization process.
Which artifacts of digitization should we preserve?
Sub-folders created by Crowley for photographs within a sub-series should be recorded. These were artifacts created in the original digitization process and we are unsure why Crowley created the folders. The original sub-folders should be recorded for each image, although the value of this data is still unclear. There should also be a note at the sub-series level that the sub-series was split by Crowley initially.
What is the archiving unit?
The unit will vary by the type of record. A preliminary agreement seems to be that a scan is the simplest unit for archiving and provides the most flexibility for retrieval/access.
What should be included in an archival package?
Possible decisions: metadata + digital master only, metadata + digital master + all derivatives created so far, metadata + digital master + one access copy
Decision depends upon storage cost and size considerations on one hand and time-to-access after requesting on the other (derivatives generated on-the-fly)
Another factor in the decision could be how tightly coupled the digital library is to the dark archive.
Where would/should we add preservation notes?
At the item and series/sub-series level, as appropriate and as necessary.
What is the relationship between how the physical CSH collection is organized and the item identification numbers?
Lorrie: During the summer of 2010, Halima Davis and I started processing the CSH materials starting with the board meeting minutes. As a result, the unique item IDs given to the materials in this series unfortunately do not correspond to their respective subseries numbers like the rest of the collection material. Instead, the IDs start with the state board minutes, which were loose-leaf photostatic copies stapled together by month. All board minutes were given the series number 1. State board minutes start with the subseries number 0 and end at 34. with each number representing a year. So, for example, all the state board minutes that begin with 1_3_xx are from 1955 and those that begin with 1_4_xx are from 1956. [Note: There are gaps in the state board minutes series, and not all years are represented. The subseries of 0 was necessary because this item was found at a later time and had to be added in to the existing series structure.]
Continuing on with Series 1, the Local Building Committee Minutes are all 1_35_xx and the State Hospital Board Photographs are 1_36_xx. The six volumes of early CSH Board Minutes were given the unique IDs of 1_37 through 1_42. [Note: the IDs for the CSH board minute volumes do not have three placeholding numbers like the other items in this series because they are bound volumes covering multiple years. If I could process these over again, they would have IDs that indicate their subseries...]
The relationship between how the rest of the series in the collection are organized and the unique ID numbers is much more obvious. The first number in the unique ID is the series, the second is the subseries, and the third is the item number within that subseries. For example, item 5_2_6 indicates that the item is part of Series 5, which are patient records, and is under subseries 2, which are ward books. It is the sixth item of this subseries.
As a final note, only the digital files that correspond to the materials that Halima and I processed utilize this naming schema. The digital files created from the microfilm (i.e., patient files) and from the CSH collection at the Library of Virginia following Crowley's naming conventions. These are the files on HDs 1, 2, 3, and 7.
As per the original finding aid on Box, subseries 2 is represented by folders 01_000_* to 01_034_* on disk8. And all these folders represent board minutes from 1939 to 1986 (with a few years missing as per the finding aid). So for this particular group of folders, what should our item group ids be? (E.g., for series 1, subseries 1, we have item groups 37 thru 42.)
Also, what should be the item group ids for series 1, subseries 3, represented by folders 01_035_* on disk 8?
Similarly, what should be the item group ids for series 1, subseries 4, represented by folders 01_036_* on disk 8?