I went in to TAMI yesterday for another day of research. The first article I opened to read stated within the first paragraph that you can not get from theory to method without practice. After reading about appraisal for about a month, I felt like I was kind of treading water, pretty ready to move on to the next phase of the project. Most of the literature about appraisal is written with regard to traditional paper archives, with a strong focus on government records. I have read a few pieces during this research period that are more media-oriented, but most of the literature is saying more or less the same thing, and none of them can address problems specific to TAMI. We are not a traditional archive, and our concern is really more with our online library. I was ready to talk structure, how to get from theory to method, so I sat down with my advisor for a couple of meetings.
First, we decided that I would begin work on a draft proposal of VHS appraisal guidelines. I will have a draft ready by next Monday that outlines a tentative process for evaluation. We can test those guidelines then adapt and improve them as necessary.
We discussed some of the issues I outlined during research and possible solutions. First was representation and if we should be concerned with representing events and culture within an individual collection or a set of collections from a Round Up or sub-region of Texas. We decided we should only evaluate videos against others in its own collection, but that representing eras or time periods within that collection should remain a concern. If we use segments of a lengthy VHS in the online library, each segment should remain its own file (not a string of segments edited into one video), and each segment should reference the original VHS. I will need to use TAMI’s established naming convention for segments so that their source is clear and provenance remains intact. Videos not selected for use will still be saved on external back-up drives, so even if they do not go into the online library, they do exist on our servers and can be accessed if need be. Lastly, in the interest of determining the use of our human resources, I will track my time spent on cataloging and indexing a VHS collection so that we have an idea of how much time we are investing in each tape.
An aspect of online curation that my advisor raised is that she employs a strict rule when choosing videos for the library: Is this video appropriate for a 4th grader? Many of our videos are used in lesson plans or as primary source materials for classroom teachers, so this is something that she must keep in mind when selecting videos for our very public online library. While some videos may be uncomfortable to watch (in the sense that they are challenging or about topics such as war or the Great Depression), no video should blatantly disgust a viewer. This element of appropriateness is one that I need to include in any criteria for evaluation.
Another possible criteria, aesthetics, is one that is very relevant for moving images. Film often has a great aesthetic, as do many scenic videos of Texas. Some of that is lost in VHS, and we discussed the possibility of adapting such a criteria more to one of “watchability.” Is there minimal white noise? Is there light? Is it steady? Is it valuable to watch, if not beautiful. I will work to flesh out this idea as I write my proposal.
A final idea that I will work on is that of sampling. As I mentioned in previous posts, so much time is spent looking at every single video to identify those that are needed for our library. Even in this post, I again referenced how much employee time is spent looking through these videos, many that we do not end up using. Currently, the practice at TAMI is to create a record of every digitized video that comes through our archive. In the database, the video file is linked to an assigned file name for the video in the collection. We discussed doing very light cataloging, giving the video a cursory viewing so that we can document at least an idea of its content, and, to maximize employee time, only performing deep cataloging on selected videos. This could very well be the best practice, but as a part of this project, we are also interested in pursuing the idea of a statistical sample. Is it possible that by taking a sample of a set of tapes, either small selections of every tape or somewhat larger selections of fewer tapes, using either random sampling or systematic sampling, we would end up with an accurate or equitable representation of a collection without expending significant time and resources? Using a collection of video tapes, I will apply a yet to be determined sampling method to the tapes. Then, using the same collection, go through each video individually, making a record of each one. We hope to see how this would affect VHS appraisal, and if a statistical sample might serve to cut down on a great amount of time while still producing a true representation of the collection.