About this blog Translator's Shack is a collection of links, news, reviews and opinions about translation technologies. It's edited and updated by Roberto Savelli, an English to Italian translator, project manager and company owner of Albatros Soluzioni Linguistiche, a team of English-Italian translators, which hosts and supports this blog.
The Life as a PM category, managed by Gabriella Ascari, contains topics that are less technical in nature, but which we're sure will be appreciated by owners of small translation businesses and freelancers.
Here are links to my pages on some social networks:
Highly recommended:
|
 Riccardo Schiaffino’s excellent blog About Translation had a new intriguing post yesterday about circumventing a limitation in Trados that will not let you run two copies of the freelance version on a LAN. He suggested a hardware-based solution. Read more about this on his post.
However, this reminded me of a quick hack I pulled off a couple of years ago when we were still using Trados quite a lot:
Hi Riccardo,
good tip, but I can recall a better one from the time we were still using Trados quite a lot (we later transitioned to memoQ and never looked back) and some freelancers were hooking up their laptops at the office for the purpose of importing/exporting files and memories.
As a workaround for this license limitation I installed a free copy of a software-based firewall (I think it was Comodo Firewall, but the functionality is similar in other firewall packages). I then disabled some of the packets that the newly-connected laptops were sending out on the LAN. I may be wrong, but I think these were the multicast packets. You may need to fiddle a bit with the options before finding the type of packet you need to disable.
With some luck, file sharing will not affected by the modification, while Trados will not complain any longer about the extra licenses.
Legally speaking, I’m not sure how such a modification can be considered in view of the software license agreement. But since you are not touching Trados in any way and are only changing some of the networking features of the OS, let’s say there’s room for interpretations…
The people in charge of Wordfast development have certainly worked hard since the introduction of Wordfast Pro (the multi-platform tool that uses its separate editor, as opposed to Wordfast Classic, which is Microsoft Word-based). Here is the change log for the latest version, released yesterday. Of particular interest is the "Pre-translate entire file" feature, which may simplify the process of translating Wordfast using a different tools, as explained here.
Added Word 2007 and Excel 2007 Filters
Glossary Stemming (Wildcard searches using *)
Text based formats such as XML and HTML can be opened directly using File > Open
TransCheck supported in table mode
TransCheck can be run while translating
Enable/disable Autopropagation
Edit placeables in the target segment
Pre-translate entire file
Open multiple windows
Adaptive leveraging
via Wordfast: Release Notes.
 Kilgray yesterday announced version 3.6 beta of MemoQ. Although a beta, this is a minor version number change (from 3.5. to 3.6), so, unlike what happens with the numerous revision number changes (e.g. 3.5.23 to 3.5.24) that Kilgray uses to solve bugs, this one contains some significant new features. Let’s take a look at the change log posted by the developers:
Our long-awaited DOCX filter.
This is definitely a very welcome change. MemoQ will now be able to handle Word 2007 files natively and, from what I heard, it does that very efficiently and fast. By the way, MemoQ has always been my top choice for handling complex Word files (performing consistently better than Trados TagEditor 2007).
A PDF filter built on the open-source Xpdf tool that allows you to extract plain text from PDF files for alignment and translation.
Like some users pointed out, this is probably a catch-up feature that allows MemoQ to be as buzzword-compliant as Trados Studio 2009. From what I’ve heard about Trados’ implementation of this filter, it’s definitely not a one-stop solution that will solve all the translator’s PDF problems. (Hint: charge extra when you have to translate PDF files. Half of the time, the client will miraculously find those editable source documents).
MemoQ can now be officially installed on 64-bit systems.
This is very important in view of the introduction of Windows 7 that will take place this Autumn.
A feature to auto-insert the best hit when you enter the next segment after Ctrl+Enter. Look under Translation / Automatic Lookup and Insertion.
This is the single feature I’m most happy to hear about. Until now, if you pretranslated a file and automatically inserted fuzzy matches, MemoQ would not warn you if a better match was saved into the TM since your last pretranslation, so you could end up editing the older, lower-matching segment contained in the pretranslated file if you failed to notice the better match in the TM.
An option (by default, on) to show TB hits in their order of appearance in the source, not alphabetically. Double-click the orange icon above Translation results for settings.
Until now the term base matches were shown in alphabetical order. This makes more sense and it should indeed be turned on by default.
The Concordance window can now be left open as you continue to translate.
Another very welcome new feature for users who like to have more reference windows open. Until now the concordance window was a modal dialog that needed to be closed every time you needed to go back to editing mode.
You can now safely use local projects stored on a network drive (only one user at a time).
I was never affected by this problem since we use MemoQ server to connect to remote projects through the internet and we save all project files locally.
We reinstated our old friend, F4, for inserting the fragment assembled hit. The shortcut can be configured, of course.
This is a minor change. I’m sure some users felt very strongly about it and the developers decided to re-introduce this feature.
An improved terminology check in the QA module that will yield fewer false positives for missing terms.
I have not tested this feature. All I know is that previous versions of MemoQ’s QA module left a lot to be desired, so hopefully this will be a step in the right direction.
MemoQ : Message: MemoQ 3.6.2 [beta] announced.
 We sometimes receive files that have been processed using the latest version of WordFast Pro. These are recognizable from the .txml extension.
This format is just a specific XML structure, and as such it should be possible to translate the files using MemoQ after formatting them properly.
Open the files in WordFast and check the total number of segments
Open the TXML file in WordFast and go to the last line, taking note of the number of segments contained in the file. In the picture below note the presence of WordFast’s tag {ut1}.
Also check for the presence of text in the target column. This will cause problems, so, you will need to delete it. The presence of < > signs in the source text may create problems, so you will need to replace them with some different placeholders.
Once you have done these checks, close the file.
Open the .txml file with an editor and replace some strings
- Open the .txml file using the jEdit text editor. Click here to go to the download page. It is important to use this editor because it allows for a very simple search/replace syntax that takes care of “greedy” wildcards. You can obtain the same results using a different editor, but the syntax to use might be different.
- After opening the file in JEdit, place the cursor at the top and choose Search > Find…
- In the Search for field, insert the string below (be careful not to add superfluous spaces if copying from this page):
<segment(.*?)>(.*?)<source>(.*?)</source>(.*?)</segment>
- In the Replace with field, insert the following string:
<segment$1>$2<source>$3</source>$4<target>$3</target></segment>
- Check that the search options are configured as in the screenshot below:
- Click on Replace All, save the file and quit jEdit.
Open the modified file in MemoQ
- First perform a quick check by opening the file you just saved with WordFast. Now the target column should be identical to the source column, tags included. The total number of segments should be identical to the value you saw when you first opened the file in WordFast. After checking this, you can open the file in MemoQ.
- Add the .xml extension to the file name (e.g. filename.txml.xml), since MemoQ likes this better.
- Open MemoQ and create a new project. Call it for instance “Wordfast”, so you can re-use it easily for subsequent projects that involve translating WordFast files.
- Go to the Settings > Source segmentation rules pane. (Warning: NEVER modify these settings using the Tools > Options > Segmentation rules menu because this will affect the global segmentation rules. We only want to change local rules for this project.)
- Select the various segmentation rules in the Rules list on the left and delete all of them. We need to do this (only once if you the re-use the same “WordFast” MemoQ project) so that MemoQ’s segmentation has no effect on WordFast’s.
- Go to Translations > Add document as…
- Select the file with .XML extension and open it.
- The Document import settings window is displayed. Download this this MemoQ XML definition file, then click on the … on top to import this file.
- Click OK at the bottom of the window. The window closes and the file is imported.
- Open the file in MemoQ and check that the total number of segments is identical to the number you checked when opening the file in WordFast initially. MemoQ should also have inserted the any tags in the correct positions corresponding to the tags contained in WordFast.
- Translate the file normally.
- When ready, export it with Export (dialog)
Check the translated file in MemoQ
- Restore the .txml extension and open the translated file in WordFast. You should get no error messages. Check that the total number of segments is still the same, check the tag position, etc.
- Confirm all segments in WordFast (the only way I know is to press Alt – down arrow in each segment.
- Save and deliver the file.
A brief manual, including screenshots, has been posted for the open source CAT tool Anaphraseus. Here are some of Anaphraseus’ main features:
- Text segmentation
- Terminology recognition
- Plain-text TM (Unicode UTF-16)
- Fuzzy search in Translation Memory
- Unicode UTF-16 TMX export/import
- User glossary
- OpenOffice.org extension
Update: the previous link does not work any longer. Apparently the domain has expired. A new version of the manual is being published on this page.
 An e-mail sent from XTM today describes some of the new features offered by XTM (“the leading web 2.0 complete translation environment”) version 4.0:
- Documents can be split into bundles of segments to allocate to multiple resources.
- Multiple translators and reviewers can work simultaneously on one document.
- Greater flexibility within the workflow to reroute jobs and reallocate resources.
- Many more standard workflows available for selection.
- Enhanced project metrics showing the progress of each step of the workflow.
- Individual translator statistics for progress monitoring and reporting.
- Improved performance from XTM Engine for analysis and matching.
- PDF filter – XTM can now process PDF files.
- Faster and more consistent processing of complex .doc files.
To sign up for a free test drive of XTM version 4.0, click here
Jaap van der Meer of TAUS (Translation Automation User Society) says in his recent post about Google Translation Toolkit:
Translators using the Translation Toolkit ‘share’ their translations with Google. If 100,000 translators start using the service, Google will be harvesting 50 billion words of good quality translation data per year to help Google improve their automatic translation engines. In addition translators may be uploading their own (or their customers’) TMs.
Read the ret of the post at the URL below:
Google Translation Toolkit | Technology.
This link came through Twitter this morning. Google has taken one more step towards implementing a web-based translation environment that supports both human and computer-generated translations. Here is a brief description from Google Translator Toolkit Help:
Google Translator Toolkit is part of our effort to make information universally accessible through translation. Google Translator Toolkit helps translators translate better and more quickly through one shared, innovative translation technology.
Here’s what you can do with Google Translator Toolkit:
- Upload Word documents, OpenOffice, RTF, HTML, text, Wikipedia articles and knols.
- Use previous human translations and machine translation to ‘pretranslate’ your uploaded documents.
- Use our simple WYSIWYG editor to improve the pretranslation.
- Invite others (by email) to edit or view your translations.
- Edit documents online with whomever you choose.
- Download documents to your desktop in their native formats — Word, OpenOffice, RTF or HTML.
- Publish your Wikipedia and knol translations back to Wikipedia or Knol.
How is this different from Google Translate? Google Translate provides ‘automatic translations’ produced purely by technology, without intervention from human translators. In contrast, Google Translator Toolkit allows human translators to work faster and more accurately, aided by technologies like Google Translate.
Here’s a 1 1/2 minute YouTube video that illustrates the main features.
Google Translator Toolkit basics
Here is the changelog:
- TM Precedence – Adds function in TM Profile to allow arranging of TM order. Presents TM matches to translator in that order.
- Download All Offline Files – When translating or reviewing offline, new button allows for download of All Files from All Tasks with status “In Progress” with one click.
- Consolidate TM’s for offline work – When translating or reviewing offline, allows user the option of consolidating TMX files into one when working a job with multiple source files.
- TermBase API’s – Allows access to Terminology Management through Web Service API’s. Perform tasks such as add, edit or delete of terminology entries from an external application.
- New TermBase Search – Provides more flexible and granular search capability on terminology entries.
- SID Support – provides support for String ID fields in GlobalSight TM’s.
- XML Rules Testing – Allow ability to test an XML file when building XML Rules (similar to testing SRX).
Via GlobalSight 7.1.3 Streamlines Offline Translation Process
 The English-Bulgarian freelance translator blog has a post detailing some of the new features that will be available in Trados Studio 2009.
Judging from the content posted on SDL’s website, it seems that Trados Studio 2009 will be presented during the SDL Trados Roadshow Maidenhead (on April 22nd).
Let’s take a brief look at some of the features:
An unified interface – no more Translation Workbench + TagEditor or Word, everything will be in a single application – SDL Trados Studio. All file filters, project management tasks, and the actual translation work will be done in the same window.
Wow, this could certainly be seen as a “bold” move, considering the vast amount of translators who still use the Word macros + Workbench duo to do the job. Although I welcome this change, I’m not sure how well it will be accepted by freelance translators in general.
A two-column view in the translation editor just like SDLX, Idiom, Déjà Vu or Accross with a stronger orientation toward WYSIWYG mode. The placeables and term hits in the source sentence are presented like in Workbench (blue and red lines) and the upper case words work now as placeables.
The two-column view seems to be the general trend nowadays and Trados is one of the last tools that is adhering to it.
A live preview capability enabling the user to see the changes that are being performed in the translation editor pane.
I have seen this feature in action in MemoQ and I think it is essential for any serious computer-assisted translation tool.
The translation memories will be stored in MS SQLite databases (for the freelance editions of SDL Trados Studio).
I wonder if this will have any effects on speed. So far Trados always shined in terms of access speed to local translation memories and was disappointing (in my opinion) when accessing shared TMs though the internet. Perhaps the migration to an SQL database format will mean not-so-great performance on local access but acceptable access speed for internet access.
The TTX format will be replaced by XLIFF and XLIFF will be the actual translation format from now on. You can still open TTX files created in previous versions. The new projects in the XLIFF format however will be no longer compatible with the previous Trados versions. The TM exchange format TMX will continue to be supported.
This is the single best piece of news in the whole package and it will mean (hopefully, unless SDL have screwed up the XLIFF format by creating their own flavor) much better interoperability between different translation environment tools (TEnTs). Ideally, translators will be able to use the tool they prefer and deliver files that can be read by Trados (or any other tool that properly supports XLIFF).
The termbase exchange standard TBX (more info here) will now be supported.
Same as above, although not as big a deal as XLIFF.
Real-time quality assurance features will be supported, which means that the errors will be displayed as you translate.
Hmm, sounds like one of the options I’ll turn off immediately.
The AutoSuggest feature will interactively suggest terms as you type, drawing terms from the content within the translation memory. The freelance version of SDL Trados Studio will not be able to build the AutoSuggest dictionaries but will be able to use externally created ones. When you begin typing the translation text a little drop-down menu appears under the cursor suggesting possible word choices.
I’m not so sure about this one, either. If properly implemented, it might be useful.
The PerfectMatch feature (i.e. in-context matching) will be now available in the freelance edition but it will leverage the matches from the existing translation memory, not from the previously translated files.
This is another feature I love in MemoQ and which has been introduced in Trados. Although my impression is that several translators do not know what “in-context matching” really is, I think it’s a feature that can largely improve the reliability of 100% matches, which, if inserted blindly, can create serious problems. To think that some agencies refuse to pay for 100% matches and are happy enough to insert them automatically!
Via New features of SDL Trados Studio 2009 | News | Bulgarian freelance translator
|
