[IxDA Discuss] Managing Translated Text Length

[Barbara Ballard]

On 10/19/06, Chris Stygar <cstygar at daed.com> wrote:
>
> We are working on a project for an embedded system GUI that will be
> translated into several languages. We have designed the layout around a
> set of US English text strings that will be presented in a proportional
> (non-fixed width) fonts. We did our best to allow for expansion when the
> strings are translated, however there a still several instances where
> there will only be a finite amount of space to fit a translated string.
>
>
> Does anyone have any reliable strategies for managing translated string
> lengths that will minimize the chance of forced truncations?
>

I would make a table of character widths in the font, then create a
little program or spreadsheet to calculate the width of an arbitrary
string, then give the target pixel count to the translators along with
the program or spreadsheet.

Of course, in some mobile design you don't know the character width
since you are using the native font on the device.  This strategy
won't work there.

Depending on the platform you are using, you can use some
platform-native strategies. Java ME, for example, lets you specify
minimum and preferred size for fields in a Form.

Finally, you can simply give a best guess for string length, in
typical-width characters, and ask for an abbreviation for any string
that exceeds something like 5 characters fewer than that.

-- 
Barbara Ballard
barbara at littlespringsdesign.com 1-785-550-3650