W3C home > Mailing lists > Public > www-forms-editor@w3.org > May 2002

Localization-related comments on XForms

From: Yves Savourel <ysavourel@translate.com>
Date: Mon, 20 May 2002 15:35:19 -0600
Message-ID: <PCEIJMPLBANPKDPOPEMLGEDCCNAA.ysavourel@translate.com>
To: www-forms-editor@w3.org
Hello,

The W3C i18n Working Group made recently some of us in the localization
community aware of the progress of the XForm specification and encourage us
to review the current specification.

Here are the one I came up with:


-1--- Translatable elements

I would suggest to provide in the specification a list of the transltable
elements in XForms. This would help clarify any misunderstanding, as several
elements and attributes can have non-translatable data that looks like
source text.

As far as I can see the elements with transltable content would be:
<title>
<caption>
<hint>
<alert>
<help>


-2--- <output/>

Can the result of <output> be formatted? It seems the ref attribute provide
an XPath expression, but Xforms seems to include only the Core Library
Function for XPath and, I think, this does not include the format-number()
function.

For example: the example given in
http://www.w3.org/TR/2002/WD-xforms-20020118/slice8.html#ui-output shows a
value '100.0'. How this value can be formatted to represent, for instance, a
French number ('100,0')?


-3--- Length constraint

The length constraint of an input field is expressed according its type,
defined using XML Schema. The unit for a string is expressed in characters
as defined in the XML specification
(http://www.w3.org/TR/2000/WD-xml-2e-20000814#dt-character).

How the problem of a storage not being in Unicode character is addressed?
For example, a form may define an input field that uses a type of at most 8
characters long, but if the storage for that field is in a database encoded
in UTF-8 or Shift_JIS 8 Kanji characters would be encoded in 24 or 16 bytes:
Entering more than 2 or 3 Kanji would result in some truncation.

Specifying 2 or 3 for maximum length is also not good since it limit
un-necessarily the entry for non-Kanji characters, and leads to possibly
having different types of data for different user-interface languages.

I realize this is not a XForms specific issue. XML Schema uses characters as
the unit for length: it works OK when you stay within the Reference
Processing Model, but breaks down as soon as you leave it, which is often
the case with the data gathered in forms.


Regards,
-Yves Savourel
RWS Group
Received on Monday, 20 May 2002 17:35:21 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 10 June 2009 18:12:11 GMT