Re: [PATCH 1/3] docs/zh_TW: add translations for zh_TW/arm64

[Akira Yokosawa]

On Tue, 24 Aug 2021 13:40:40 -0600, Jonathan Corbet wrote:
> Hu Haowen <src.res@xxxxxxxx> writes:
> 
>> Create new translations for zh_TW/arm64 and link them to index.
>>
>> Signed-off-by: Hu Haowen <src.res@xxxxxxxx>
>> ---
>>  .../translations/zh_TW/arm64/amu.rst          | 104 ++++++++
>>  .../translations/zh_TW/arm64/booting.txt      | 251 ++++++++++++++++++
>>  .../translations/zh_TW/arm64/elf_hwcaps.rst   | 244 +++++++++++++++++
>>  .../translations/zh_TW/arm64/hugetlbpage.rst  |  49 ++++
>>  .../translations/zh_TW/arm64/index.rst        |  23 ++
>>  .../zh_TW/arm64/legacy_instructions.txt       |  77 ++++++
>>  .../translations/zh_TW/arm64/memory.txt       | 119 +++++++++
>>  .../translations/zh_TW/arm64/perf.rst         |  88 ++++++
>>  .../zh_TW/arm64/silicon-errata.txt            |  79 ++++++
>>  .../zh_TW/arm64/tagged-pointers.txt           |  57 ++++
>>  Documentation/translations/zh_TW/index.rst    |   5 +
>>  11 files changed, 1096 insertions(+)
>>  create mode 100644 Documentation/translations/zh_TW/arm64/amu.rst
>>  create mode 100644 Documentation/translations/zh_TW/arm64/booting.txt
>>  create mode 100644 Documentation/translations/zh_TW/arm64/elf_hwcaps.rst
>>  create mode 100644 Documentation/translations/zh_TW/arm64/hugetlbpage.rst
>>  create mode 100644 Documentation/translations/zh_TW/arm64/index.rst
>>  create mode 100644 Documentation/translations/zh_TW/arm64/legacy_instructions.txt
>>  create mode 100644 Documentation/translations/zh_TW/arm64/memory.txt
>>  create mode 100644 Documentation/translations/zh_TW/arm64/perf.rst
>>  create mode 100644 Documentation/translations/zh_TW/arm64/silicon-errata.txt
>>  create mode 100644 Documentation/translations/zh_TW/arm64/tagged-pointers.txt

Hi Jon,

> 
> I have gone ahead and applied this set.  I really do have to ask,
> though: is there really nobody with sufficient interest to review these
> patches?  The zh_CN translators have done a good job of organizing
> themselves to review this work; could you please try to do the same?
> Otherwise I'm being asked to add lots of material that, as far as I can
> tell, nobody has looked at...

I expected there would be some response from Haowen to your concern.
I'm an outsider of the zh_TW translation effort (except for the font
setting for pdfdocs) and does not read Chinese, but let me describe
my observation of what Haowen is up to in the hope of making you feel
better.
(I might be missing something important, though.)

The zh_TW work is not a direct translation from English documents,
rather, it is based on the zh_CN translations.

This is not explicitly mentioned in the change logs of Haowan's patch sets,
but if you compare the "Translator" fields under zh_TW to that of under
zh_CN, the chain of translation/conversion can be seen easily.

For example, the head of Documentation/translations/zh_CN/arm64/amu.rst
reads:

------
    .. include:: ../disclaimer-zh_CN.rst
    
    :Original: :ref:`Documentation/arm64/amu.rst <amu_index>`
    
    Translator: Bailu Lin <bailu.lin@xxxxxxxx>
    
    ==================================
    AArch64 Linux 中扩展的活动监控单元
    ==================================
------

, while the head of Documentation/translations/zh_TW/arm64/amu.rst
reads:

------
    .. SPDX-License-Identifier: GPL-2.0
    
    .. include:: ../disclaimer-zh_TW.rst
    
    :Original: :ref:`Documentation/arm64/amu.rst <amu_index>`
    
    Translator: Bailu Lin <bailu.lin@xxxxxxxx>
                Hu Haowen <src.res@xxxxxxxx>
    
    ==================================
    AArch64 Linux 中擴展的活動監控單元
    ==================================
------

My interpretation of above is that the zh_TW work is based on the
zh_CN translation by Bailu Lin, which has got proper reviews by
the zh_CN community.

IIUC, changes in zh_CN -> zh_TW are mostly "Simplified Chinese" ->
"Traditional Chinese" character substitutions.
They can be done semi-automatically based on a character mapping.
Exception is that some of the character mapping are one (SC) to
many (TC) [1].
And simply substituting characters might end up in unnatural wording
in the zh_TW documents due to vocabulary/convention differences.
This is where human intervention is required occasionally.

So, Huawen's patches can be regarded nearly as good as that of
the zh_CN counterparts even if they don't have anyone's
Reviewed-by tags, I suppose.

Of course, it would be much better to have regular reviewers. 

Haowen, if I'm saying something far from reality, please let
me know.

[1]: https://en.wikipedia.org/wiki/Ambiguities_in_Chinese_character_simplification

        Thanks, Akira
 
> 
> Thanks,
> 
> jon