Digest Topics

Authoring & editing
Budgeting, reporting, & metrics
Case studies
Content management
Enterprise knowledge management
Government applications
Graphics & posters
Learning & Education
Legal applications
Ontologies & topic maps
Personal knowledge management
Point of view articles
Publishing
Reviews
Studies & surveys
Research & searching
Slides & presentations
Taxonomies & metadata
Technologies & tools
Trends & strategies
Usability & user behavior
see also A - Z index

Metaphors in user interfaces

Article in focus: "Interface lift" (Amy Wohl, IEEE Spectrum, November 2005).

More articles on this topic: See "usability and user behavior" in the Digest index or the Montague Institute Review index.

About the author. Amy Wohl is a consultant, newsletter publisher, and columnist for VARBusiness magazine. Prior to that, she was Executive Editor for Datapro Research Corporation, an IT research firm now owned by Gartner Group. She has a B.A. and an M.A. in economics. Her focus is primarily computer technology news.

About the publisher. IEEE Spectrum is the flagship publication of the IEEE (Institute of Electrical and Electronics Engineers), a professional association with more than 380,000 members.

Article summary. The next big thing in user interfaces is unlikely to be a single, all-purpose design, such as the current one that mimics folders and file cabinets. Instead, three new "metaphors" are emerging:

1) Web browsers (e.g. Firefox and Internet Explorer);

2) special purpose interfaces for browsing large document collections (e.g. Grokker and IBM's WebFountain);

3) personal knowledge management programs (e.g. Evernote).

In designing new user interfaces for computers, you can either use radically new metaphors, software, or devices — and risk alienating experienced users — or you can make incremental improvements that exploit the familiarity and work habits of a large user base. It's hard to do both.

What I liked. Interface metaphors are useful in expanding our view of how we interact with computers. For example, Wohl points out that FedEx drivers, retail clerks, and nurses have joined the ranks of "knowledge workers" alongside research scientists, business intelligence researchers, and marketing managers. Computer interfaces for these new knowledge roles may come not from the office but from the life sciences, the health care industry, or even the world of computer games.

Commentary. I guess I shouldn't be surprised when IT commentators miss the two most important information metaphors of the last 500 years — books and libraries. When libraries morph into portals, reference librarians cede their role to search engines, and books become electronic files, why bother with quaint artifacts like card catalogs, A - Z indexes, glossaries, and tables of contents? Two reasons: familiarity and utility.

The problem is that while old metaphors still have value, they need to be adapted to today's publishing realities, where databases, Web sites, text, graphics, and interactivity merge. It's a delusion to think that people who've grown up using card catalogs and A - Z indexes will be satisfied with search engine results — even with fancy starburst graphics or embedded taxonomies.

On the other hand, it's crazy to ignore ways to extend the library and book metaphors with modern electronic tools, giving users the ability to:

• search the full text of a book;

• create and annotate a personal "card catalog" of sources (including books, Web pages, electronic documents, and people);

• connect source materials to contact information about the people who authored or recommended them;

• share and link personal libraries with colleagues through a system of categories and keywords.

For an example of a traditional back-of-the-book A - Z index updated for use with an electronic journal and Web site, see the Montague Institute index.

Is the the library/book publishing metaphor is too expensive for today's infoglut, as many IT professionals seem to believe? After all, who has time to catalog and index millions of documents by hand? This view makes sense if you look at corporate intranets as giant information warehouses. More realistically, they are aggregations of personal and departmental libraries, coffee houses, war rooms, jam sessions, labs, and news rooms as well as factories and counting houses.

The concept of "expensive" is relative. A $1 million expense might look significant if potential savings per unit of output are relatively small, largely invisible, and hard to measure. On the other hand, the same figure might look paltry if the upside is large (e.g. a blockbuster new product or a multi-million dollar deal) and the outcome is highly visible. The "expense" might shrink even further if participants are willing to create, categorize, and tag documents on a volunteer basis (as they do in information sharing communities like Flickr and del.icio.us).

Metaphors based on things (e.g. books or inventory items), transactions (e.g. sales), or numbers (e.g. finance) tend to be backward-looking, passive, and focused on cost reduction. Metaphors based on relationships and collaboration (e.g. customer lifetime value, new product development, or new business models) tend to be forward-looking, proactive, and focused primarily on quality and results. Cost is a secondary consideration.

Metaphors are seductive but tricky. For example, Wohl uses a retail metaphor to describe how clerks become "knowledge workers" by entering inventory data into a store's database as they stock the shelves. But this is only the back end of the retail metaphor. On the front end, where the customer meets the inventory, many metaphors are available — independent boutique, global discount chain, "embedded" store (e.g. college bookstore or hospital gift shop), electronic garage sale, or something not invented yet. In designing user interfaces, there's too much focus on back end metaphors and not enough on the front end.

I like to play with new programs and devices as much as anyone, but what I find most interesting are products and services that make it easy for people to express their own metaphors using computer tools. This, I think, explains the popularity of services like Google maps (which allows you to add a geographic dimension to data such as crime reports and real estate listings) and Filemaker (which allows non-technical people to create database applications). These products turn the idea of "usability" on its head and change the focus from back end to front end. Traditionally, programmers write applications and test them on users. Another alternative is for users to create applications and tinker with them until they work hand-in-glove with the task at hand. User-developers would only need to consult a programmer when they get stuck or need to exchange data with another application.

Bottom line. Metaphors are great, and we should certainly open our minds to new models for human-computer interaction. But at the same time, we need to broaden our perspective by adapting metaphors from libraries, print publishing and other models, and reassess the way we calculate the cost of creating, organizing and presenting information.

For original articles by Jean Graef, see the Montague Institute Review.