Re: [ACAT] What users want

Posting to Autocat

On 12/07/2012 21:57, Brenndorfer, Thomas wrote:

Being partly though Daniel Kahneman’s Thinking Fast and Slow (, I’m skeptical that using “what users want” as the sole or primary baseline for consideration for change is a good idea.

Rather it seems one should always turn to looking at results. As an example, adding MARC records for e-books to the catalog improves usage, even though there is a separate, “book-store” like interface that users can access. I doubt that if asked users would say for certain that they “want” one versus the other, or describe things in extensive meaningful and actionable ways that will result in interface or catalog redesign.

There might be broad outlines. Users may overwhelmingly indicate that they prefer an online catalog versus a card catalog (I know which one I would pick!). Users may reflect on the pleasure they get with instantaneous results from Google searches, but this can lead to false intuitive perceptions, as Kahneman would put it.

“Want” seem like it has little to do with it. I would rather have things work to the point where users don’t have to think about things, but get the results regardless. That doesn’t always happen with full-text, keyword-based, relevancy-ranked searching. When shopping for books, or computers, or other spec-laden products, I always prefer a good database-driven catalog, where I can sort and filter comprehensively on plenty of fields, with lots of controlled vocabulary terms and facets, in addition to list-making capabilities where I can rank products under different categories. There’s a tremendous sense of accomplishment, due to the recognition of the slow, hard work involved, when I’ve feel I’ve completely mapped out a subject domain in my mind. Can’t do that with a Google search result, even though I might occasionally search out a term or company using Google, and be pleased enough with the fast results to be almost deluded into thinking that I’ve achieved some information-seeking skill level that’s better than others.

The problem with this way of thinking, that “want has little to do with it”, the extreme flip side would conclude that we should create products that “the users don’t want”, because “we know better than they do” and I hope nobody thinks that. The trick is to approach the task from the user’s point of view, not from the supplier’s view. As an example, when computers were introduced, many in the typewriter industry would think, “What new kind of typewriter do people want?” Their own world was focused on typewriters and they couldn’t really imagine a world where people didn’t care at all about typewriters. People just wanted a decent way to not have to write something out by hand and when PCs appeared, it turned out that people didn’t want any kind of typewriter anymore. The technology had changed in fundamental ways, and this changed the public’s expectations. The typewriter industries that couldn’t accept that fact failed and disappeared. It is only after the supplier has discovered in an unbiased manner what the public wants, can those suppliers go on to thinking how to respond to the wants. It doesn’t mean that the supplier just gives whatever the users say they want–it is much more subtle than that, of course. For one thing, you look at what your competitors are building.

In my paper at ALA, I showed some areas where the public has become quite vocal in their wants: variations on information overload, filter bubbles, paradox of choice and so on. This is not at all a complete list, but I think that the library field as a whole can help solve each and every one of those areas I noted. To do so however would/will demand deep changes in the library’s traditional work. I have absolutely no doubt that someone will give the public what they want eventually, just as took place when people wanted PCs instead of typewriters. The typewriter companies that kept making typewriters withered and died unmourned by most, but lots of other organizations that didn’t care one whit about typewriters were happy to take their place. We live in a world where most people care as much about our catalogs as they do about typewriters! That is an inescapable fact. The patrons don’t care about library catalogs at all. They want something else.

Still, there is some residual good feeling toward libraries themselves, although very few patrons really understand what libraries are and what they do. I hope libraries can find it within themselves to change enough so that they can provide people with what they want, but even if they don’t, I am sure that someone will, someday.

Relating this to catalogs, I think that the library catalog–when seen within the entirety of the library enterprise–provides far more to the public than the arcane and antiquated FRBR user tasks, or anything that can be found in Cutter or the rest of the cataloging literature. To figure out what the catalog really and truly provides will demand radically new thinking however, and this new thinking can be helped along by finding out what the users really want. Then, librarians can imagine new tools that address those needs and they can be created and tested.