Buy or upgrade to Studio 2019 today at a special pre-launch offer price

AND get Studio 2021 for FREE on release!   

In 1991, Sprachen- und Dolmetscher-Institut in Munich offered a terminology course for bachelor's students in the translation program. There were three of us on the course, two chose to work with MultiTerm. I established some three hundred terminology entries for the medical domain with a DOS-based tool called Termex. I still have them on a 5 ¼ inch floppy disk. Technology had just started to exist for this budding and constantly evolving grace field. But I don't recall ever getting a book or document to read. Written material for practical terminology work was simply scant.

In the decades since, I met and became friends with Termex inventor, Alan Melby; MultiTerm has competition; terminology is a subject that everyone, from freelance translators to language service providers to large companies and organizations, at least dabbles in. And yet, there isn't a wealth of practical material on the market. Except, in 2010 colleagues from industry and academia put together an easily updatable folder in German under the umbrella of the German Terminology Association, Deutscher Terminologie Tag e.V. Those of us who could read it were thrilled. There was finally a comprehensive handbook that covered the basic theory, tools, project and process management, the ROI perspective, even copyright, and many practical aspects that so many of us had had to learn the hard way.

Granted, Terminologiearbeit-Best Practices presents the Austrian, German, and Swiss angle on the subject. Precision and detail-oriented, with a focus on skills and process, efficiency- and quality-driven. But why not? Even those sectors of the translation industry that may be more speed and quantity-oriented have recognized that consistency is a good thing. Anyone starting out needs a vision to aim for. Budget, infrastructure, available skills, existing processes, corporate, country or industry culture will influence where your terminology project lands on the spectrum. You may be a freelancer doing terminology work while translating. You may be a PM at a language service provider. Or you may be working inhouse with a larger team. You may start with a spreadsheet or a full-blown terminology management system. Terminology Work-Best Practices will provide guidance.

The English translation was just published by DTT this March. I started the translation during a lull in business. Since DTT had put together the original with volunteers, there was no money allocated for translation, although the association provided me with a small stipend. Prof. Sue Ellen Wright took work done by her former student, Julian Bisping, for his Master's thesis, and transcreated the module on term formation. She also updated the part on TBX, and she and Mark Childress proofread my translation. Prof. Klaus-Dirk Schmitz provided the layout work and project management on the DTT side.

To be clear, none of us is benefitting financially from the sale of the handbook. But we all see a win for the community in having this resource more widely distributed. For more information, including the contents, check here. To order follow the instructions on this page.

You can reach Barbara at bikterminology.com or [email protected].  

Translate your Office documents or CAT bilingual files
using the MT of your choice

GT4T is an intuitive MT add-on to all your Windows/Mac apps. It is now also a file translator that supports a great number of file formats. Get your Office documents or CAT files pre-translated without having to upload to a server. Click here to get a free license:  

gt4t.net/en/gt4t_get_free.php 

The magazine Christianity Today or CT might not be a familiar publication to many of you, but it's a really big deal in the Christian sphere here in the US (and for a short while it became a big deal beyond that when its editor in chief released an editorial calling for Trump's removal from office). (It's very well-known to me since I've published a couple of articles in CT about Bible translation over the years.)

Outside the US, it has a presence in South Korea with a Korean edition, and there had been some Spanish translation activity for a while. But last year, the then-new CEO decided there needed to be a much stronger emphasis on international expansion. A third of the articles online were already read by folks outside the US, but, with the few exceptions mentioned, in English. So, a three-year-plan was set up during which the first two years were supposed to be spent building relationships with other organizations around the globe, with some increased translation activity in the third year. Then COVID-19 happened, and any international travel had to cease, making the relationship-building component a lot more difficult.

So, Jeremy Weber who was and is in charge of this global outreach instead decided to go full throttle on translation a couple of years before originally planned. As a result, within the last two months Christianity Today has published more than 100 translated articles in Spanish, Portuguese, Chinese (S+T), Korean, Indonesian, Arabic, Russian, French, Italian, and Tagalog. You can see a collection of some of these articles, these mostly related to COVID-19, right here.

In the past, I've written about publications like the New York Times that I lauded for translating some of their articles into relevant languages (such as articles about China into Chinese, articles about Korea into Korean) but I was quite frankly gob-smacked to find this array of languages in a publication that is a) not as well-financed as the New York Times and b) has a very different focus.

So, I talked to Jeremy and asked him about his goals, his selection of languages and how he organizes the translation efforts.

Just like me, you'll have been surprised at the languages being covered (or not covered) by this effort. It's surprising to see Indonesian and Tagalog in the array of languages, for instance, and yet, according to Jeremy, the Indonesian articles have, to his and CT's surprise, actually been the most accessed of all translated versions.

He describes this whole effort as a sort of experimental (a let's-see-what-sticks kind of attitude) but also very user-driven. Readers are regularly polled about what languages they would like to see translations in and which articles they think should be translated and are then involved in the translation effort. This is by doing an initial machine translation which is then looked over by reader-volunteers and lastly handed over to a professional translator (who eventually actually gets to sign the translation with his or her name).

Most of you will agree this is probably not the most productive or time-efficient way to get translated articles published. But it seems to me to be a very effective way to build enthusiasm about translation within your user group and at the same time pride and a sense of ownership of the product.

While the process was different, it reminds me very much of what Facebook did many years ago with its first attempt at translation. It crowdsourced the translation, received a lot of publicity (admittedly negative from many translators) but from which it benefited overall. It then realized that it's much more effective and reliable to have professional translators involved from the get-go. I told Jeremy I think this is likely to happen here as well -- if of course there are funds to continue the translation efforts in the long run. But that's the other thing I told Jeremy: Maybe not all of this will stick, but certainly enough to make a difference, even long term.

To circle this back to the introduction of this Tool Box Journal: I l-o-v-e this story. It's a story about an attempt to deal with the current crisis by not cutting translation, but adding it. In a new and unexpected way, one that doesn't fit any preconceived ideas that we have and shows that you and I will continue to have plenty to do: as translators, consultants, or even as cheerleaders.

AIT's team response to help translators through difficult time:

Tool Box Journal readers can get the software 60% off to manage your projects and count words in 70 file formats! Order at https://www.translation3000.com/stay-safe. 

If you or your company has been hit by economic crisis, tell us and we will send you 1 year complimentary license of our tools.

If you are a medical translator, write to [email protected] and get TO3000, Projetex, or recently released AnyCount 4D for FREE forever!  

For a while, I've been interested in how much translators who use Trados Studio AND MT use its internal SDL Neural Machine Translation solution in comparison with the big, generic MT providers: Google, Microsoft and DeepL.

It's an interesting question to my mind because unlike its generic counterparts, SDL's solution is customizable and virtually free (to up to 1,000,000 characters a month during this crisis, otherwise half that amount). Plus, SDL Trados Studio remains the market leader among translation environment tools so whatever happens in that environment is an important indication of what machine translation engine translators are using (if any).

SDL's Daniel Brockmann was kind enough to look at some data for us from the time the SDL NMT offer was productized (he used the telemetry numbers of the voluntary Customer Feedback Program in Studio which 40-50% of users opt into). While he didn't divulge actual numbers (aside from this: "We have just recently crossed the 1 billion characters for our free offering") he gave us a graph that shows the amount of data being processed with the different engines (excluding Microsoft, which has a much lower usage). It turns out that SDL NMT is being used about twice as much as Google or DeepL. Or, expressed differently, DeepL and Google combined are being used about as much as SDL NMT.  

 

There are a couple of things to take into account when looking at those numbers (especially when it comes to comparing DeepL to the rest). DeepL serves 12 languages (and any combination with those languages via pivot languages, i.e. a language that serves as a bridge between two other languages), Google serves 109 languages (and uses pivoting as well) and SDL serves 52 languages (and doesn't use pivot languages but has some languages, including Spanish, Portuguese, German, Arabic, and Italian that offer up to seven direct combinations to other non-English languages, plus Chinese (S)<>Japanese -- see here for a complete list of languages).

What this means is the languages that DeepL serves have a proportionally larger share than indicated here (and the fact that DeepL makes it so difficult to purchase a license outside the European Union also muddies the picture a bit.)

Concerning the recent spike of Google in comparison with DeepL Daniel remarked this: "Also it is interesting to see that Google seems to have picked up recently while DeepL seems to have plateaued. This can always shift, however, depending on who offers what in the market of course, and we can then see this also in our numbers."

Just for the sake of consistency (knowing full well some of you will likely fall asleep as you read this because you've heard it so often), I still maintain that unless you have a highly trained engine, machine translation is probably used most productively, and ensures the highest quality when you use more than one engine at a time, "harvesting" fragments from different ones.

Save on memoQ now, and upgrade to v.9.4 for free

Don't you have your own memoQ translator pro license yet? memoQ offers 30% off on all new translator pro licenses for Tool Box Journal subscribers! Read our guide on the purchase process and apply code CAM_TBJ202030_PS in the memoQ webshop. Offer expires at midnight June 17 in any time zone.

Get memoQ and upgrade to memoQ 9.4 -- which will be released in June 2020 -- for free!

"This is the age of implementation." -- Kai-Fu Lee

Five Years of Experimentation

Let me start with the big news. This will be my last Tech-Savvy Interpreter column. I've accepted a senior leadership position at the multilingual web conferencing company KUDO Inc. Let me explain why.

Where do I even begin? Do I start with March when all conference interpreting work disappeared in less than two weeks around the globe? Do I begin by mentioning that just about every interpreter's livelihood vanished as assignment after assignment was cancelled? Or do I highlight how medical interpreters suddenly found themselves on the front lines of the COVID-19 pandemic? No. I think I should turn the clock back about five years.

The Tech-Savvy Interpreter column first debuted in the Tool Box Journal in September 2015. In that first installment, I wrote this: "The influence of technology on interpreting is a rising tide that interpreters need to understand to remain competitive and successful." That statement seems patently obvious today, but back then, not so much. We were still in the age of experimentation with remote interpreting technologies. Scrappy, innovative startups on shoestring budgets were building remote interpreting platforms, interpreting management systems, and terminology management and preparation tools specifically for interpreters on the belief that "if you build it, they will come." Just like Kevin Costner in "Field of Dreams," they were following that seductive and often elusive voice of inspiration.

Back then, the idea of using a computer and USB headset for simultaneous interpretation was scoffed at by many. Technology was something interpreters didn't worry about because we just interpret, right? Or so we thought. But it was clear to me that the interpreting profession was on the cusp of radical change. Digital communication technologies were improving and dropping in price at breakneck speed, and artificial intelligence was beginning its next leap forward with the application of deep neural networks. The interpreting profession was on the edge of a cliff waiting for the right breeze to come along so we could safely test our new technological wings and then quickly fly back to safety. Then the winds of the COVID-19 pandemic began to howl, and everything changed.

March 2020: The Month the World Went Online and Interpreting Stood Still

With the emergence and spread of the novel coronavirus and the subsequent COVID-19 pandemic, the world first shut down normal operations and then held its breath hoping things would "get back to normal" soon. When it became clear the way we all live and work would have to change for the duration of the pandemic, governments, international organizations and businesses the world over scrambled to keep operating as social distancing measures were implemented around the globe in an effort to stem the spread of the disease. Suddenly, almost overnight, the world moved online. As a result, videoconferencing became the principal way to interact.

As international organizations, corporations and governments realized they would need to continue to operate virtually for the foreseeable future, remote interpreting quickly became a must-have.

The interpreting profession was faced with a dilemma. For years, professional associations and labor unions had resisted the introduction of remote interpreting in all its forms, concerned about it displacing face-to-face or on-site work and the hard-fought working conditions and remuneration many interpreters enjoyed. But now, that work was all but gone. Conference interpreting associations had been inching toward middle ground through the cautious approval and incipient use of remote interpreting hubs, but now strict social distancing measures around the world made the use of hubs impractical if not impossible.

Interpreters everywhere were initially asking: "Can I work from home?" and "Should I work from home?" But those questions gave way to "How do I work from home?" because for the vast majority of interpreters it was the only possible way to work. It has been heartening to see so many interpreters tackle this challenge head on. From webinars to YouTube videos and from online practice groups to technology demos, forward-thinking interpreters are helping their colleagues transition to this new reality that few foresaw. These initiatives are extremely valuable and should be encouraged and supported.

Everything about how we deliver interpreting has changed. Period. Nothing is what it was just a month ago. The initial shock is beginning to subside, and now we must face the very real challenges of helping an interconnected and interdependent world communicate across languages during a pandemic. It is my hope that we will see at the center of this unprecedented challenge the opportunity to highlight the vital role interpreting plays in global communication. In this, the shift to remote interpreting is not unlike the advent of simultaneous interpretation after World War II at Nuremberg. The circumstances are different, to be sure, but the long-term effects of this shift will likely be as far reaching and profound. This shift will be written about as a seminal moment in the profession.

The Age of Implementation  

So, why would I stop writing this interpreting technology column at a time when the information is needed more now than ever? I'm glad you asked.

The COVID-19 pandemic has moved remote interpreting from the age of experimentation to the age of implementation. For the last five years, I have observed, opined and advised from a distance. I have experimented. I have reviewed technologies as an independent professor and technology geek. I have shared experiences from an interpreter's point of view for interpreters. I have sought to explain to the world what interpreting is and why it matters. And I have learned much by writing my columns, and hopefully, you have too by reading them. But, for me at least, the time has come to stop opining and observing and start creating and implementing.

That's why when I was asked to join the executive team at the multilingual conferencing company KUDO in late May as the Vice-President of Customer Success, I knew it was an opportunity I couldn't let slip by. KUDO is in the center of a broad worldwide effort the help international organizations, NGOs, corporations, educational institutions, and governments transition to cloud-based multilingual meetings. Remote interpreting is in the age of implementation, and I want to be part of it at this crucial time. In a sense, I'm donning a team uniform. I'm stepping out of the bleachers and onto the playing field. As I do, I'll be helping make the transition to virtual and hybrid meetings as smooth as possible for interpreters and clients alike, not just rooting from the sidelines as others do it.

Thank you to all my faithful readers. Thank you, Jost, for having the foresight to include interpreting technology in the Tool Box Journal and for trusting me to write about it for the last five years. It has been an honor and a privilege.

• Integrated into Memsource, memoQ, ONTRAM
• Compatible with all CAT tools.
• Fast log-in for Proz members.

Every once in a while, I add to my collection of "Characters with Character." Here's one that both looks kinda funky and is unique to one language, even though that language s using the Latin alphabet as a base for (one of) its own alphabet(s).

If you have a Javascript-enabled browser, hold your cursor over the image of the letter to read the associated story.  

I've rarely had quite such a response to the suggestion to use the keyboard shortcut WinKey+V on Windows 10 to store and access past clipboard entries. I do love it and use it a lot. But I thought I'd mentioned it so many times before in the newsletter and elsewhere!? But, hey, who am I to think that everything I say is always being read. . . <

A few weeks ago, I had asked the translation environment vendors what they would do to help translators in this current crisis and published their responses in the Journal. Yves Champollion from Wordfast, in turn, asked Google and Microsoft to lift payments of their API access for professional translators. This hasn't exactly happened but Google has lifted payment for access to their "website translator" for government, non-profit, and/or non-commercial websites (e.g. academic institutions) that focus on COVID-19 response (see here). Does that also strike you as a bit (too little and) too late?

And lastly, I've been holding on to this quote from Jay Marciano, formerly of Lionbridge and arguably the most important bridge builder (not sure how intended that pun is. . . ) between translators and machine translation. Maybe now's a good time to use it:

"Neural networks are based on an idea of how the brain works when we didn't actually know how the brain works. Today neutral networks are used to find out how the brain works."

Stay well!