[lkml]   [2003]   [Apr]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRE: kernel support for non-English user messages
Hi Robert.

> Actually, my final point had been that doing it inside the kernel
> itself, or indeed inside klogd, was probably a very bad idea. If
> the translation always happens after-the-fact based on properly
> harvested message semantics then any segment of messages
> distributed into this mailing list (among other uses) would be
> A) Still in English.
> B) Translatable after the fact there too.
> Also after-the-fact translation makes the language translations a
> scalar problem instead of a matrix one. That is, if you always
> pass the message stream around in English (treat it like n opaque
> source file) and then translate it as necessary, it will "always
> work".
> If you try to do the translations at message generation time, then
> the translation must be any-language-to-any-language capable during
> post-even discussions. Not good.

I can see the points you're making, and that is precisely why I believe
that message codes would be required to implement this idea. As Linus
has vetoed the idea of having message codes in the kernel, I can't see
it ever coming to fruition.

> Also, you will always have leakage as people add new strings to the
> set.

That's the easiest aspect of dealing with it - the tool that generates
the language set to use just grabs the "English" language version for
any message codes not in the selected translation.

> As for the #define issues, when you process the source tree to build the
> source matrix you just "gcc -E file.c | collector" and now the printk
> case you mention is handled. Any module designer who does uglier things
> can make a dead-code procedure that expresses his possible output strings
> for collection (if he cares.)

> {Satire}

> Speaking as an arrogant (U.S. of) American who knows that God(TradeMark,
> all rights reserved) decreed that he never had to learn any language but
> his own, I can honestly state, that it is nearly certain that you will
> get no real support for the multi-language kernel out of a us
> We can't even get ourselves to write decent comments, and on the average,
> we all secretly believe that if we just speak slowly enough everybody
> really knows English. After all, that's how our condescending "wouldn't
> want to fail Johnny, it would be bad for his self-image" public schools
> taught us in the first place.... 8-)

Speaking as an amused (U.S. of) American, I long ago learned how to tell
somebody is speaking "God's Language"(tm) - that's simple to work out. After
all, the most likely people to speak "God's Language"(tm) are those that
just left His presence - the new born babies - so if we want to listen to
language, we just listen to them speak it. What could be simpler???

However, I understand "God's Language"(tm) is not currently understood well
enough by the kernel developers for any of them to translate the kernel
messages into it...

> {/Satire}

Best wishes from Riley.
* Nothing as pretty as a smile, nothing as ugly as a frown.

Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (
Version: 6.0.471 / Virus Database: 269 - Release Date: 10-Apr-2003

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
More majordomo info at
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:34    [W:0.128 / U:1.752 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site