Good Interaction Design Trumps Smart Algorithms

Over on the new CACM blog, researcher Tessa Lau has an interesting post on three common misconceptions that folks have about HCI.  I recommend reading the full article, but I would like to call attention to her provocative opening statement (emphasis mine):

I come to the field of HCI via a background in AI, having learned the hard way that good interaction design trumps smart algorithms in the quest to deploy software that has an impact on millions of users. Currently a researcher at IBM’s Almaden Research Center, I lead a team that is exploring new ways of capturing and sharing knowledge about how people interact with the web.  We conduct HCI research in designing and developing new interaction paradigms for end-user programming.

One of my biggest grievances with web scale search engines is that they have made the assumption that smart algorithms (or, at least, simple algorithms trained with enough data to be made smart) are more important than good interaction design.

The amount of interaction design in web search is negligible.  The user is given a single-line input box, 10 blue links, and almost no ability to interactively refine or re-sort results, and no explanation (why the result was retrieved) or contextualization (in what information neighborhood the result is located) to help understand the results. Web search engines prefer handling all the intelligence behind result ranking on the back end, non-transparently, in a black box and hidden away from the user’s view.  They place more value on their smart (data-rich) algorithms than they do in good interaction design.

It would be interesting to know how, throughout her career, Tessa Lau came to exactly the opposite conclusion.  At the end of her blog post, she says that there will be “more on this in posts to come.”  I look forward to those posts!