Mini-Qwerty Substitution Errors
The majority of mini–QWERTY keyboard substitution errors are row errors accounting for 55.3% of the substitution errors and 22.2% of the total errors.
Xuebing Du
Peter Solarz
let's talk about Bridgerton tea, my ask is open

@theartofmadeline
KIROKAZE
🪼

blake kathryn
almost home
styofa doing anything

pixel skylines

Kiana Khansmith
Claire Keane

Love Begins
hello vonnie
Misplaced Lens Cap
we're not kids anymore.

shark vs the universe

No title available
Monterey Bay Aquarium
trying on a metaphor
seen from Ireland
seen from United States

seen from Malaysia

seen from United States

seen from Germany
seen from United States
seen from United States

seen from United States
seen from United Kingdom

seen from T1

seen from United States

seen from United States
seen from Germany

seen from United States
seen from United Kingdom
seen from Switzerland
seen from France

seen from United States
seen from Malaysia

seen from Malaysia
@pushhandle
Mini-Qwerty Substitution Errors
The majority of mini–QWERTY keyboard substitution errors are row errors accounting for 55.3% of the substitution errors and 22.2% of the total errors.
Errors on mini–QWERTY keyboards are...
Evenly distributed with substitutions still occurring most frequently (40.2%) followed by insertions (33.2%) and deletions (21.4%).
In comparison, there are relatively few transpositions (5.2%).
The Twidor software Phrase Set
The Twidor software was configured to use the MacKenzie and Soukoreff phrase set [MacKenzie and Soukoreff 2003], a set of 500 phrases representative of the English language. The phrases range from 16 to 43 characters with an average length of 28 characters. The phrase set was modified to use only American English spellings and display only lower case letters and spaces (no punctuation or capitalization).
T9 Text Entry Method
Common words:
“The user only presses each key containing the desired character once. T9 computes all of the permutations of letters that correspond to the sequence of key presses entered by the user and uses a dictionary to disambiguate the results. It then presents a list of words to the user with the most likely word appearing first in the list. For example, to input the word “truck” using T9 a user would type “8-7-8-5-2”. T9 displays the most commonly used word with the “8-7-8-5-2” number sequence: “usual”. In this case, “truck” is the second most commonly used word with that number sequence. In order to get T9 to correctly output the desired word, the user must select it from the list presented.”
Custom words:
“In cases where the user inputs a word that is not in the dictionary, the user is asked to input the word a second time using multi–tap. She then has the option of adding that word to the T9 dictionary for all future lookups. “
Typing rates:
“T9 typing rates range from 9 wpm for novices to 20 wpm for experts [James and Reischel 2001].”
8 - 16 WPMÂ Multitap Phone Keyboard Performance (2001-2)
“Novice multi–tap users begin typing at approximately 8 wordsper minute (wpm) [Butts and Cockburn 2002; James and Reischel 2001] with experienced users reaching speeds in the 16–20 wpm range [MacKenzie et al. 2001; Lyons et al.2004].”
Are we satisfying desire for handling mobile email in 2015?
“Wireless email is a feature desired by 81% of consumers” (2006)
31 - 59 WPM on Tactic mobile keyboards (2006)
“Subjects average over 31 words per minute (wpm) at over 95% accuracy for the first session and increase to an average of 59 wpm at almost 93% accuracy by the twentieth.”
To me, error analysis is the sweet spot of improvement.
Don Norman