how to test different key-board lay-outs?

13 Jan 2009 - 4:51pm
5 years ago
1 reply
593 reads
majid dadgar
2008

*Greetings,
*
*1 -* *what are the approaches/ methods towards the most optimized lay-out
on key-boards?*(in other languages)
According to the nature of the language and alphabet, for example to
minimize cognitive load, to enhance typing speed, to distribute letters on
keyboards equally for both hands, ... .It would be a stage before the user
test.e.g. that if I have to find the 100 most frequent words, what will be
the approach to identify those words?

*2 - In the event that I have come up with the most optimized lay-outs(e.g.
5 final lay-outs), How could I test them by the users feedback?*
The challenge for me is how to present final lay-outs to the defined users
so that I can have their feedbacks for choosing the most optimized one. to
see how they re-act to the new lay-outs. Actually I am seeking the simple
practical and scientifically acceptable way(if any). How can I specify
different lay-out letters to the keys on key-boards? Is there any program
that could define different lay-outs(manually) for keyboards?

*Best Regards,
Majid.
*

--
Industrial Design, BA,
University of Tehran.
P.O.Box: 15635-119, Tehran, Iran.
once upon a time Design
http://1to3Design.blogspot.com

Comments

14 Jan 2009 - 9:01am
Barbara Ballard
2005

On Tue, Jan 13, 2009 at 3:51 PM, majid dadgar <majid.dadgar at gmail.com> wrote:
> *1 -* *what are the approaches/ methods towards the most optimized lay-out
> on key-boards?*(in other languages)
> According to the nature of the language and alphabet, for example to
> minimize cognitive load, to enhance typing speed, to distribute letters on
> keyboards equally for both hands, ... .It would be a stage before the user
> test.e.g. that if I have to find the 100 most frequent words, what will be
> the approach to identify those words?

Testing keyboard layout is difficult ... not in the actual task
(simply develop a set of normal text entry tasks for your device) but
in getting participants who don't know on particular layout.

I'm going to assume you are designing for a mobile device. Why?
Because I get about 5 inventors per year asking me to endorse their
new mobile text input. And they're going about it the wrong way. It's
not my input they need; they need to invest $5m and 3 years in
business development.

Why else do I assume mobile? Because keyboard layout is far more variable.

Some keyboard inventors analyze letter combinations occurring in the
language, and lay out keys accordingly. FrogPad was designed in this
way.

Optimized layout based on letter frequency or common words does mean
you'll have a large learning curve in the resulting design. So you'll
need to add a training task to your user studies.

Don't just use the most common 100 words. Use the power of a computer,
use the most common 10,000 words but weigh them by frequency of use.
Add in common abbreviations in the language, decide whether you want
to actively support slang, and don't forget NAMES. Consider whether a
key might enter multiple keystrokes, like "Al " to make typing "Al
Jazeera" that much easier.

Layout isn't the only issue. Key size is huge. If you have limited
space you have to decide whether to use very small keys or larger keys
that do similar things.

Don't require participants to learn more than one new layout. You'll
have extra errors as a result; your results won't mean anything.

~~~~
Barbara Ballard
barbara at littlespringsdesign.com 1-785-838-3003

Syndicate content Get the feed