There was a lot weeping and gnashing of teeth (or as Wycliffe so lovingly put it: "wepyng, and grynting of teeth") among translation environment tool vendors this week. You may know that for the last few months, many tool vendors have been working hard to respond to a tender of the EU to replace Trados as the main TEnT used by the EU. This tender was truly of historical scope (as far as our little industry is concerned) -- many would make the case that it was that original EU contract that initially gave Trados the push to become what it is today. Supplying the largest translation client and supplier in the world just might do that for you. So I and many, many others have been eagerly waiting to hear the outcome.

AND THE WINNER IS . . .

. . . err, no one.

In the words of the official release: "The awarding procedure has been declared unsuccessful" because "none of the submitted tenders met the minimum quality criteria."

Well, I happen to know that some of the vendors worked very hard and invested considerable resources to live up to the desired criteria. (At one point, SDL, Wordbee, Kilgray, Lionbridge, XTM, and Star were in the running, and there may be others that I didn't know about.) So the question is: What now?

I asked the EU for some more clarification, and after a few attempts I received this response:

Thank you very much for your interest in our call for tenders.

The contract has not been awarded because none of the submitted tenders met the minimum quality criteria, as specified in the notice published on the Official Journal of the European Union.

However, I have to ask for your understanding, as I cannot give you any information about why the individual tenders did not fulfill the quality criteria.

Did we learn anything new from that? No!

I hope that others find out more and let us know at some point. I have a hard time believing that the whole tender process will be launched again. So the question remains: What now?

Is there going to be an award given to someone to custom build a tool? Will there be a whole different approach with a stronger reliance on machine translation? Will things stay where they are?

Hopefully we'll know soon.

Successful Translator Rules

Rule #1: Invoice Promptly, Get Paid in Time

Rule #2: Never Miss Deadlines

Rule #3: Build Strong Client Relationships Succeed with Translation Office 3000!

Start FREE trial!

Get it with 20% discount from ProZ.com: http://www.proz.com/TO3000 

Files from CAD (computer-aided design) tools have been notoriously hard to translate. And with the exception of Star Transit and our little Polish friend Cafetran, they are not supported by their near-namesake CAT tools (which, of course, are now typically called "TEnTs" or "translation environment tools"). Over the years, I've often reported on CAD workarounds or the failures of those workarounds. This is crucial for most of us because the original CAD tools are . . . Well, let's put it this way: If you think CAT tools are expensive, you ain't seen nothing yet!

I have good news, however. There's a little Mexican tool, TranslateCAD, that can do it all! To be fair, I'm not sure it can do it all, but it works extremely well in .dxf files -- the exchange format that was introduced by the industry leader, AutoCAD. It allows you to seamlessly extract the translatable data into a text file and merge the translated file with the original CAD drawing.

If you want to make sure that it actually worked, you can download the free viewer DraftSight and check everything out before sending it back to the client (who might have to do some reformatting to account for expanded text, etc.).

Now, I know what you're thinking: If it's true that CAD software is so expensive, this tool must cost a fortune.

Well, it's all a matter of perspective, but if your project is not worth the $29 the software costs, it might be a good idea not to accept it in the first place.

With the introduction of new pricing for the premium edition of this newsletter at the beginning of this year, I also introduced "corporate" pricing, which refers to a subscription by folks in a company or organization who plan to share the newsletter with colleagues. And the good thing is that it's a mere $23,000 a year.

Uhhh, just kidding.

The good news (for you) is that it's only $150 a year for up to five users and $250 for more than five users. Not bad. What makes it even better is that you also get a full 90 minutes of free tool consulting with me when you purchase one of these corporate options. This offer is valid for two months after the purchase.

You're welcome! For more information, I'll see you at http://www.internationalwriters.com/toolkit/. (And I look forward to consulting with you soon.)

One of my favorite countries, Poland, is a hotbed of tool development for translation professionals these days.

Not only is there the workflow tool XTRF (whose owners have so far successfully managed to escape my pleas for a demo), but there is also CafeTran, the no-nonsense, inexpensive Java-based tool that has a small but loyal following; the once hopeful but now defunct AidTrans Studio; sweet little tools like TMBuilder, the subject of a recent Tool Kit profile; and now there is also Aliado, a translation environment tool that is clearly geared at the freelance translator, as expressed both in price and features.

I looked at Aliado's website awhile back and admit to having shrugged it off a bit. The tool seemed like a strange combination of lightweight translation environment tool intermixed with some speech recognition technology. So I let it sit for a few weeks, but I decided to come back to it for this newsletter. As it turns out, I was partly right: It is indeed a translation environment tool, and it does offer some speech recognition technology. But I may have erred in my dismissal of it as "lightweight," and I was very impressed with its use of technology.

Aliado comes in two versions:Aliado.SAT and Aliado.CAT. SAT stands for speech-aided translation, and CAT stands for -- well, you know what that stands for. (I'm happy to report that though these names may sound a little jarring, the English of the user interface is flawless.) There is a slight price difference between the two versions (99 versus 69 Euro, and it's also possible to lease the software on an annual basis); the difference is that the first (Aliado.SAT) has all the features of the second (Aliado.CAT) plus the voice recognition.

My first thought was, How can the voice recognition be any good? How can it compete with the Windows-internal voice recognition or tools like Dragon that have invested so much into development? The answer: It doesn't compete. It simply offers access to already built-in voice-recognition programs. (Of course, Windows Vista and 7 come with built-in voice recognition, and the tool also offers access to programs like Dragon NaturallySpeaking or IBM ViaVoice if you have those installed. Only if you still run Windows XP and have no other voice recognition program does Aliado offer to install rudimentary voice recognition for English, Japanese, and Chinese.) After you install Aliado, you can prompt the program to look for available voice recognition applications. And after those are found you can choose your favorite, which from then on will be the one that will be invoked (in the background) when you start the voice feature to dictate.

What are the advantages of running the voice recognition like this rather than in the normal way? There really are not too many differences, but there is a slightly deeper integration that allows the control of helpful details like automatic capitalization, automatic punctuation, and trailing white-spaces at the end of segments. Furthermore, there is a cool feature that allows you to hover over a certain tag or word in the source segment with your mouse and copy that word to the target just by clicking. I'm usually not a big fan of mouse-based features, but this one is helpful, mostly because you are dictating anyway and don't have your hands on the keyboard to start with.

Finally, it seemed that Dragon started up more quickly and was a little less sluggish when running only in the background.

The potentially biggest advantage of this that I see is that it pushes the availability of voice recognition into your face and reminds you that this can be a great usability feature. (Of course, you'll need to be aware that even in Windows Vista and 7, the Windows-internal voice feature is available only in the language of your operating system which is furthermore limited to Chinese, Japanese, German, French, Spanish, and English.)

And the "normal" features? The actual translation interface is in grid form, very much like Déjà Vu or memoQ, including the numeric tag display in curly brackets ({001}). The translation memories -- one main TM and one reference TM -- can be in a proprietary format or directly (!) in the original .tmw Trados 2007 format (unfortunately, it crashed when I tried that, plus you do have Trados installed on your computer if you get it to work). Glossaries are simple text files that display matches right underneath the translation grid, plus there are some nice filter functions that allow you to show only segments with a certain property or status.

You won't be surprised that there is also MT support -- Google MT and Apertium. We'll see how long the Google support will last. As in any other tool, the support can be activated or deactivated to your heart's content.

Any translatable file is internally converted to the XLIFF format -- which has become sort of standard these days -- and the translatable file formats include XLIFF, MS Office 2007/2010, OpenOffice, HTML/XHTML, TTX, RESX, Java .properties. TMX, PO, SRT subtitling files, and text files. In keeping with the spirit of a single translator tool, files are (quite unfortunately, I might say) translated one by one rather than being able to create projects into which many files of the various supported formats can be imported and translated at the same time.

Still, I'm excited about this tool because it adds yet another challenge from the bottom-up to the big (and much more expensive) guys such as Trados, Déjà Vu, memoQ, and Transit. Although this small tool certainly does not have all the bells and whistles of its big competitors, it does have some clever things that "they" don't have, it comes with a price tag that everyone can afford, and it might just have the feature set that some of you are looking for.

I'm always on the lookout for new data sources, and I actually complained several years ago that the many open-source translations are often done in self-contained spaces with little communication between the different ongoing projects or between the open-source and commercial communities. It looks like I was not the only one who did not like that, and Jacek Śliwerski from -- yes, you guessed it -- Poland, started Open-Tran.eu in 2007 with these goals:

Although everybody has full access to the translations of many open source programs, this opportunity is hardly ever used. The knowledge of hundreds (if not thousands) of people translating applications into their native languages is not centralized and thus hardly available to the others. This is the reason why the translations of even the most popular programs are not consistent with each other, which discourages people from using them.

I'd like the Open-Tran.eu service to fill this gap and to provide the translators with a useful tool, which may improve the quality of their translations and increase the consistency of terminology in the open source software.

(You might be wondering: Who is Jacek Śliwerski? He expected that you would ask that, so he provided his Geek Code:

  -----BEGIN GEEK CODE BLOCK-----
  Version: 3.12
  GCS d s+:- a- C+ UL+++ P L+++ E+++ W++ N o? K- w--
  O-- M- V PS+ PE++ Y PGP++ t--- 5? X R tv- b+ DI-- D+
  G e+++ h- r- y+
  ------END GEEK CODE BLOCK------

As I always say: A code block is worth a thousand words!

Need a translation? You might want to check right here -- just make sure that you remove any choking hazards before you read it.)

So, to come back to the matter at hand, Open-Tran.eu is a project that assembles translation data from open-source projects such as KDE, GNOME, OpenOffice.org, Fedora, Mozilla, and various others. It stores all of that in a large database that can be queried by entering a search term (and the correct language direction) on the homepage, entering it into your browser's search field, downloading the whole thing if you're technically inclined, or using search tools like IntelliWebSearch to perform searches. The searches include fuzzy matches, ignore articles and other noise words for many languages, and even use some stemming.

And the languages? There are actually more than 90 of those, including languages such as Mongolian (don't you love the Cyrillic spelling: Монгол?), Swahili, Icelandic, and Amharic.

Every Windows user should have Arial Unicode MS preinstalled on his or her computer. This is a font that covers a LOT more characters than your typical Arial or Times New Roman -- 38,917 characters, to be exact. That's the whole shebang of characters that are covered in Unicode version 2.1. Not bad. Unfortunately, we're at version 6 of Unicode at this point, and there are many living and historic languages that are not covered by the font.

What are the options? Well, you could download specific Unicode fonts for all the languages that you might get to work with at some point (this is specifically directed to project managers), but that would slow your computer to a snail's pace -- the more fonts the slower the computer. Or you can get yourself a few fonts that cover pretty much everything.

You should probably start with Code2000, arguably the most important font that a multilingual project manager should have on his or her system. This is the most comprehensive Unicode font, with more than 60,000 glyphs and an impressive array of languages. I don't want to bore you with endless lists of languages and scripts that you have never seen or heard of (and I can almost guarantee you that there are many among the ones that Code2000 covers) -- but let's just say, if you ever wanted to display Klingon alongside Inuktitut and Amharic, Code2000 is for you.

The same person who created Code2000 also created Code2001 and Code2002, Code2000's significantly much smaller siblings that mostly support certain historical languages. You could download these, but it's probably not essential if you don't deal with historic languages.

On the other hand, a font that you should download is called Everson Mono. It covers fewer scripts than Code2000 -- it includes Armenian, Canadian Syllabics, Cherokee, Cyrillic, Georgian, Greek, Hebrew, Latin, Ogham, and Runic -- but some of these are not covered by Code2000.

And finally, you should also look at Doulos SIL. This font does not cover as broad a range of languages, but it includes virtually every feasible character needed for almost any Roman- or Cyrillic-based writing system. This crazy collection of "r" glyphs would not be possible with any of the other fonts listed above.

If you round up your existing assembly of fonts with these babies, I can (almost) promise that you won't ever have to scratch your head again when you see odd-looking squares or other place-holders when you should be seeing text. Whether you actually understand what you're seeing, however, exceeds the terms of this warranty.

As a subscriber to the Premium version of this newsletter you have access to an archive of Premium newsletters going back to May 2008.

You can access the archive right here. This month the user name is toolkit and the password is blacklake.  

New user names and passwords will be announced in future newsletters.

If you would like to promote this newsletter by placing a link on your website, I will in turn mention your website in a future edition of the Tool Kit. Just paste the code you find here into the HTML code of your webpage, and the little icon that is displayed on that page with a link to my website will be displayed.

© 2011 International Writers' Group