CHAPTER 6
Data Input
6-20 Writing, Drawing, and Editing
No matter how you have configured recognition for a text field, users can
input the wrong type of text if they try hard enough. For example, a user
may manage to input numbers where words are the proper type of input.
This happens because every kind of text recognition uses the built-in symbols
dictionary, which includes all digits and some punctuation together with all
letters of the alphabet. Users have to try hard to input an improper type of
text because recognition is weighted towards the proper type of text, and
that is what the recognizer usually returns. But if an input stroke is vague
enough, the recognizer might return an improper type of text, such as a
number instead of a letter. Thus users can manage to write virtually anything
in any text field.
Here is a complete list of recognition options that applications can control:
■Allow users to control recognition with a Recognizer button on the status
bar, in a slip, or both.
■Separately enable or disable recognition of words, shapes, and
editing gestures.
■Recognize words letter-by-letter without using dictionaries.
■Improve recognition of complex stroke groups in which users tend to
put large spaces, such as telephone numbers. This is accomplished by
disregarding spatial cues (distance between gestures or strokes), and
relying solely on temporal cues (time between the end of one stroke and
the beginning of the next one) to determine when a user has completed
a group of strokes.
■Make the Names dictionary or any other built-in dictionary the primary
dictionary. (The Names dictionary contains common names for a user’s
locale, not the names contained in the built-in Names application.)
■Use custom dictionaries with specialized words such as product names or
plant species.
■Limit word recognition to one or more of the built-in dictionaries (proper
names, surnames, first names, names of days and months, names of
countries, names of states, names of cities, names of companies, abbrevia-
tions of states, common words, and user’s words).