|
|
Translation Manager |
|
| Home | Project Documentation | Mailing Lists | Site Map |
A tool that helps
| Status | Source Tree | Download | Developers |
|---|---|---|---|
| Being designed | Available | Unreleased | Joseph Koshy <jkoshy[@]users.sourceforge.net> |
Open-source documentation changes very rapidly, like open-source software development itself. A tool that can automate tasks involved in managing translations would help increase the productivity of the translation teams.
We need a tool that can help translation teams track ``source'' documents across revisions and changes. Source documents for a translation team could be structured documents like SGML text or unix manual pages (troff input), or unstructured ``flat'' files like READMEs and HOWTO documents. Changes between two documentation releases could involve structural changes (pieces moving around) as well as content changes. Whitespace changes interspersed with content and structural changes can make the task of determining what has changed difficult.
The translation manager tool is meant to help translators by making it possible for them to concentrate as far as possible only on content changes. Whitespace changes and structural changes are to be brought forward automatically.
The tool interfaces with popular document revision control systems like CVS allowing translators to work with the benefits of having a revision control system backing their work.
Further, in order to assist translation tools working on systems lacking builtin support for Indian languages, the translation manager will provide multiple kinds of input methods, including transliterated input and support for using fonts having arbitrary internal encodings.
There are quite a few tools around that provide assistance to translation teams translating GNU gettext message catalogs, for example, EMACS has a gettext-mode and there is also a full-fledged GUI based message catalog editor poedit. There don't seem to be tools that can work with multiple documentation input formats and which provide for different input styles.
Information about suitable Indian language input methods is somewhat scarce. Hopefully the Indic-Computing Handbook will have more information for developers on this topic soon.
This project is being prototyped in Python so Python programming skills are a must. Awareness of cross platform portability issues is a plus.
None
4 months (approx)
Currently on the task list of: Joseph Koshy <jkoshy[@]users.sourceforge.net>.
This project can benefit from having many eyes and hands on board. Please join up.
|
Copyright © 2001--2009 The Indic-Computing Project. Contact: jkoshy |
View document revision history
|