Image

Catalog Enhancement: The Struggle Continues

I’m a member of the Public Access Working Group (PAWG, which rhymes with frog, tee hee hee) comprised of members from both Carleton and St. Olaf. We’re in charge of making decisions about our shared catalog, and among other things, PAWG has been working on two decisions this year: whether or not to upgrade to Innovative’s WebPac Pro catalog interface, and whether or not to include “record enhancement” in the form of book cover images from Amazon.

The WebPac Pro decision proved to be easier than we thought. Turns out, the ranking algorithm for it’s much-touted relevance ranking stinks. If any of you are making this decision right now, try searching for the keywords Global Warming. You should notice that even though there are some books that have this exact phrase as a subject heading, they get buried pages and pages down your result list unless the exact phrase “global warming” appears in the title. No kidding. Effectively, the default search is a keyword in title search. Now, you can get around this if you put quotation marks around each word (i.e. “global” “warming”), which deactivates the relevance ranking altogether, but there’s no way I’m going to start teaching that in my instruction sessions. So we’re not “upgrading.”

But the more interesting decision is yet to come. Will we include book cover images from Amazon in our catalog? I’ve heard a lot recently about the benefits of allowing one-click access to book reviews, and I’ve heard that our students are used to seeing images on web pages, and I’ve heard that we should do our best to make the experience of searching the catalog as enjoyable and visually interesting as we can. I agree with all of these arguments. Absolutely. After all, as librarians we’re supposed to aid in locating, collocating, and making decisions, right? And book reviews and book covers both move us in the direction of making decisions much more effectively than, say, the height of the book’s spine in centimeters.

But I think our students are used to much more than pretty pictures. They’re used to everything on a screen being clickable, and everything doing something meaningful and more or less useful. Imagine what you would think if you saw an image to the left of a title in a result list. You’d click on the image, right? I would. And I’d get sucked to Amazon. Talk about mixed signals.

Now, I would realize that this wasn’t what I wanted or what was expected of me, and I’d hit the nice little “back” button in the browser and click on titles from then on. But my co-workers and I have noticed a distinct trend among our students. Instead of using our A to Z list to see if we subscribe to a journal or magazine, they find the publisher’s web site. Then they come to us and ask, in all seriousness, if the library has a fund that will help them buy this article online. They’re usually very happy when we point out the A to Z list. But last week one poor kid came back to me after he’d verified that we did not, in fact, subscribe to his journal and wondered if now I’d help him pay for it from the publisher’s site. That lucky kid learned about the A to Z list and interlibrary loan all in one evening.

These students who expect to be able to do whatever it is that site they land on tells them to do. What’s more, they expect that we expect them to do these things. The web isn’t just about seeing things; it’s about doing things. And if we put something in front of our students that implies they should do something that we don’t actually want them to do, that’s a problem.

Of course, we could by images from places like Syndetic Solutions, but that’s expensive. Still, this would be a way to make our records look pretty without threatening mass confusion.

So the upshot of all of this is that I’m willing to be convinced, but I have grave reservations about the amazon solution to our boring catalog. If we’re really interested in providing what users are used to, we’d have to redesign everything, not just slap on some pretty pictures. And that “everything” would have to include all sorts of clickable and interactable options.

But in the spirit of being convinced, is anyone willing to share their experience with the amazon lipstick option? I’m particularly interested if you have any statistics to show how often those images get clicked on. Do your ILL stats change?

Just as a side note, we’re also pondering how best to configure our link resolver… To offer choices of electronic options or not to offer choices, that is the question. Whether ’tis nobler to privilege one-click access to beauteous PDF full text or to deepen the understanding of young minds as they develop access preferences from among the panoply of database options. To learn, perchance to dream. To dream of other uses to which they can put databases that appear often in their list of options….

Sorry Will. You’re writing was a lot better, and certainly more moving.

Technorati tags: ,

5 thoughts on “Catalog Enhancement: The Struggle Continues

  1. >The WebPac Pro decision proved to be easier than we thought. Turns out, the ranking algorithm for it’s much-touted relevance ranking stinks.

    This drives me nuts. While the current way of returning keyword search results isn’t great, at least you can choose to sort by title, date, or “relevance.”

    Under WebPac Pro, there is no way around the grouping by relevance. Which is problematic, to say the least.

    I think it is great that III is taking on the big problem of relevance ranking in the catalog, but the current version seems pretty half-baked.

    This is frustrating because there may be things that we want (like getting RSS feeds from catalog data) that we can’t get without upgrading to WebPac Pro.

    I hope that III hears these complaints (voiced loud and clear on the IUG email list) and figures out a way to let us disable this “feature” while still using WebPac Pro.

  2. Exactly, exactly, exactly. Up until last week, we were pretty set on the idea that we’d go to WebPac Pro. We were just testing it to see if there was any reason not to do it. And then there was.

    In other news, though, somebody from III must have read this post because they called us to see if they could better explain RightResult. I look forward to hearing what they said, but it won’t change the results of the test searches we did. I’m looking forward to their 2007 upgrade, though, if they can figure this one thing out. I really want RSS, not to mention all the other things they say are coming.

  3. Well, I’ve heard it from the horse’s mouth, so to speak. But I’m not quite sure I should post the story. So, if you’re interested, drop me a line.

    Short story: we were right, but they can’t tell us why because it’s proprietary, but there may be a fix of sorts coming. End of short story.

  4. Regarding a clickable Amazon book cover image, how about just displaying the image w/o it being clickable? That’s what we do (although we use Content Cafe images): http://opac.cadl.org. Our original desire was to change the link so that it went to the full record page of our catalog, but that proved impossible. So plan B was to remove the link. Then the title is the most tempting click, which gets them to the full record and keeps them in your catalog. In the full record view, we use what we’ve labeled “more info” to link to review content. That content opens in a new window. The relevant WWWoptions are WEBADDRESS and BIBIMAGE. That’s how we resolved these issues.

  5. We thought about making things un-clickable, but unfortunately Amazon won’t let you use their content if you don’t link back to them.

    If we do end up including the link to Amazon reviews, I’d like to have it in the form of a button that says something like “Look for reviews on Amazon,” but there’s no legal way for us to get the image without the link, unfortunately, and it’s the images that people really seem to want.

    I like it that I can’t click on your images, and the images look nice, too, and there don’t seem to be many books that don’t have an image. (We noticed that there were a lot of “no image available” returns with the Amazon option). Does your Content Cafe license require you to include some attribution, and if so, where to did you display it?

Comments are closed.