I recently moved and I seem to be turning into the User Interface Police. I am sure that, back in 1982 when they wrote the user manual for my "Generation II" microwave, pushing nine buttons just to set the clock made perfect sense. The UI Police will let these folks get off with a warning—it was written in 1982, after all. I had hoped that household appliance manufacturers had gotten a clue in the last two decades, but my spiffy new vacuum cleaner came with a 40-page manual, complete with obscure line drawings and text in 9-point font. Evidently, there is still a need for good user-manual writers.
In my work, however, I'm now on the other side of the equation, and suddenly I'm realizing that it's all those stupid users out there, not the quality of the documentation, that's the issue. In one of my volunteer roles for an association, I walk people through the procedure for signing up for the association's private email discussion list. It doesn't matter that there are large links for "Log-in Help" and "Change Password." Perhaps RTFM (shorthand for "Read the F***ing Manual," stupid) should now stand for "Read the Fabulous Monitor," fool.
A more humbling experience has been in building a wiki that I will be using to gather information from a wide range of info pros. I am struggling to build navigation tools, home page links, a FAQ, and a concise explanation of the wiki's goal. It's all clear in my head—the challenge is anticipating where people will encounter problems or ambiguities, and figuring out how best to resolve their issues. What is painfully obvious to one person is utterly opaque to another. (Well, I still don't understand what part of "Log-in Help" is ambiguous, but that's another matter.)
I have been reviewing the advanced search pages of the major search engines and have come to the realization that we still have a long way to go in terms of intuitive search interfaces. Take, as an example, how search engines present advanced search options.
Both Google and Yahoo! do a good job of including links from the main page to the advanced page and providing good documentation. Unfortunately, Live.com and Ask.com hide their links to the advanced search option. Worse, Gigablast has an advanced search page with very few options, but it also provides an extensive list in a help file of some really powerful tools for advanced web searching. The advanced search page isn't actually for advanced searchers. Go figure.
A9.com has some interesting tools, including the ability to compare web search results with Amazon.com and a number of library catalogs, news sources, image databases, RSS feeds, Furl, and a number of other non-traditional web sources. You can even create your own specialized search engine, selecting among 400 specialized search tools. However, it seems they went out of their way to make their search results screen as user-hostile as possible. When I show the search results to A9 virgins, they practically recoil in horror.
On the other hand, my favorite among the search engines is Exalead. Its advanced search page not only explains the various search filters with examples, but automatically inserts the proper syntax into the search box when a user clicks that filter. Their UI designers really nailed it.
So, will we ever see a truly innovative and well-documented search interface? Yes, most econtent sites have figured out how to simplify the process of building a search query and reviewing the results. Some sites even provide some clustering and sense-making of the results—most commonly, an analysis of the most frequently-occurring keywords in the retrieved articles. My hunch is that most amateur info-seekers have no idea how those "suggestions" or clusters are compiled, and we info pros already know how to search by controlled vocabulary terms, thank you very much.
Better yet, I eagerly await a well-designed graphical interface for building search queries and analyzing results. Factiva has taken a stab at this in its Discovery Panel in its Search 2.0 search results page. But I wish there were more initiatives from the econtent companies to help info pros visualize the most common elements among search results, find unexpected relationships among concepts, and identify the outliers that may provide an unanticipated counter-perspective to the common wisdom. I don't know whether a visual search interface could be truly self-explanatory. My hunch is that this will remain a long-term dream of info pros. Let's just hope that, when it's rolled out, it has a good help file.