Personal methods in translation

Dan Lucas wrote:
Like Chris, I am very specialised and experienced, but my Japanese clients insist that the translation follows the precedent of previous translations, if such translations exist, so there's a clear cultural difference there. I have seen edited copies of my translations, and rather than errors, 99.9% of the changes made consist of the editor rearranging sentences to make them look more like previous translations. Rather sad, in a way.

When the client offers me a job I will scan the whole document, then go into it more carefully, and see what resources I can find for areas with which I am unfamiliar. If I think I can take it on, I get back to them with a price and a deadline.

Becca Resnik wrote:



What I find delightful about this is that you seem to translate the way one would write a program, which you obviously have extensive experience in (I took a peek at your profile after reading your post and noting "software strings" in your list above ). Understand the overall goal; work line by line, checking syntax carefully; compile; fix two out of 539 "errors" to yield the perfect result.

Jennifer Caisley wrote:

Like many posters above, I have a multi-stage approach. It seems fairly similar to what Chris does above (albeit without dictation) - so I'm glad I'm not the only one!

1. Do a very quick first draft of the text in my CAT tool, essentially at the speed it takes to read the ST. I don't stop to look up terminology or puzzle out tricky structures: this stage merely serves as an opportunity to familiarise myself with the text and make a mental note of areas that might need extra attention later on.

2. Go back through the text, segment by segment, and amend the quick draft so that it accurately matches the ST. Depending on the type of text and how accurate the first draft was, this could be a very quick process or like wading through treacle. This is where I look up terminology that's unfamiliar, cross-check against my termbase etc.

3. Preferably the next day, or at the very least with fresh eyes, re-read the text through in my CAT tool, doing a segment-by-segment check for accuracy and any easy slips like the spelling of people's names, mis-typed numbers etc. I then run the QA process as a final double-check.

4. I export the text from my CAT to its native format (Word, Excel etc.) and review the text for flow. It's amazing how differently the text reads when it's in its intended layout!

5. If there's time, I print out the translation and do a final review on paper.

Becca Resnik wrote:
Or is it different because you have long-term clients, because you sign NDAs

Susan van den Ende wrote:

Wonderful question and some interesting approaches, thanks for sharing!

For manuals, I like to create a view and then sort it by string length, so I can translate the shorter ones first. This way MemoQ suggests them for auto-complete when they get referenced in other segments, which speeds up my work and helps consistency.

Example:

Somewhat shorter segment:

* Chapter twelve: How to do this complicated thing with the machine.

Sentence in chapter 1:

* If you want to do the long and complicated thing with the machine, refer to [Chapter twelve: How to do this complicated thing with the machine.]

In this case, I would also filter the view for any segments that start with "How to", so I can make sure that all translations of chapter titles use the same grammatical structure.

You don't catch everything like this, obviously, and it doesn't work for every manual. But for manuals with this type of structure, this method tends to work *really* well. It's not just the time you save with auto-complete, but also the QA time you need to verify that your references are consistent.

Yolande Hivart wrote:
...
[If I have to use a CAT tool, I check for constitency and the wording that was already in the TM, I make a quality check and proofreading]
...
Extra if I do not have to use a CAT tool:
...

Becca Resnik wrote:

Thank you very much for detailing all of this out! If I'm inferring correctly that you don't prefer to use CAT tools, why don't you, and what determines if you "have to"? Client requirement, current workload, or something else?

Nadja Balogh wrote:

I usually don't bother much about going through the document first (except to find out if it is something I can handle or not). Somehow my brain will only start working properly when I'm in translation mode - any useful questions I might have will only start forming themselves when I'm actually translating, not when simply reading through the text.
This often means that the stuff I translate at the beginning will need to be revised again after a while as everything starts to become clearer, which is why I try everything I can to avoid projects with partial delivery halfway through.

Nadja Balogh wrote:

I usually don't bother much about going through the document first (except to find out if it is something I can handle or not). Somehow my brain will only start working properly when I'm in translation mode - any useful questions I might have will only start forming themselves when I'm actually translating, not when simply reading through the text.
This often means that the stuff I translate at the beginning will need to be revised again after a while as everything starts to become clearer, which is why I try everything I can to avoid projects with partial delivery halfway through.

Mark Harris wrote:
I also work in a similar way. I'm fairly early into my career so I generally don't get paid the greatest rates, so time is often a crucial factor to me to maintain profitability. I obviously try and get a good sense of who the end client is, what the text is generally about, etc. before I start translating, otherwise I would have to change everything afterwards, but I don't have time to read through the entire ST before I even start translating, particularly if it's a long text. I tend to just get stuck into the translation and then change some of the wording when I go back through it. It does help that I work a lot with the same end clients, so I can tell pretty quickly what the text is going to be about since I'm very familiar with who they are and what they do.

One thing that I have changed, similarly to others here, is skipping difficult segments and going back to them later. I used to spend ages agonising over what they meant as I came across them, only for the answer to become clear later as I carried on working through the text, once I had more context. So rather than wasting time and energy battling with an incomprehensible phrase, I leave it for later to see if the rest of the text will yield an answer.

I do hope to start earning better rates as I progress through my career so I can relax and take my time a bit more though. I do work hard to guarantee good quality of course, but I also have to work fast to ensure I can make a living!

Becca Resnik wrote:

I am new as well, so I truly am grateful for your input. Thanks for taking the time to respond! Isn't that so frustrating when you dwell on a term only to see that the later context would have instantly clarified it regardless?

Have you taken any measures toward increasing rates, or have you had to hold them steady to keep clients? I know this is a hotly-argued topic, so I'm just curious about your personal experiences/plans, if you don't mind sharing.