Browse Prior Art Database

XML Automation for Localizing String Resources

IP.com Disclosure Number: IPCOM000019863D
Original Publication Date: 2003-Oct-02
Included in the Prior Art Database: 2003-Oct-02
Document File: 2 page(s) / 35K

Publishing Venue

IBM

Abstract

Abstract: Disclosed is method for automating the translation process of localized strings in software application to allow for deployment to multiple markets. The localized strings can be part of graphical user interface or message files.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 2

XML Automation for Localizing String Resources

Overview The automated translation method is based on placing localized strings in a locale specific XML (extensible markup language) text documents and then automating the translation process through the use of a translation software that utilizes the XML document and a 3rd party translation engine to automatically generate the locale specific versions of the original XML file. The locale specific XML document can then be loaded at run time based on the current user machine locale.

Solution Details The solution is broken down into multiple stages:

a. Development stage: Primary language strings (such as English) are defined in the primary language XML file. XML file is fed into the automated translation application to automatically generate the language specific versions of the XML file based on supported languages. Specifically, an automation model receives as input the name of the primary language XML file. It then interacts with the underlying automated machine translation engine to auto generate the equivalence of the primary strings into the targeted language. This interaction involves the automation model engine's usage of its knowledge banks and heuristics gained from the feedback it has received from previous encounters with the human translators acting in the capacity of translation advisors and reviewers. Furthermore it should be noted that the automation engine model does not depend on any particular machine translation engine (i.e. the translation engine can be that of IBM or developed by any other 3rd party vendor). The result of this phase is a new text file containing the translated strings in the same XML file format as expressed earlier above. The translated file is included in the product install image under locale specific sub directories. Application components can look up and display the locale specific string from the XML file using a string ID which corresponds to the XML node (see file format detail below).

b. Verification stage: The translated product is installed on supported locale specific machines. Human translator verification of application interface ensures proper context-specific translation. Any defects found in this stage are automatically applied to the generated XML file (no re-build is required). Defect(s) are reported back in an intuitive grap...