Translation in Practice
Translation Process
A - SL writing
B - Translation
- Read SL Text
- Read to understand
- Identify unknown words/terms
- Read to understand
- Research terminology
- Online
- Reference works
- Decide on competing usage (Google as large collection of corpora)
- Online
- Write translation
- Question yourself
- Am I sure this is the best translation?
- Does this sound right and familiar only because I always translated it like that?
- Different techniques
- Use CAT tools
- Write directly in word processor
- Overwrite original
- Write first draft with optional wording, to be self-edited later
- Example:
- Write first/rough draft using optional/preliminary/candidate terms/words...
- Write first/rough draft using optional/preliminary/candidate terms/words...
- Am I sure this is the best translation?
- Self-edit translation
- Read your translation
- Does it read like a similar document written by a native speaker of the TL would read?
If it does not make sense, there is a problem (and "but the original did not make sense, either" or "but I used the same words as the original" are not good excuses)
- If possible, set it aside after the first draft, to look at it with fresh eyes
- Send translation to editor
- Does it read like a similar document written by a native speaker of the TL would read?
C - Quality Control steps
- Editing
- Review translation against SL text
- Errors of meaning
- Errors of form
- Errors of compliance
- Preferential changes
- Track changes
- Change translation directly
- Errors of meaning
- Proofing
- Review only the TL text (checking the SL
only if necessary)
- Review only the TL text (checking the SL
- Other quality control steps (These take place all during the process, from terminology work onwards)
- Verification that translation is complete (all files translated)
- Other QC steps
- Verification that translation is complete (all files translated)
D - QA
- Techniques used in order to assess the effectiveness of QC work
- Translation quality measurement
Technology, tools, reference materials and publications
Hardware
Computer: PC or Mac?
- PC environment much more widespread
- Greater availability of CAT tools
- Easier to meet customers' requirements
- Cheaper computers
- Greater availability of CAT tools
- Mac preferable when working with graphic designers
Desktop or laptop?
- Laptop permits to work anywhere
- Desktops
- More powerful
- Cheaper
- More robust
- More powerful
Peripherals
- Scanner (Necessary if much work is received as hardcopy)
- Printer (Desktop lasers are the most economical choice)
- External hard disk (For removable back-ups)
- UPS (Uninterruptible power supply)
Fast Internet connection
- Essential when working with large files
- Useful when surfing the web for research purposes
Software
- Normal software tools (Office software)
- MS Office
- Specialized search techniques using wildcards (Regular Expressions)
- Using Excel to create "ad hoc" translation tools for the translation of software strings (especially when there are size limitations and such)
- Using Excel to create translation memories from text glossaries
- Specialized search techniques using wildcards (Regular Expressions)
- Open Office
- Star Office
- MS Office
- CAT (Computer-Aided Translation) tools
- SDL
- Trados
- Currently the market leader
- Buggy
- No free "lite" version
- SDLX
- "Lite" version available (Works only with projects created by customer)
- Currently the market leader
- Déjà Vu
- Omega T
- Freeware
- Freeware
- Wordfast
- Mostly compatible with Trados
- Cheaper than most other CAT tools
- Works with the Mac
- Mostly compatible with Trados
- Transit
- Other translation memory tools
- SDL
- Terminology management tools
- Additional software tools
- Localization tools
- Catalyst
- Multilizer
- Other localization tools
- Catalyst
- Specialized search tools
- Adobe Acrobat
- Professional (Permits editing, annotating, etc.)
- Reader (Freeware)
- Professional (Permits editing, annotating, etc.)
- Text editors
- Windows' Note Pad is free
- TextPad
- Ultra Edit
- Windows' Note Pad is free
- DTP programs
- Framemaker (Text-intensive documents)
- Quark
- InDesign
- Framemaker (Text-intensive documents)
- HTML editors
- Front Page
- DreamWeaver
- Etc.
- Front Page
- Graphic editing tools
- Photoshop
- Photoshop Elements (Cheaper alternative, permits editing text in Photoshop files)
- Illustrator
- Photoshop
- OCR (Optical Character Recognition) tools, such as Omnipage (Useful when one receives documents to translate in hardcopy or pdf, and needs to transform them into editable files)
- Word counting software
- Translation project management tools (e.g., Translation Office 3000
- Clipboard utilities (e.g., ClipMate)
- Screen capture tools (e.g., Gadwin PrintScreen
- Localization tools
- Google and other online search aids
- MT (Machine Translation)
Reference materials
- Jost Zetzche's Translator's Tool Box
- Dictionaries
- Printed dictionaries
- Dictionaries on CD-ROM
- Differences v. paper dictionaries
- More powerful searches
- WordWeb Pro
- Differences v. paper dictionaries
- Online dictionaries
- Printed dictionaries
- Glossaries
- Do-it-yourself (See above, "Software for terminology management")
- Online glossaries (WARNING: separate the wheat from the chaff!)
- WikiWords
- Do-it-yourself (See above, "Software for terminology management")
- Other reference materials
Assignments for next class
Assignment: read chapters 6 and 7 in "Becoming a Translator", and Chapter 4 of "How To Succeed As A Freelance Translator"Notes from the previous lessons in this course:
Foundations of Translation - Course Description
Foundations of Translation - Lesson 1: Difference between translation and interpreting
Foundations of Translation - Lesson 2: Jobs for translators
Foundations of Translation - Lesson 3: Characteristics of a good translator
Nice post, thanks!
ReplyDeleteHowever, I find it interesting that you list Babelfish and Systran as two different options for Machine Translation, yet they both use the same technology.
How about including links to SDL's FreeTranslation.com website, which uses SDL's Enterprise Translation Server as its machine translation engine?
[Disclaimer: I work for SDL...]
Hi Tony:
ReplyDeleteBear in mind that these are just my notes for the course, so what's missing is what I told the students about the various things. Babelfish was included, in fact,
a) to provide a free example of what MT looks like (and specifically, what Systran looks like since, as you say, it uses the Systran engine),
b) to provide an example of what not to do (i.e., "do not send out tests translated with Babelfish: I've received a few, they look terrible and you will be found out")
I mentioned machine translation in the course mostly to warn people away from it.
Hi Ricardo,
ReplyDeleteThanks for this usefulpost.
I have just inherited a huge translation contract that could be back-breaking andf will investigate some of the tools suggested here.
I would like to add that I will be using voice recogntion (Dragon Naturally Speaking)French prefered edition - apparently this can be used for English as well but the opposite is not true.
The last edition is especially performant.
How does one use Excell for terminology retrival?
I would not used Babelfish but was recommended Reverso Prof. Edition.
Is this still valid info?
Thanks
Diane
English-French French-English translator in Québec.
Oh my Goddess,
ReplyDeleteTypos!
As regards Reverso, I don't know the program, but I'd generally be wary about any MT program: with some exceptions, in order for them to be useful a lot of preliminary terminological work, pre-editingand post-editing is necessary.
ReplyDeleteAs regards Excel: it can be useful to translate software strings, since with a few formulas you can set it up to check for things such as string length, etc. Also, if you have a glossary you were maintaining in Excel, it is fairly easy (though a bit more complicated than can be explained in a blog comment) to use Excel to transform the glossary into a MultiTerm database or even a translation memory... I'll probably explain the steps involved in a future post.
Thanks Ricardo
ReplyDeleteCan I create the terminology list in Word or should I try to make it Excell right away?
I am getting Dragon Naturally Speaking - French edition. I was told that it also recognises English but I am not sure of that. I will throw Antidote, a very powerful French correction software and hope I can switch the text around without crashing anything.
What should I get next as terminology recognition software or any new step?
Diane
Hi Diane:
ReplyDeleteI would do it directly in Excel: Word is really not suitable for something such as keeping a glossary.
To do it in Excel in a way that may later be easily exported to other applications (like MultiTerm), I would use the following format:
Start from row 1 column one (i.e., from cell A1).
I the first column, cell A1 put your source language, in A2 your target language, in A3 a definition, and, if you want, further fields in the next columns (e.g., context, comments, and so on); then fill the terms as necessary starting from cell B1.
As regards the voice recognition programs, I have no experience at all with them: I'm not very good at dictating, and even if I were, there is two of us working in the same room, and dictating our translations would be absolute chaos.
I know translation offices who use them and find them very useful. I would too because I am a lousy typist but a fairly quick translater.I work on my own.
ReplyDeleteConcerning the voice recognition, I was told that you must edit random noise that show up as random words in the text.
Someone else also told me that this modus oprandi had to do with existing individualistic cerebral pathways that are hard to retrain.
Diane