If I Could Build a Library Catalogue

10.12.2009

I know many librarians blanch at the idea of blending commerce and information retrieval – and to be fair, those concerns aren’t completely without merit. Still, I think that there are certain user experience practices that are common place in commercial enterprises that, if we introduced them into our own library catalogues, would delight library users. The catalogue is a logical place for this kind of integration.

cecily's library catalogue design

If I could build a library catalogue, the search results page would look a little like this. Starting from top to bottom, I’ll highlight some of what I think are the more useful features:

  1. Users will be able to sign into their accounts using their library credentials, or, if they are not yet library cardholders, they can click the “New Member?” link to register for a temporary library card.
  2. The shopping cart convention that is so familiar to anyone who has ordered items online is carried over into the interface. When a catalogue visitor adds an item to their cart, the book is held until the patron is ready to check it out. Ideally, a patron would be able to choose if they wanted to pick the book up in 24 hours (for an additional fee), or if they would like to pick the book up in 48-72 hours for no charge. When the book is ready for pickup, the ILS would send a text message, email, or automated phone call to the patron to let them know the item is ready. Delivery would be lovely, but I don’t think it’s feasible.
  3. The results page would display items that matched the search in a grid, and would include cover art for each item that would help the user identify the item on sight. The results page would highlight the search term at the beginning of the list, complete with the number of items that were returned.
  4. The Quick Info button would trigger an overlay that contains detailed information about the desired item. The user could read a review of the item, add a rating, and add the desired item to their cart.
  5. Clicking the “Add to Cart” button that appears underneath the title in the results grid adds the item to the patron’s cart automatically. This allows them to bypass the ‘quick info’ layer if desired, which speeds up their transaction.
  6. Patrons can customize the interface by viewing results on a grid or in a list.
  7. One of the complaints I hear most often from patrons is that when lots of results are returned, pagination becomes a chore. By allowing patrons to choose how many results to display – either in smaller chunks, or viewing all items at once (a feature that isn’t available in our current catalogue), we put them in control of how much information they want to see.
  8. The sidebar contains links that help users refine and limit their searches further. For example, if the user only wanted to see DVDs that match the search results, she would click the DVDs link under “Narrow by Format” and the results display would change to show only those items.

I’m aware that many products like Aquabrowser and Bibliocommons already have these features rolled into their catalogues. We don’t have this type of discovery layer in our catalogue (yet), but I’m pretty sure it’s on the horizon. There is much room for improvement in a design like this. For example: is it too much information? Are there too many options? How can such an interface be made even more simple and streamlined? Is there a danger that we aren’t showing the user enough information?

If you could design your own library catalogue, what features would you include? What would you leave out?

  • http://twitter.com/sheatsb @sheatsb

    I know you've probably seen this already, but what about Koha (http://koha.org/)? It's pretty user-friendly from the end-user and admin perspective.

  • http://twitter.com/sheatsb @sheatsb

    I know you've probably seen this already, but what about Koha (http://koha.org/)? It's pretty user-friendly from the end-user and admin perspective.

  • Cecily Walker

    Yep, I've seen it, though I've never worked with it personally. I like the idea – but the conventional wisdom I've been hearing is that systems like Koha and Evergreen aren't well suited for library systems as large as mine. I don't know how true that is because I haven't done the legwork, but in just looking at the list of libraries who are using Koha, I don't see anyone that's a comparable size.

    Still, it's good to see competitors adopting these kinds of features for catalogues. I think the major vendors are definitely taking notice.

  • http://intensedebate.com/people/cecily cecily

    Yep, I've seen it, though I've never worked with it personally. I like the idea – but the conventional wisdom I've been hearing is that systems like Koha and Evergreen aren't well suited for library systems as large as mine. I don't know how true that is because I haven't done the legwork, but in just looking at the list of libraries who are using Koha, I don't see anyone that's a comparable size.

    Still, it's good to see competitors adopting these kinds of features for catalogues. I think the major vendors are definitely taking notice.

  • http://brettlwilliams.wordpress.com/ brett

    A lot of this stuff is structurally impossible with the current crop of ILS without significant hacking, The best projects I've seen (including Aquabrowser) require you to export your data to them in order to display it properly.

    My ILS wishlist would include direct links to each item, a clear, non-authenticated (or simply authenticated) POST search support, an open, XML based API that does not require you to hack around with YAZ, RSS, the ability to easily integrate with social media (share on twitter, facebook, etc).

    Then I could make an interface do all the things your asking about without having to have a degree in computer science.

    You might want to check out the Drupal for Libraries project as well as the Scriblio project. Solr shows some promise as the basis for an OPAC., Solr projects like the Blacklight OPAC are still being activly developed and are easier to hack.

  • http://brettlwilliams.wordpress.com brett

    A lot of this stuff is structurally impossible with the current crop of ILS without significant hacking, The best projects I've seen (including Aquabrowser) require you to export your data to them in order to display it properly.

    My ILS wishlist would include direct links to each item, a clear, non-authenticated (or simply authenticated) POST search support, an open, XML based API that does not require you to hack around with YAZ, RSS, the ability to easily integrate with social media (share on twitter, facebook, etc).

    Then I could make an interface do all the things your asking about without having to have a degree in computer science.

    You might want to check out the Drupal for Libraries project as well as the Scriblio project. Solr shows some promise as the basis for an OPAC., Solr projects like the Blacklight OPAC are still being activly developed and are easier to hack.

Previous post:

Next post: