Hi All,
It looks that the "Unicode and the Linux Console" thread is losing direction,
so I am opening a new thread.
The question whether system messages should be localised or not comes up often
in mailing lists, especially localisation project mailing lists.
They main reason why they should not (mentioned in the previous thread) is to help
people in finding online references with a search engine like Google.
I suppose an example of a translation domain with system messages is glibc (Translation Project).
There have been cases that translators have been "strongly advised" not to translate specific translation domains
because of the Google searching issue.. 8-).
A. If people deem that a translation domain should not exist (such as glibc?), you should talk to the people
that did the internationalisation, and get the response from them. A translator's job is simply to translate
all messages that are available.
B. If a developer has the need to see the messages in English (but is ok with rest of messages), simply
remove the offending .mo file from /usr/share/locale/LL/LC_MESSAGES/xxxx.mo
C. It looks interesting to have global error codes in error messages (such as "GLIBC0003: File not found").
It will require however extra work for the maintainance of the list of errors, the changing in the source code
and the translators updating the resulting fuzzy messages. Is there enough manpower to actually do this work?
Simos
Martin Wiss wrote:
Often when I search for the solution of an error message I find the solution in german, spanish, chinese or japanese language. On the homepage of a german, spanish, japanese or chinese LUG.
So the global sharing of knowledge is already lost, has always been lost, and will always be lost. But that deasnÂt mean that this is a big problem, or that there are language barriers always, since people do learn other languages, and communucate with each other in languages other than our mothertongues. (Like we are doing now)
I think displaying an unique errno would be a good behaviour, wheater or not the messages are localized or not, since it would make it easier to locate the answer in any way.
Regards, Martin
P.S. I feel that maybe this discussion is a bit out of topic, that should not be kept for too long... since it is more about what people think generally about languages, than about practical issues regarding implementing Unicode in Linux. So I apologize that I couldnÂt refrain from sending this message. D.S.
By the way, people generally do not like the idea of a universal language if it is not their own mother tongue.
-- No virus found in this outgoing message. Checked by AVG Anti-Virus. Version: 7.0.300 / Virus Database: 265.6.13 - Release Date: 16/01/2005
-- Linux-UTF8: i18n of Linux on all levels Archive: http://mail.nl.linux.org/linux-utf8/
