On Sun, Feb 2, 2025 at 8:27 PM Akira Yokosawa <akiyks@xxxxxxxxx> wrote:
>
> Tsugikazu Shibata wrote:
> >> Do you read the list that way?
> >>
> >> I don't think so. It's a mixture of suggestions made in various grammatical
> >> structures.
> >>
> >> For example, here is an item in original submit-checklist.rst:
> >>
> >> 3) All memory barriers {e.g., ``barrier()``, ``rmb()``, ``wmb()``} need a
> >> comment in the source code that explains the logic of what they are doing
> >> and why.
> >>
> >> I don't think this is in the form of TODO. Rather, it implies what you
> >> should do, in a sentence whose subject is "All memory barriers".
> >> So I don't think it is worth preserving wording in the original.
> >> List of "What you are supposed to have done before submission"
> >> is much straight forward, isn't it?
> >
> > In your translation, most of the sentence is saying "whether you have done"
> > or "Have you done something" in Japanese. Those are asking each item
> > to the readers as a result.
> > However, the original text is not asking the readers. That is what I
> > pointed out.
> > I feel we, open source developers, are historically working for their
> > own motivations.
> > So, this checklist would be written for the developers for their self
> > checking purpose.
> > That is the reason, the list is just saying "Do something" and it
> > became the TODO list..
> > It is not "Have you done something" because that sounds like "Another
> > person is asking you"
> > and that's not preferable for the developers in my opinion.
> > I really hope that kind of non-written understanding would be included
> > even in translated documents.
>
> Thank you for elaborating.
>
> So, you read authoritative tone/nuance in my translation.
>
> Let me see ...
>
> You suggested a change from:
>
> +1) 利用している機能について、その機能が定義・宣言されているファイルを
> + ``#include`` したか。
>
> to
>
> +1) 利用している機能について、その機能が定義・宣言されているファイルを
> + ``#include`` すること。
>
> Actually, I don't see much difference in the nuance/tone between them.
First one is a stronger question than the second one, I feel.
So, the long list of stronger questions is becoming a pressure to the
developers.
> 「〜すること」 is sometimes used to indicate authoritative ordering.
>
> For example, "You should test both of cases A and B." can be translated into
>
> A と B の両方をテストすること。
>
> , and this can be translated back into:
>
> I ask you to test both of cases A and B.
>
> , depending on contexts.
>
> Instead, if you just say without 「こと」:
>
> A と B の両方をテストする。
>
> in a list of what to do, chances of such an interpretation can be
> reduced.
>
> Do you prefer this approach?
Yeah, This looks better than my opinion.
> Following diff (on top of patch 1/2) is my attempt to address your
> concern for the first three items in the checklist:
>
> Here, I changed each sentence to have "patch or change in the patch" as
> its implicit subject or "主語", and describe its preferred state.
>
> コードのレビュー
> ================
>
> -1) 利用している機能について、その機能が定義・宣言されているファイルを
> - ``#include`` したか。
> - 他のヘッダーファイル経由での取り込みに依存しないこと。
> +1) 利用する機能について、その機能を定義・宣言しているファイルを
> + ``#include`` している。
> + 他のヘッダーファイル経由での取り込みに依存しない。
>
> 2) Documentation/process/coding-style.rst に詳述されている一般的なスタイル
> - についてチェックしたか。
> + についてチェック済み。
>
> -3) すべてのメモリバリアー (例, ``barrier()``, ``rmb()``, ``wmb()``) に、
> - その作用と目的、及び必要理由についてソースコード内にコメントしたか。
> +3) メモリバリアー (例, ``barrier()``, ``rmb()``, ``wmb()``) について、
> + その作用と目的、及び必要理由について説明するコメントが、ソースコード内
> + にすべて付いている。
>
> Do they sound better/acceptable to you?
This looks good to me!
> [...]
> > Fault injection on Linux was developed and contributed by Akinobu
> > Mita, a Japanese guy.
> > He spoke about it at the Japan Linux Symposium in 2007.
> > In that time, we have discussed how "fault injection" can be
> > called/translated into Japanese language.
> > As a result, he chose just "Fault Injection" in English, not using
> > Japanese Kanji or Katakana.
> > You can see his presentation at
> > https://www.static.linuxfound.org/jp_uploads/seminar20070710/LinuxFaultInjection-2.pdf
>
> Good to know I'm not alone in disliking transliteration of the term!
>
> > With respect to him, I would recommend using English "Fault Injection"
> > instead of Japanese words. I hope this may be a good solution.
> > (I found the name of Jonathan Corbet at the symposium and saw Randy at
> > the previous event :-)
> >
>
> So how about the following?
>
> 4) 最低限、slab と ページ・アロケーションの失敗に関する誤り注入
> (訳註: fault injection) によるチェック済み。
>
> This still diverges slightly from its English counterpart of:
>
> 4) Has been checked with injection of at least slab and page-allocation
> failures.
Again, Looks good now.
> , though ...
>
> Translation is hard!!
Thanks,
Tsugikazu Shibata
