However, to our knowledge, none of the alternative virtual keyboards have been very successful so far, although they have been available for several years. This is a clearly shorter learning time than the over 100 hours found in the Dvorak vs Qwerty typewriter study ( Yamada, 1980). The user studies by Smith and Zhai (2001) and MacKenzie and Zhang (1999) indicate that the initial typing speed is moderate (around 10–15 words per minute), and it takes about 4 hours net training time to surpass the typing speed of the Qwerty layout. The results indicate that the alternative layouts could indeed increase text entry speed, even as much as about 50% ( Zhai et al., 2002)! It should be noted that speed is, however, not the only usability factor, and very few studies with actual users have been carried out. Therefore, their typing speed can be estimated by Fitts' law analysis ( Fitts, 1954 see also MacKenzie, 1992). The use of alternative virtual keyboards is based on straightforward pen tapping with one hand. Some proposed layouts are, for example, Fitaly ( tested by MacKenzie et al., 1999), OPTI ( MacKenzie & Zhang, 1999), Metropolis ( Zhai et al., 2000), and ATOMIK ( Zhai et al., 2002). Since the Space character is very common, many alternative layouts provide more than one Space key. Sometimes also completely new keys are introduced.

Since the virtual keys are pointed at with a stylus, the alternative layouts typically change the positions of the keys to minimize the time and effort needed for stylus movement. Therefore, it is no surprise that alternative layouts have been proposed also for this category of keyboards. Virtual keyboards are presented on the device screen and are therefore easy to change through software.

Miika Silfverberg, in Text Entry Systems, 2007 Alternative Virtual Keyboards

Virtualkeyboard browser