Defaults or Standard Practices

back to introduction

In certain cases we will have to keep the default/standard options or interactions, the users have been practicing for a long time. Changing the defaults might demolish the complete interaction and usability of the interface.

Principle: Defaults within fields should be easy to “blow away”

When a user activates a field, the current entry should be auto-selected so that pressing Backspace/Delete or starting to type will eliminate the current entry. Users can click within the field to deselect the whole, dropping the text pointer exactly where the user has clicked. The select-on-entry rule is generally followed today. (Sloppy coding, however, has resulted in the text cursor dropping at various unpredictable locations. )

Principle: Defaults should be “intelligent” and responsive

Not everything should have a default. If there isn’t a predictable winner, consider not offering any default. It takes precious cognitive cycles to look at a default that covers maybe 25% of the cases and make the decision not use it. That same time could be spent entering the choice actually desired.

Principle: Replace the word “default” with a more meaningful and responsive term

Users rarely have any idea of what “Default,” in a given situation means. (They do know its literal meaning, of course, which is that the bank is going to take away the user’s house. That always cheers them up.) Replace “Default” with “Revert to Standard Settings,” “Use Customary Settings,” “Restore Initial Settings,” or some other more specific terminology describing what will actually happen. User test to find out what terms enable your users to accurately predict what your software will actually do.

Principle: Both your vocabulary and visual design must communicate the scope of a reversion

Make sure, through user testing, that users understand the extent of the restoration: Are they signing up to a benign restoration of just a few recent and localised items, or are they about to spend the next four days re-entering usernames and passwords in every app they own?

User-test your restoration options to find out what users think the result of pressing the button will be. If you are going to do something benign, but they interpret it as potentially destructive, they won’t use the option, leaving them with the same broken or partially broken system that made them consider using it in the first place. Likewise, if you wipe out hours of careful customization without properly preparing them, they may not be nearly as grateful as you might expect. (I once had a young chap in India help me out with a minor problem on my DVR. When he was finished, he had led me to reinitialize the hard disk, erasing every single program on the machine. That was a bit more restoration than I was looking for. After that, I was able to carry out the rest of my conversation with him without even using the phone. I had no idea I could yell that loud.)

When designing tabbed objects, such as properties and preference windows, ensure that the visual design makes the scope of a restoration button clear. Individual tabbed “cards” should be visually separated from the surrounding window so that buttons may be placed either within the individual card or in the surrounding area, indicating whether the button action will apply only to the current tab or all tabs. There is never an excuse for leaving such a scope ambiguous. This is not a fashion decision.