I put together this Excel-based Holdings Overlap Scenarios Toy as a prop for thinking about cooperative print archives during the run-up to WeST and some shared print projects with the University of California system. On opening the spreadsheet, you will see a randomly generated scenario for a 100-part serial held among a batch of institutions, each of which has a different level of completeness.
The number of copies of each issue that is available to the community is show along the left, along with some indicators of the likelihood of begin able to reconstruct a complete version of each issue, and thus the whole serial, from the partner holdings. You can twiddle the drop down lists to change the mix of institutions along the top and in the small table on the “inputs” tab you can change the level of completeness assigned to each type of institution.
This is a toy, so there are any number of things that it does not account for: the famous article or illustration that was stolen from so many institutions, the relative likelihood of older issues being unavailable versus newer issues, and the list could go on. Furthermore, this toy was created in the context of regional planning. The “alert” icons (a check box for more than 5 holdings, an exclamation point for 1, and an X for 0) are intended to be meaningful in that context. Candace Yano’s report, “Optimizing the Number of Copies Required for Print Preservation,” work that was reflected in the Ithaka S+R “What to Withdraw” report, suggests a curve where 12 copies often emerges as the safe threshold. The holdings overlap toy uses 6, on the assumption that we would want to have several print retention networks.
What the toy does suggest is that the ingredients for a successful cooperative project ought to exist. There is good reason to anticipate that a cooperative group of libraries construct complete series and (the crucial and for a successful collaboration) that each library involved has access to a more complete version with its partners than without.
In almost any scenario, the level of completeness across the entire group is notably (10-20%) higher than the completeness of the most complete individual members. Also, in many scenarios, it is interesting to note the way that many “small” or “tiny” libraries can add value to the whole system. There is, I find, some received wisdom at the intersection of the preservation and collection development community that assumes the big academic collection stand alone. What this toy suggests, and this has been born out in the WeST experience, is that the big collections are necessary hubs and natural archive providers, but that they are dependent on the broader network to achieve completeness.
Herewith, The Holdings Overlap Toy.
While the number 12 seems like a good and safe number of copies, the resource that is used to verify the copies is of some concern. If it is World Cat then it is important to remember that many instituions do not update their holdings when materials are withdrawn. While it is a great deal of trouble to verify it seems necessary before disposal. In a random sampling done some years ago to verify the actual physical volumes recorded in World Cat the number of “missing” or otherwise not available copies was high. This was a very small sample but suggests that further research into the area is warrented before mass weeding is done.