A little while ago, I saw a video of the first iPhone tests. When I saw the gesture & keyboard input using the fingers, I was thinking : I can do that as well...
So I developed my own finger input keyboard.
Since I don't have a mobile PC atm, I've only been able to run the demo on the pocket PC emulator.
Currently it feels a little slow to me, so I was wondering how it was holding up on a real pocket PC, and if anybody would be able to test it, and maybe even post a little video of the test-app in use.
I think atm the dictionary might have the biggest performance impact. In the current version I support about 75000 English words, while most predictive systems use a dictionary consisting of +/- 15000 words. If the performance part is lacking, I might try to reduce the dictionary to 15000 words also.
A few notes :
- currently only the keyboard support is available for WM5/6, so no gestures
- .Net compact framework 2.0 must be installed.
- only English is supported.
- the prediction is intelligent. a few examples of intended first matches:
* SURROGATE (input text : XUERMFSRE )
* TWEAKER (input text : YSRALDR )
* ABSOLUTE BEGINNER (input text : ZHDIMYTR JRHOKKER)
Since I don't have a pocket PC handy atm, I was also wondering if the finger input method is usable on a real Pocket PC. Please let me know what you think...
Here you can see a screen shot of the app in the windows emulator :
USAGE:
- Just type the words, and use space or ENTER to confirm the first match.
- Press clr to clear the current word.
- Press the text box containing the key input to use this word (currently the words do not get added automatically)
- further keyboard enhancements (QWERTY, numbers,... & skinning) will be developed in the future if this turns out to be usable.
Update 2007-06-19 : new version available (at the same link)
- performance improvements
- QWERTY,AZERTY & numeric keypad
- Tactile feedback (sound)
- ...
Update 2007-06-20 : new version available (at the same link)
- performance improvements after an online chat with tene
You can download it at this page .
Please test it and give me your feedback !!! Thanks !!
So I developed my own finger input keyboard.
Since I don't have a mobile PC atm, I've only been able to run the demo on the pocket PC emulator.
Currently it feels a little slow to me, so I was wondering how it was holding up on a real pocket PC, and if anybody would be able to test it, and maybe even post a little video of the test-app in use.
I think atm the dictionary might have the biggest performance impact. In the current version I support about 75000 English words, while most predictive systems use a dictionary consisting of +/- 15000 words. If the performance part is lacking, I might try to reduce the dictionary to 15000 words also.
A few notes :
- currently only the keyboard support is available for WM5/6, so no gestures
- .Net compact framework 2.0 must be installed.
- only English is supported.
- the prediction is intelligent. a few examples of intended first matches:
* SURROGATE (input text : XUERMFSRE )
* TWEAKER (input text : YSRALDR )
* ABSOLUTE BEGINNER (input text : ZHDIMYTR JRHOKKER)
Since I don't have a pocket PC handy atm, I was also wondering if the finger input method is usable on a real Pocket PC. Please let me know what you think...
Here you can see a screen shot of the app in the windows emulator :
USAGE:
- Just type the words, and use space or ENTER to confirm the first match.
- Press clr to clear the current word.
- Press the text box containing the key input to use this word (currently the words do not get added automatically)
- further keyboard enhancements (QWERTY, numbers,... & skinning) will be developed in the future if this turns out to be usable.
Update 2007-06-19 : new version available (at the same link)
- performance improvements
- QWERTY,AZERTY & numeric keypad
- Tactile feedback (sound)
- ...
Update 2007-06-20 : new version available (at the same link)
- performance improvements after an online chat with tene
You can download it at this page .
Please test it and give me your feedback !!! Thanks !!
Last edited: