Re: Emacs as a translator's tool

[Eric Abrahamsen]

Jean-Christophe Helary <jean.christophe.helary@traduction-libre.org>
writes:

>> On May 30, 2020, at 2:39, Eric Abrahamsen <eric@ericabrahamsen.net> wrote:
>
>> I've thought many times over the years about what I would really want an
>> Emacs-based translation environment to provide for me. I don't do
>> technical translation, so there's not a whole lot of value in
>> sentence-by-sentence correspondences.
>
> Most translation tools I know (or I've used professionally) rely on a
> segmentation scheme set by the user. If the user wants paragraph based
> segmentation, so be it. What people call "sentence" segmentation is
> actually a regex based system that takes into account various signs in
> the source language.

Okay, that's good to know. I guess I would just set it to split by
paragraph, but would also like manual control in some cases.

>> But as Yuri mentions it can be
>> very useful to keep track of how you've translated certain names, or
>> certain important terms, in different places throughout the text.
>> Basically I would want two things:
>>
>> 1. A way to keep track of location correspondences between the source
>>   text and translated text. CAT tool split the text up by sentence,
>
> (not true, see above)
>
>> but
>>   that's not very useful for fiction (particularly Chinese->English
>>   translation) because there's rarely a one-to-one correspondence.
>>   There /is/ a more reliable correspondence between paragraphs, though,
>>   and I'd like to know which paragraph equals which. The point would
>>   mostly be to find my place again when I start translating at the
>>   beginning of the day, and to implement a more useful follow-mode.
>
> I'm not sure I understand what you mean. What's the difficulty that you are 
> facing ?
>
>> I
>>   imagined this would happen when the mode was turned on: it would run
>>   down the file and insert markers that would be used to find
>>   correspondences. Special characters could be inserted into the file
>>   to indicate that two paragraphs should be joined, or one paragraph
>>   split.
>
> What would be the use of such a marking ?

A follow-mode, as I mentioned above. And just finding my place. I do my
translation in two sibling Org sub-trees, original and translation,
displayed in two side-by-side windows. I don't want to mess with
two-column-mode or anything like that. I want to be able to go to the
bottom of the translation, run a command, and have the second window
display the corresponding original. If I realize I've done something
wrong a couple of chapters previously, and I skip back up to that
location in the translation, I want to run the same command to display
the corresponding spot in the original.

>> 2. Link terms in the translation to a glossary pulled from the original.
>>   This would be character names, places, special terms, etc. They might
>>   not always be translated the same way, but I need to know how I've
>>   handled them earlier in the document. Glossary terms would be
>>   highlighted in the source text, and when you came to the equivalent
>>   spot in the translation, you'd use a command like
>>   insert-translation-term that would prompt for the translation,
>>   offering completion on earlier translations, and then insert that
>>   term into the translated text with a link to the original in the
>>   glossary. There would also be two multi-occur commands: one that
>>   prompted for a translation and showed all the places in the source
>>   text where it came from, and another that did the opposite: prompted
>>   for an original glossary term and showed all the places in the
>>   translation where it was translated.
>
> Very nice ideas.

Maybe this will inspire me to write some code! The nice thing about the
glossary is that it wouldn't have to just be vocabulary. You could just
as easily use it for "every time the car crash is referenced", or
something like that. You'd just have to manually mark the passage in the
original, rather than automated marking by text search.