Recognition vs. Discovery
The Da Vinci Code by Dan Brown	Cannibals and Kings by Marvin Harris

Have you played with Zoomii yet? It’s a new bookstore—a skin on Amazon.com—that uses a very attractive and dynamic cover-browsing interface. Instead of text, or a mix of text and graphics, Zoomii is all covers, laid out as if they were on an “endless shelf.” The effect is very impressive but also, and with due praise for the ingenuity involved, unsatisfying.

There is no shelf. Part of the problem stems from the “physicality” of the idea. The limits of shelves are the limits of the physical world. Importing physical limitations into the online world is a familiar error. As Clay Shirky remarked in 2005, we ought to be over it.

“People have been freaking out about the virtuality of data for decades, and you’d think we’d have internalized the obvious truth: there is no shelf. In the digital world, there is no physical constraint that’s forcing this kind of organization on us any longer. We can do without it, and you’d think we’d have learned that lesson by now.”

In Shirky’s analysis, not “learning that lesson” results in information architectures like that of the original Yahoo directory:

“Yahoo, faced with the possibility that they could organize things with no physical constraints, added the shelf back. They couldn’t imagine organization without the constraints of the shelf, so they added it back.”

In Zoomii’s case, the whole point was to add the shelf back. It was surely a conscious reversal, and therefore an audacious one, but like swearing off email in favor of handwritten correspondence or communiting in cars in favor of horses, not an efficient one.*

Covers and usability. Zoommii also helped me answer a question I have been struggling toward for some time but never fully worked out for myself: What are covers good for?

If you had asked me a month ago, I would have mentioned Gardnerian “Theory of multiple intelligences,” and the contrast between visual learners and those who do better with text. This concept has a lot of relevance in my own life.** And I would have mentioned how covers were a great way to browse other people’s library.

The truth is, I think, much more simple:

1. Covers are great for recognition, because visual memory is faster than reading.
2. Covers are terrible for discovery, because reading covers, with all their different typefaces and layouts, is slower than reading words.

Transferred to web design, these are fundamentally uability principles, and for the bookstore or OPAC developer up there with any overbroad dictum of Jacob Nielsen—not the full story, but a good rule-of-thumb and starting-point.

In retrospect, this patterns can be seen all over LibraryThing. On the new home page, your recently-added books are shown as covers because you are expected to recognize them at sight, but recommended books are in list format by default, because you probably aren’t familiar with them. This principle also solves why list and cover view are both useful. Cover view is, in particular, a great way to scope out someone else’s library quickly—when you’re looking for commonality, not making a detailed assessment.

Obvious as this discovery is in retrospect—and you may have known it all along—I think it was worth spellng out carefully. In my estimation, bookstores and online library catalogs lack a clear rationale for when covers should be used and when they shouldn’t. Often the idea seems to be that covers add “panache,” which to some extent they do.

But there are some deeper principles at work in the decision to use covers, and the decision to put them on virtual shelves.

*In this vein there’s a good deal to be said from David Weinberger’s Everything is Miscellaneous. In Weinberger-ian terms, Zoomii is a throwback to the “first order of order.” Incidentally, for a quick fly-by of both Shirky and Weinberger, check out Mike Wesch’s Information R/evolution.

**Although obviously a reader, I am an unusual visual person. I learned this when in a group of graduate students preparing to take Latin. We all took a standard learning-styles test so that we understood the idea. The class was perfectly split between visual and textual learners—the archaeologists were visual, the philologists textual. Except for me. I showed up on the visual side. It was a revelation to me because I couldn’t even understand my fellow philologists. Confronted with the task of navigating to an unknown place and offered a choice between a map and a set of directions these people chose the directions? Were they insane?