Special guest post from Jack Dempster and Simon
“Dear mom
Life is good here at Camp RM. Camp Counselor Towell has been very patient in teaching me how to swim in the spreadsheet ocean of inventoried digital records. Pretty soon I will be able to do so without drowning!
Jack Dempster and Simon
There is no surly lunch-lady, no negligent teen counselors, no crazed weirdos lurking in the woods—only the nicest people you can possibly imagine!
Tomorrow night we will swim to the other side of the lake and egg the Archivists’ cabin.
Love,
Jack”
Ok, so maybe doing a Work Learn at UBC’s Records Management Office wasn’t actually like going to summer camp. But it was fun, friendly, and “active” in the sense that I got to test out and apply in real life the principles that I’ve so far only learned about in theory.
As part of my Work Learn position, I undertook a range of tasks. The first, in fact, was not a conventional records management task at all. Every two years, UBC Records Management circulates a survey soliciting feedback on its services, and I was tasked with updating the survey for the 2024 circulation. Although the program I used, Qualtrics, is relatively user-friendly, it still proved surprisingly challenging. A lot of thought goes into designing the architecture and flow of a survey, like who sees what questions and when. And then there are the survey questions themselves. When you ask a question in a survey, you need to know why you are asking that question, and what you would like to know, so that you can gather data that is focused enough to consolidate but open enough to provide you with insights you hadn’t anticipated. And of course, because respondents are giving you their time, you do not want to risk making the survey so tedious that they give up part way through. So it’s important to keep your questions—and the survey as a whole—brief and to-the-point. Designing a survey, therefore, is a very involved process. However, I think we pulled it off. The resulting survey functioned not only to gather information about problems related to records management at UBC; it also advertised services and resources to folks who might not have known about them. We sent the survey out to roughly a thousand people on our contact list. After some delicate cajoling in the form of two judiciously-spaced reminder emails, we received responses from about 236 people, working out to a response rate of 23.6%. Except for our first survey in 2015, this was the highest response rate we have received.
Following completion of the survey, I moved on to my next task, which involved classifying the digital records of the recently decommissioned Peter Wall Institute for Advanced Learning (PWIAS). The PWIAS has a history of eclectic, engaging programming, with conferences, public lectures, and podcasts exploring some of the world’s most pressing issues, and I was excited to dive into their digital records. Using an inventory generator created by previous Work Learn Student, Morteza Rezaei, I created an excel inventory for each series in the PWIAS shared drive. The inventory listed down to item-level, so every single folder and item contained therein had its own row, with information on title, filepath, and size. This was at first overwhelming. Some series were quite large, and their corresponding inventories were tens of thousands of rows long. The spreadsheets were a veritable sea of data. Eventually, however, I became familiar with the structure of these inventories and became adept at scanning through them fairly quickly. I tried to determine function according to title, and at the folder level, since it would be impractical to open and examine each item individually. Occasionally, however, it was necessary to open a file, or sampling of files, to determine the function they served. I would then assign a classification to the record/groups of records by referring to the Record Manager’s bible: the retention schedule. Using UBC’s house-made retention schedule, I determined retention periods and dispositions for thousands of records in the PWIAS shared drive.
Once I had classified the PWIAS records, it was mainly a straightforward matter of determining how long, or whether, they had any time left in their retention period. Where things got a bit tricky, however, was when the disposition instructions for a given classification dictated “SR” or “selective retention” by the archives. In this case, I had to exercise my own judgment to determine whether the record in question had any archival worth; in other words, I had to appraise. I quickly learned that appraisal is one of the most challenging tasks related to records. The responsibility of such a task can feel heavy: with appraisal, the archivist/records manager takes on the role of the grim reaper, by whose judgment a record may live or die. Although there are certain core principles that an archivist may use to guide them in their appraisal decisions (and these I had to learn on the fly from my supervisor, since I hadn’t taken the appraisal course yet), a person might still feel some anxiety about relegating a record, any record, to the dustbin of history. Particularly when you are a person who worries constantly about doing something wrong, and when the results of a decision are irreversible (a destroyed record cannot be resurrected) appraisal can be a stressful task.
Once I had classified all the records and determined their retention periods, it was then time to transfer them to UBC’s e-warehouse for either to temporary managed storage (for records still requiring a period of retention before destruction) or archival back-log storage (for those records with enduring archival value). For this phase of work I used Move-It, another mercifully user-friendly program, which is a godsend for those less-tech savvy amongst us. With Move-It, the archivist/records manager has merely to click and drag the desired folder/file to the upload zone, before filling out the appropriate paper-work, which includes fields for describing the content and context of the transferred records. In the course of uploading the records for entry into the system, Move-It generates checksums for the records as well as a unique bag number. Once these have been generated, the user confirms that the information accompanying the package is correct, then submits the package to the system. One limitation moving information from a shared-drive through Move-It is the file-path length capacity: Move-It cannot accept files with a file-path exceeding 256 characters. Because many file names were exceedingly long and because items were often nested inside many layers of folders, Russian Doll-style, it was often necessary for me to abbreviate or simplify names to shorten the file-path (making sure to log these changes in the inventory). Another limitation of Move-It is its transfer size capacity: individual transfers cannot exceed 20 GB. Because some of the series within the Peter Wall shared drive were quite large (they often included many audio-visual files like videos and podcasts, which tend to be bulky) it was often necessary to split the series in parts for transfer, or even to transfer single items on their own. In these cases it was particularly important for the original context of the records to be expressed in the accompanying transfer records, so that original order could be reconstituted by the archivist on the other side.
And that brought me to the end of my Work Learn term. It was a valuable experience in that I was able to dig my hands into the soil of records work, rather than gazing at the garden forlornly from the high porch of Theory. I also met some cool people, and was able to expand my network of records professionals. And it was paid! A Work Learn position with UBC Records Management comes highly-recommended.