Re: Roadmap for AbiWord 3.0

Several benefits include:

* Many (the majority?) of translators are familiar with and already use gettext
* There are tools (including automated tools) that work with gettext
* The code is maintained by someone other than us

You get none of those benefits with a proprietary format.

2009/11/17 SSK <eskaer_spamsink@ngs.ru>:
> ----- Original Message -----
> From: "J.M. Maurer" <uwog@uwog.net>
> To: <abiword-dev@abisource.com>
> Sent: Tuesday, November 17, 2009 4:03 AM
> Subject: Roadmap for AbiWord 3.0
>
>
>>
>> Hi All,
>>
>> After AbiWord 2.8 comes AbiWord 3.0 (let's skip 2.10, people will be
>> confused), which means we need to decide what to put in it.
>>
>> Fridrich already committed some work towards a Windows 64 bit port, so
>> that will be in 3.0 to start.
>>
>> Then we have some stuff lined up in branches that we need to decide upon
>> when to merge:
>>
>> 1. gsoc2009gettext - port abiword to gettext, making plugins localizable
>> 2. gsoc2009unicode - make the Win32 port a true unicode application
>> 3. gsoc2009tablelayout - improve (table layout) performance, includes
>> the new PieceTable.
>
>
> Why use gettext? In my opinion, it is not an elegant system: it stores translations in binary format, which requires special development tools to (de)compile; also it has some ugly hacks like embedding of non-printable characters in keywords to prevent undesired string merging, and run-time interpreting of expression for plural form(s). The current *.string format is good enough: usable and clear.

-- 
"I like to pay taxes. With them, I buy civilization." --  Oliver Wendell Holmes