Subject: Re: Translated manual pages
To: None <current-users@netbsd.org>
From: Paulo Alexandre Pinto Pires <p@ppires.org>
List: current-users
Date: 10/29/2002 14:28:40
On Tue, Oct 29, 2002 at 03:05:27PM +0900, Bang Jun-Young wrote:
> In Korea, a lot of technical books are poorly translated by people who
> don't have a good knowledge on the subject, and even Korean grammar.
> I regretted that I bought such books quite a number of times, so these days
> I no longer read or buy translated books. I also have heard a lot of people
> complained of poor quality of translated books.

Please  add  me  to your list (when translation is to Portuguese,
though).

> The same thing could happen
> with translated manpages, if they were not thoroughly proofread by
> developers (and other translators in the team).

There's always user feedback, too.  We already have a category in
send-pr for problems in the documentation.

Translating  can sometimes be a very difficult task.  Frequently,
translated text end up looking clearly like translated  text;  an
ordinary  speaker of the target language would never have written
it that way.  Proofreading must go further than looking  for  in-
correct  words and information, in should also look for bad style
and fix it.  It is too big a task for a team of voluntary  trans-
lators, who would work mostly in their spare time.  User feedback
would be very important, not to overwhelm translators (who  would
live perfectly with just English docs).

Translation  teams would vary considerably in size and structure.
I believe that translation to Japanese would turn to being faster
than  would  translation to Portuguese -- yet I hope I am not the
only Brazilian running NetBSD and not the only  willing  to  con-
tribute with translation efforts.  If translations are to be con-
sidered part of the system, it is important that we define how we
will  deal  with different pace and stages for different transla-
tions of the very same documentation.  Will there be any  set  of
minimum  requirements  (quantitative  and/or  qualitative) before
start shipping any given  translation?   What  will  be  fallback
policies (which require changing software)?

We also need to look at what other projects are doing or have al-
ready achieved is this area.  While this could help  us,  we  can
also  help  them  --  I believe, for example, that *a lot* of man
pages could be shared between all the BSDs.  Perhaps we can  also
learn from e.g.  SuSe and Conectiva, which have long engaged into
translating Linux, and maybe even pick some  material  in  common
areas.

> Jun-Young
> 
> -- 
> Bang Jun-Young <junyoung@mogua.com>

-- 
	Pappires

... Qui habet aurem audiat quid Spiritus dicat ecclesiis.