For the last eight weeks, I have been sitting at the Smithsonian looking at pretty pictures. Specifically, I've been sorting through three thousand digital objects selected for addition to the Galaxy of Images, figuring out what they are and what to write about them. These images, culled from texts headed for the Biodiversity Heritage Library and holdings selected by Cooper Hewitt, have depicted everything from Mayan ruins to equine scabies — okay, so they’re not always pretty — and each have required the creation of descriptive metadata in order to be made searchable in the Galaxy database. It has been a fascinating summer journey — although I’ve been pretty stationary, I feel like I’ve traveled and seen a lot.
Résultats des campagnes scientifiques accomplies sur son yacht, 1889-, Mastigoteuthis magna Joubin (Oceanic squid)
United States, Alaskan boundary tribunal, 1903, The British Dominion in North America …
Sometimes coming up with metadata — simplistically, searchable data about data — was simple: an easily transcribed image caption was provided by the author or publisher below the picture, or an accurate and concise index within a digital copy of the original volume could be consulted for plate or picture information. But frequently it was trickier: sometimes the text was faded, misspelled or inconsistent, in another language entirely, or not even there at all — in cases like these, further research was necessary to figure out what exactly what I was looking at and how best to describe it for others. In the case of the multitude of mollusks — and all the other pictures I looked at from the animal world, whether birds, bees, or the Hoary Bamboo Rat — a common or taxonomic name at the time of publication might have evolved or changed completely over time, which also necessitated tenacious investigation. And the transcription and creation of captions was just the first step; each image also required the creation of simple keywords and assignment to an existing Galaxy collection heading.
Keywords were certainly another challenge: it was immediately necessary to reflect the caption, the catalogue record, and the intent of the author, but should tags also include, say, the color of an image, describe multiple or minor elements within the frame, or strive to consider the depiction from more abstract vantage points? Optimally, as much as possible should be provided, but finite restraints of time only allow a picture to be examined for so long. It was often the case that when I had the chance to return to an image, I found something else to add, or a new way of looking at the object in front of me based on later related discovery. Tagging all of these pictures was definitely an iterative process, and something I got better at it with time and practice. The project was also instructive as to how helpful it is to encourage user participation when it comes to cataloguing visual objects — the Galaxy of Images invites users to submit tags, and I encourage everyone reading to visit the Galaxy and contribute in this manner. Please improve my keywords! Point out any spelling mistakes! Improve my German translations!
John Riddell, Architectural designs for model country residences, 1861, Villa. No. 4, Front Elevation.
Overall, I am truly fortunate to have been able to intern at the Libraries through the Alberta Smithsonian Internship Program, a government initiative in my home province (in Canada) established to fund and facilitate learning opportunities for Canadian students here at the institution. Because of this cross-border partnership, I have had the opportunity to participate in a unique project enabling the practical application of skills that I’ve been acquiring as a Master of Library and Information Studies candidate at the University of Alberta — and the chance to meet and work with many of you! I'm grateful to have spent two fun and humid months in DC, and extend my thanks once more to the Smithsonian, Erin Clements Rushing (my supervisor), and each of you for being such excellent hosts.