5 Replies to “WorldCat Discovery and Pluralistic Searching”

  1. I really appreciate your post because, entering a week where access is our main topic, I hadn’t necessarily been thinking about the search tools people use to find content, but these things are so important to our discussions, because how people find things dictates so much of WHAT they find as well!

    I think something that Sadler and Bourg surprisingly didn’t discuss is the hierarchical nature of search tool readouts: almost all of them give a list of “hits,” from top to bottom. There are certainly ways of destabilizing this: Instagram and image-based websites will give not a single column of images, but rather three, and sites like Padmapper and Zillow can give map-based results for apartment and house hunting. I wonder if there aren’t more visual ways to represent search results– that would not only deconstruct some of the narratives set in place by search engines, but also give users a better sense of the interrelated nature of items?

    1. That’s a really great point, they don’t so much challenge the concept of hierarchy or relevancy rankings, rather they seek to somehow make the hierarchy more fluid by changing how something is deemed more relevant.

      I think one option that I’ve heard talked about before is displaying search results as books on their shelf, so showing the user the book on its shelf and the surrounding books. A criticism of library catalogs and discovery systems is that they prevent browsing. When you look for a book on a shelf, people are often drawn to look at the books shelved around the book the were trying to find. This is also an effective research tool; when you’re looking for a book for your research, related books will be shelved around that one book. I think this would be a great way to display results not in a hierarchy, give a sense of the interrelated nature of items, and just genuinely provide a better, more effective search experience. Creating such a display… not sure how that would work though!

      Edit: thinking about this more a bit later, I think such a display would still end up with the same hierarchy (the bookshelf with the “top hit” book would still be on top). I’m not sure if this would be effective, but maybe display an entire book case and highlight a range books that would be relevant. The user could mouse over and see what each book is and then click on the book to see its record. I think that would certainly could be a better option but seems like it would be a difficult project.

      1. I think the “bookshelf” display solution can reinforce certain hierarchies, but it brings back ideas of order that are deconstructed by search engines. The items can be listed in an order deemed to be best by an archive, rather than what the engines deem to be “relevant.”

        I like the idea, though, because you would want to display a “book” or entry in the center, and move out from there. The question is, if someone had multiple hits, how would they be displayed? If you have a series of “shelves,” then we still have this list-like solution, but you could also highlight “books” on a couple of shelves– with different colors for levels of relevance, depending on the search terms? Maybe? This still is a bit wonky, but might provide some solutions to established hierarchies in search features. I think this is an interesting case where form defines function– how we design things to look greatly affects how users can interact with them.

  2. When I worked in cataloging, I loved/hated thinking about ways to implement browsing in the library of the future. I still think it is extremely feasible technologically, if only we could get a few top software engineers to care about browsing for a couple of weeks. (Given our discussions recently, maybe we just need some software engineers who are NOT white men?)

    I think it would be simple to implement virtual bookshelves that you could filter or re-arrange based on various criteria. You could just use the traditional call number order for starters, but then reconfigure in different ways. You could even have historical data: let me browse a replica of the Cambridge library that existed in 1850, or whatever. Or limit a results set to recent publications, or to those in the public domain, or whatever.

    A fun (though a little clunky) example of a virtual library is John Peel’s record collection, which you can browse at http://johnpeelarchive.com/albums/

  3. I am so glad you highlighted this issue, Ben. The Sadler and Bourg reading (and especially that heartbreaking story you mention) really underlines the degree to which the assumptions inherent in the tools we use can unexpectedly shape user experiences in ways that are counter to our values as archivists.

    In an ideal world every library would be able to create their own bespoke access software, but given the resources and skills available to the average institution its probably not realistic. I think a more realistic path is for archivists like us to find ways to participate in the development of our own tools, whether they are open source or commercial projects.

Leave a Reply

Your email address will not be published. Required fields are marked *