Questions for Déjà Vu X2 experts Thread poster: septima
|
|---|
Hi, I'm experimenting with the new Déjà Vu X2 trial. Could any users kindly answer a couple of questions?
1) Abbreviations: Is there an intelligent way to import abbreviations into the segmentation exception rules? This is obviously very important right at the beginning, but the only way seems to be by hand.
2) Footnotes: Footnotes in my documents break the text up and sit in the middle of the broken parts. I've researched this on the Atril forums, and someone suggest... See more Hi, I'm experimenting with the new Déjà Vu X2 trial. Could any users kindly answer a couple of questions?
1) Abbreviations: Is there an intelligent way to import abbreviations into the segmentation exception rules? This is obviously very important right at the beginning, but the only way seems to be by hand.
2) Footnotes: Footnotes in my documents break the text up and sit in the middle of the broken parts. I've researched this on the Atril forums, and someone suggests (for DOC files) importing them "as DOCX", but I can't see any way to do this. Is there a good solution for footnotes?
3) Autosuggest portions/segments: Is there a way to set up an alternative keyboard shortcut for choosing a portion or segment? On a laptop keyboard it is quite inconvenient to be going down the list and then pressing e.g. Ctrl + 51 or Ctrl + 37 all the time just to insert a little text. An arrow key function would be a better option for me.
Thank you very much for any ideas! ▲ Collapse
| | | |
septima wrote:
1) Abbreviations: Is there an intelligent way to import abbreviations into the segmentation exception rules? This is obviously very important right at the beginning, but the only way seems to be by hand.
Indeed, it's a pain in the arse.
You can edit directly the settings.dvset file in Access.
Exception column in the Languages table, search for your MS language code (e.g. 1045 for Polish, 1027 for Catalan etc.).
Not very user friendly but it's faster, you can also copy-paste data created in an external tool if you get familiar with the codes.
Make a backup before you proceed...
2) Footnotes: Footnotes in my documents break the text up and sit in the middle of the broken parts. I've researched this on the Atril forums, and someone suggests (for DOC files) importing them "as DOCX", but I can't see any way to do this.
Simply save as docx, then import.
Or select the file in the Project Explorer, select the Properties tab and force the filter you need.
Is there a good solution for footnotes?
You can also process the file in Trados and import the TTX or bilingual doc.
Generally, 90% of the jobs I receive are Trados ones, so...
3) Autosuggest portions/segments: Is there a way to set up an alternative keyboard shortcut for choosing a portion or segment? On a laptop keyboard it is quite inconvenient to be going down the list and then pressing e.g. Ctrl + 51 or Ctrl + 37 all the time just to insert a little text. An arrow key function would be a better option for me.
In fact, I always considered the Alt+number is the fastest way if one uses the NumPad.
Nonetheless, the arrow method is possible by default.
Press Ctrl+Shift+Down/UpArrow to select, then press Ctrl+R (insert mode) or Ctrl+Shift+R (overwrite mode).
You can also simply double-click the item you want.
BTW, if you have a keyboard without numpad, you can also remap the numkeys using AutoHotkey.
I used this method before I got used to do it "as is".
For some notebook brands, it's possible to emulate the numpad using Fn.
E.g. it worked on Dell, Fujitsu or older HP machines but IBM/Lenovo is a pain in the neck.
If I really need to work on a notebook, I use an external keyboard (a Mac one, it's the smallest and the lightest one, it works also as a USB hub).
Catspeed
GG
[Edited at 2011-11-14 13:21 GMT]
| | | | Selcuk Akyuz 
Türkiye
Local time: 20:38
English to Turkish
+ ...
| Ctrl+RightArrow customizable | Nov 14, 2011 |
Grzegorz Gryc wrote: septima wrote:
3) Autosuggest portions/segments: Is there a way to set up an alternative keyboard shortcut for choosing a portion or segment? On a laptop keyboard it is quite inconvenient to be going down the list and then pressing e.g. Ctrl + 51 or Ctrl + 37 all the time just to insert a little text. An arrow key function would be a better option for me. In fact, I always considered the Alt+number is the fastest way if one uses the NumPad. Nonetheless, the arrow method is possible by default. Press Ctrl+Shift+Down/UpArrow to select, then press Ctrl+R (insert mode) or Ctrl+Shift+R (overwrite mode). You can also simply double-click the item you want.
Thanks Grzegorz for reminding the Ctrl+R shortcut. I have also added the Ctrl+RightArrow shortcut key through Tools>Customize>Keyboard>Insert Category
| | | | septima
Local time: 19:38
TOPIC STARTER
Thank you very much, Grzegorz , looks like a found a real Déjà Vu expert 
I'll try and take all that information in, bit by bit, before the trial runs out.
I do think it would be in their own best interests to make this all more user-friendly for those experimenting with the tool for the first time. Obviously, an expert like yourself works things out over time, but new users would like to be able to get... See more Thank you very much, Grzegorz , looks like a found a real Déjà Vu expert
I'll try and take all that information in, bit by bit, before the trial runs out.
I do think it would be in their own best interests to make this all more user-friendly for those experimenting with the tool for the first time. Obviously, an expert like yourself works things out over time, but new users would like to be able to get a feel for what the program offers without having to go to a lot of trouble beforehand (offputting).
How hard would it really be to have a function where you can import your chosen abbreviations from a text file, for example? Not that hard.
Having said that, it does look like an interesting tool, with some good bits of design.
Thanks again, I'll know where to come with any more questions! ▲ Collapse
| | |
|
|
|
Selcuk Akyuz
Türkiye
Local time: 20:38
English to Turkish
+ ...
| Yahoo Deja Vu Group | Nov 15, 2011 |
septima wrote: Thanks again, I'll know where to come with any more questions!
You can find detailed answers also at http://tech.groups.yahoo.com/group/dejavu-l/
A great DVX group with more than 100K messages and 2300 members.
| | | | | Shortcuts and settings... | Nov 15, 2011 |
septima wrote:
I'll try and take all that information in, bit by bit, before the trial runs out.
You can always ask for some additional days...
I do think it would be in their own best interests to make this all more user-friendly for those experimenting with the tool for the first time.
Indeed, some functions need a big rework...
Nonetheless, DVX is one of the easiest tools to understand if you manage few shortcuts, i.e.:
- Cltr+DownArrow, Alt+DownArrow (segment confirmation);
- F11, Shift+F11 (adding terms with/without form), crucial for (Auto)Assemble which is one of the major DVX advantages;
- Ctrl+number (AutoSearch portion insertion);
- F10 (adding text to Project Lexicon), important for AutoAssemble (PL has always the highest priority);
- Ctrl+A (AutoAssemble), may be slow for huge DBs;
- Ctrl+Shift+B/N (shifting the selected text backward/forward), very useful for editing (mainly after AA);
- Ctrl+D or F8 (copy code), Ctrl+D is faster;
- Ctrl+Shift (clear codes);
- Ctrl+Shift+F8 (checking embedded codes), crucial (if they're messed up, DVX may be unable to export the translation or export an invalid file);
- F7 (obviously, spellcheck);
- take a look on the options available in the contextual menu in the translation grid (if a blue portion is inserted in the text area, put the cursor in it and hit the Windows context menu key on the keyboard);
- use the standard Windows Ctrl/Shift+arrows shortcuts
In fact, it's all for a beginner.
In the Options, you should:
General tab
- lower the threshold (minimum score) to, let's say, 55-60%;
- Enable fuzzy terminology matches;
- be careful with DeepMiner, it slows down DVX and the results are often crazy for heavily inflected languages (just test it with different settings);
- Allow portions containing ponctuation (very useful for languages with complex ponctuation);
- disable all the terminology checks except numerals (if you work with heavily inflected languages, these checks throw false alarms for every segment).
Environment tab
- enable Edit in separate text area and Split text area vertically (IMO, it helps a lot but the keyboard comportment may be somehow unstable);
- enable all the automation stuff (On CTrll+DownArrow) except AutoAsemble (see below);
- in the AutoSearch section, enable all (maybe except both "Hide..." options, as you like);
- set the cache expiry to the max. value (2000), it speeds up the searches.
Remember, you can open the same project in multiple DVX instances e.g.:
- in the first one, you start the pretranslation (Ctrl+P, check Assemble from portions and Insert source text, for DeepMiner see above),
- in the second one, you translate interactively (so why I disable AutoAssemble on Ctrl+DownArrow, it's already done in the second instance).
Of course, most of these choices are personal but it may be useful for you.
Obviously, an expert like yourself works things out over time, but new users would like to be able to get a feel for what the program offers without having to go to a lot of trouble beforehand (offputting).
Yep.
But first at all, you should import some termbases.
DVX without termbases loses its edge.
BTW. you have an introductory webinar tomorrow (November,16), see http://www.atril.com/
You can also take a look on http://www.youtube.com/user/AtrilVideo
How hard would it really be to have a function where you can import your chosen abbreviations from a text file, for example? Not that hard.
I can but agree...
BTW, it's a common error in the design in many CAT tools, e.g. Trados Studio doesn't allow to import 'em at all, in memoQ, the configuration files can't be edited in the batch mode in the GUI (at least they can be edited directly in an XML editor) etc.
Catspeed
GG
[Edited at 2011-11-15 12:11 GMT]
| | | | septima
Local time: 19:38
TOPIC STARTER
Thanks so much, Grzegorz (and Selcuk too for the tips)! You've really made a mini reference manual for all DVX newbies here. They should be paying you!
Real work has rudely interrupted me with my playing around with this stuff, but I can see there's plenty of fun to look forward to
Cheers!
septima
| | | | | Import of WF Classic TM | Jan 7, 2012 |
Instead of starting a new thread, I thought to just continue this one, as the title seems appropriate.
I have spent hours trying to import my TM in .txt as well as in .tmx (converted by Wf). With .tmx, DVX2 crashes regularly. With .txt, it apparently works, as DVX proceeds with the import, but then, when I open the file to be translated (and which has already been translated with Wordfast - I just wanted to become familiar with DVX), it doesn't find one single translation, which sho... See more Instead of starting a new thread, I thought to just continue this one, as the title seems appropriate.
I have spent hours trying to import my TM in .txt as well as in .tmx (converted by Wf). With .tmx, DVX2 crashes regularly. With .txt, it apparently works, as DVX proceeds with the import, but then, when I open the file to be translated (and which has already been translated with Wordfast - I just wanted to become familiar with DVX), it doesn't find one single translation, which should all be 100% matches though.
I followed the manual, but when the wizard asks to specify the "field information", I'm lost and I think that's where the error occurs. Unfortunately, the manual doesn't explain any further details.
How do I know "how the fields in the text file are interpreted"?
Furthermore, why the manual says (in the step before) that the text delimiter of .txt files are commas, whilst in the screenshot tabs are checked? I tried both, just in case...
I have gone through the whole procedure different times, each time restarting from the beginning, deleting the just created TM and creating a new one and importing my TM once more, always with the same results...
Now I am stuck and need some help.
Thank you in advance! ▲ Collapse
| | |
|
|
|
| Import of WF Classic TM | Jan 8, 2012 |
Christel Zipfel wrote:
Instead of starting a new thread, I thought to just continue this one, as the title seems appropriate.
I have spent hours trying to import my TM in .txt as well as in .tmx (converted by Wf).
With .tmx, DVX2 crashes regularly.
It may happen.
DVX is very sensitive to malformed TMX files and Wordfast sometimes produces files with illegal characters.
I would try to load the file in Olifant and repair or delete the segments with invalid characters, then import.
With .txt, it apparently works, as DVX proceeds with the import, but then, when I open the file to be translated (and which has already been translated with Wordfast - I just wanted to become familiar with DVX), it doesn't find one single translation, which should all be 100% matches though.
If the file was imported as Trados txt, this behavior is "normal".
As the structure of the Wordfast and Trados txt files is different, DVX is unable to import it although the .txt extension is the same.
AFAIR it doesn't display any error message, it's somehow confusing.
I followed the manual, but when the wizard asks to specify the "field information", I'm lost and I think that's where the error occurs. Unfortunately, the manual doesn't explain any further details.
How do I know "how the fields in the text file are interpreted"?
I see you selected te text file (CSV) as source.
I don't remember exactly the procedure but it should be something like:
- open your txt file in a decent text editor (e.g. Notepad++ or similar)
- strip the first 5 lines (the file should start with something like:
20110415=181607 GG 0 EN-US Scanning PL-PL Skanowanie EL ST
- add an empty line at the beginning and write:
Date(Tab)User(Tab)Frequency(Tab)Source language ID(Tab)Source(Tab)Target language ID(Tab)Target(Tab)Attribute 1(Tab)Attribute 2
(in fact, it's not mandatory but it should help you to get more human display in the wizard).
- in the wizard, select tab as separator and map at least the Source and Target columns (5 and 7) as your Source and Translation languages (you can also map user etc. but start with the simplest structure), select the "First row contains field names (if wyou wrote the line above).
- select the proper encoding for every language (probably Western (Windows) or something like that, the second possibility is Unicode (UTF-16)).
It should work.
Furthermore, why the manual says (in the step before) that the text delimiter of .txt files are commas, whilst in the screenshot tabs are checked? I tried both, just in case...
The CSV (Comma Separated Value) is a "traditional" name for files with separators, it's often used incorrectly
In the modern CAT world, it means mostly TSV (Tab Separated Value).
You should select only one option (Tab).
I have gone through the whole procedure different times, each time restarting from the beginning, deleting the just created TM and creating a new one and importing my TM once more, always with the same results...
Let's see now...
Cheers
GG
[Edited at 2012-01-08 18:23 GMT]
| | | | | Thank you very much, Grzegorz | Jan 12, 2012 |
and sorry for the delay.
It may well be that the TM ist damaged, it is my biggest one and I started it many years ago...
So I downloaded and installed Olifant but wasn't able to find the repair function or however it is called.
As soon as I have been able to repair it, I will try once more and keep you posted.
Anyway, the file was not imported as Trados .txt. as I don't even own Trados.
| | | | | Olifant option | Jan 12, 2012 |
Christel Zipfel wrote:
So I downloaded and installed Olifant but wasn't able to find the repair function or however it is called.
I believe the option meant by Grzegorz is: File > Import > Check for invalid XML characters (tick that).
| | | | | Finally managed! | Feb 17, 2012 |
Here I am again.
Only this week had I the time to get back to DVX2 trying once more to import my Wordfast TM. Unfortunately I have not been able to solve with Olifant, although it may be an excellent tool.
Each time I tried, both with Olifant and Wordfast, to export my TM as tmx, the file was void or nearly.
After having tried without success the Wordfast special filter "Repair and compact", finally I used "Mark suspicious entries". I had no idea what this meant... See more Here I am again.
Only this week had I the time to get back to DVX2 trying once more to import my Wordfast TM. Unfortunately I have not been able to solve with Olifant, although it may be an excellent tool.
Each time I tried, both with Olifant and Wordfast, to export my TM as tmx, the file was void or nearly.
After having tried without success the Wordfast special filter "Repair and compact", finally I used "Mark suspicious entries". I had no idea what this meant and didn't find any explanation. Anyway, it found some and at this point, I deleted them without mercy as I thought this could be the only way to get finally a .tmx file. And lo and behold, I got it! My .txt file of 25 MB has been converted in a .tmx file of 36 MB and successfully imported in DVX2. How can I see how many entries it contains in order to check how much of them I have lost?
This problem has cost me several days, though, and until now, I have never been able to use DVX.
And then they say that most CAT tools are compatible with each other!!!
Anyway, thanks to this special Wf filter I finally have solved with this one that is my biggest TM and hope that in the same way I will be able to import little by little all my other ones.
Just wanted to share this.
Thanks for your advice, Dominique! ▲ Collapse
| | |
|
|
|
| TM properties... | Feb 17, 2012 |
Christel Zipfel wrote:
How can I see how many entries it contains in order to check how much of them I have lost?
You need to:
1. open the TM you have just imported/created (the .dvmdb file)
2. from the Translation Memory menu choose Properties
3. go to the Statistics tab and
4. click Calculate.
Regards,
Chris
| | | | | Importing TMs | Feb 20, 2012 |
If you are importing formats that you are not too familiar with I would advbise you to do so with an empty DVX2 TM first, so that you can then open the DVX2 TM and inspect the contents. This way, you can look at what was imported and make sure it is correct, and if it is not, you may see clues to indicate what went wrong.
This is especially useful if you are importing formats that require you to specify things (like importing TEXT) where you may make mistakes at first.
... See more If you are importing formats that you are not too familiar with I would advbise you to do so with an empty DVX2 TM first, so that you can then open the DVX2 TM and inspect the contents. This way, you can look at what was imported and make sure it is correct, and if it is not, you may see clues to indicate what went wrong.
This is especially useful if you are importing formats that require you to specify things (like importing TEXT) where you may make mistakes at first.
Importing from Wordfast into DVX2 can be difficult. Wordfast does not always produce correct XML, but DVX2 demands it. ▲ Collapse
| | | | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » Questions for Déjà Vu X2 experts | Anycount & Translation Office 3000 | Translation Office 3000
Translation Office 3000 is an advanced accounting tool for freelance translators and small agencies. TO3000 easily and seamlessly integrates with the business life of professional freelance translators.
More info » |
| | Wordfast Pro | Translation Memory Software for Any Platform
Exclusive discount for ProZ.com users!
Save over 13% when purchasing Wordfast Pro through ProZ.com. Wordfast is the world's #1 provider of platform-independent Translation Memory software. Consistently ranked the most user-friendly and highest value
Buy now! » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
