Distinguishability of similars. Usability increases when Sara is able to distinguish among similar item features. Where a sighted user distinguishes similar features, such as among the buttons on a CD player or CD cases in a collection, due to written labels on the items or by seeing each item’s position in a larger spatial context, Sara had much more difficulty. She labeled items in Braille that were otherwise difficult to distinguish and used her hands, fingers, feet, nose, and ears to see what her eyes could not. Each opportunity for sensing and distinguishing can be exploited by technology designers. Design that aids identification and distinguishability of otherwise similar features, such as CD cases with preprinted Braille identifiers and cellphone keys with textured surfaces, enhances ease of use, flexibility, and efficiency. Sara showed how, with deft fingers, she is able to distinguish differ-ent-size measuring cups held together by a common ring. She complained that some cellphones lacked keys she could identify by touch.

Brute-force backup. One fallback problem-solving technique is to exhaustively try all possibilities. When Sara became disoriented while demonstrating her Language Master and JAWS software, she tried all possible options as much as possible. Due to her self-described disorganization, her linear search method for CDs is the most effective tactic, but also very time-consuming.

Flexibility and interoperability. Sara took notes and read books on her BrailleNote, but her model lacks external storage, except for a floppy drive, and does not include access to the Internet. By considering how people will use particular devices when carrying out larger task landscapes, 14 such as that Sara might not only want to take class notes but share these notes with friends using the Internet, designers increase opportunities for use. Despite the many uses of her BrailleNote, Sara wanted a laptop computer due to its flexibility for Internet access, communication, and storage.

conclusion

We draw two main conclusions from the study. The first is methodological. We used an ethnographically inspired

simply replacing
one interaction
mode, such as
the display of text
on a screen with
a functionally
equivalent mode,
as in speaking
the text aloud, is
not necessarily
equivalent from
the point of view
of user experience.

investigation of a single, nonsighted individual interactions with a variety of artifacts in her own home. Our concern with not only usability but socially situated meaning contrasts with experimental designs focused on measurement and control and with lab-based usability studies. Although we can state our conclusions as principles associated with this individual’s preferences and beliefs, such statements are animated in the ways in which we observed them. We are unsurprised that, for instance, Sara’s sense of self is intimately tied to her relationships in her social network. How could it not be? We are, however, surprised in the specific ways in which this was embedded in her choices about artifacts and interactions. Such surprise—in her wall of textually labeled photographs as conversation pieces for her sighted friends, her preference for a tactile watch because of how it looks and feels, and the use, when she was a child, of her talking dictionary as an ice-breaker with friends— undoubtedly reflects our situatedness (as researchers and as people) within our own social worlds, the taken-for-granted assumptions we carry with us as sighted people.

To what extent are we able to generalize these results to other contexts and other people? Though single-per-son case studies are rare in human-computer interaction, they have a long history in the social and behavioral sciences. For example, studies with single participants that have been influential were undertaken by Freud in psychoanalysis, 7 Harper in sociology, 9 and Luria in brain/mind studies. 12 One goal of case-study research is to develop theoretical propositions that can be used to guide subsequent research studies and design efforts. However, it is important to understand the difference between these analytic generalizations and the statistical generalizations common in experimental study designs. Emphasizing this distinction, Yin24 writes, “‘How can you generalize from a single case?’... The short answer is that case studies, like experiments, are generalizable to theoretical propositions and not to populations or universes. In this sense, the case study, like the experiment, does not represent a ‘sample,’ and in doing