W3C home > Mailing lists > Public > public-i18n-its@w3.org > October to December 2006

RE: Action item - Result format proposal

From: Yves Savourel <ysavourel@translate.com>
Date: Wed, 13 Dec 2006 13:07:46 -0700
Message-ID: <742F71D71E424346A47CA8BE9B70DCB00A3BA7@lupus.RWS.LOCAL>
To: "Felix Sasaki" <fsasaki@w3.org>
Cc: <public-i18n-its@w3.org>
Hi Felix,

I have no issue with using InnerXml rather than InnerText for
locNoteText/locNoteReference and termInfoText/termInfoReference, since,
as you pointed out it makes sense for ruby.... We might as well be
consistant for all 'content'.

Using a qualified namespace for the helper elements is also fine, but
I'm not sure if I get the its:translate part. I don't think we should
have any ITS makup in the result file (except if it's part of the inner
xml for selected contents).

Otherwise we would be using ITS incorrectly for no good reason.
When we say 'translate' in:

  <node path="/book" outputType="default-value">
   <output translate="yes" />
  </node>

We mean 'the node /book of the input file was translatable' not 'the
content of <output> in this file is translatable'. So why use the ITS
namespace for this (like the following code would)?

  <node path="/book" outputType="default-value">
   <output its:translate="yes" />
  </node>


Would something like:

<r:nodeList xmlns:r="urn:ITSTestResult">
 <r:nodeList datacat="translate">
  <r:node path="/book" outputType="default-value">
   <r:output translate="yes" />
  </r:node>
  ...
 </r:nodeList>
</r:nodeList>
 
Would work for you? With InnerXml for the 'content' results like locNote
and termInfo instead of text.

-ys
Received on Wednesday, 13 December 2006 20:08:01 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:12:48 GMT