The beginning of the abstract for this presentation provides an excellent intro:
"With the proliferation of rich Internet applications and interactions more closely aligned with how people think, we face some interesting challenges:
- Do we design for one common audience and common tasks, or tailor applications around specific audiences and their unique activities?
- How do we resolve the tension between creating simple applications that ‘do less’ and the demand for new features that some people really do need?
- As we move beyond usability to create desirable interfaces, how do we handle a subjective domain like emotions?
"These types of challenges could all be addressed by creating a truly ‘adaptive' interface."
Examples of ways that applications could be/are being tailored for niche audiences (i.e. "the long tail."):
- Expanding text box to accommodate use
- Using IP address to guess at correct location of user
- If the person attaches multiple files, give him the number of fields he typically needs
- Assume info based on meeting types—lunch meeting pulls preferred lunch spaces.
- Look for iterative actions and reveal features over time to tailor to need
- Increase the button size if this user consistently misses it
- Color and saturation to indicate age and importance of content—old data fades in color (see: ShaunInman.com)
- Prominence of a help link minimizes over time if you never use
- To fit smaller/different display sizes, changing layout (liquid versus fixed) and content (disappears to fit in new spacebased on display)
- For mapping application, collapse neighborhood driving instructions based on history.
- Changing [help] text based on audience? (travel agents versus travelers)
- Text changes based on regional differences (Coke, pop, soda)—just like we’d do in a conversation
- Removing L-shaped global navigation when applications are displayed
- Changing help based on what you know about the person (novice user gets instructions for simpler tasks like drag and drop)
Challenges:
- These kinds of options/features are currently being done primarily with cookies. Could also use rich profiles.
- OpenID may provide more options when small bits of info about a user can be put together to paint a bigger picture (which, of course, if also a danger).
- Requires a clean separation of content and the application.
- First, get the basics right (metaphor of using rough grit sandpaper first; fine detail later)
- Disclose what you are doing.
- Provide opt-outs (i.e. "Please logout if you are not [Fred]."
- Be wary of changing spatial organization (users will look for the same content in same spot)
- Ease into this kind of info
- Could use Web 2.0 model of testing live on users
Links for more info:
- Slide show presentation from SlideShare
No comments:
Post a Comment