There is a new service release for SDL Trados 2017. And in the good (recent) tradition of SDL to actually make its service releases worthwhile beyond error fixes, there are a couple of interesting things.

One is the "LookAhead" feature, which starts to search for matches and terminology data for the next segment while you are still working on the current segment. Why is this helpful? Because it at least potentially eliminates any wait time for finding matches when you go to the next segment. You likely have the same question I had, so in the finest Jeopardy form (Jeopardy is an American game show where you have to guess the question to a given answer), let me give you Daniel Brockmann's answer sans question: 

"Indeed, it will skip pretranslated segments and look at the first fuzzy match down the line."

That's good.

Also, embedded Excel files within Word and PowerPoint files are now automatically processed. That's something others have provided for a while, but it's nice to have it in SDL Trados as well. Of course, it would be nice to have other embedded (Office) file types processed, too.

And then there is the beta of Language Cloud Terminology. This is an interesting proposition aimed at freelancers or small companies who don't want to (or can't) afford one of the relatively expensive GroupShare licenses but still want to share terminology databases between different translators. According to Massi Ghislandi, "it is not meant to replace MultiTerm but it is an additional resource at this stage -- not as sophisticated as MultiTerm -- to create, store and share terminology in the cloud." And to quote Daniel again, it's

"a 'true' cloud-based terminology offering that is more than a glossary with a more or less basic data model behind it. The beauty is the ease with which you can ... share a termbase with anyone else -- i.e., a terminology service for everyone -- or import an Excel glossary, or set up even a sophisticated termbase (or ... a basic one - just as you please), all from within the browser (and then access from within Studio)."

He's right, there is a choice between a basic and an advanced, concept-based data model when you create the termbase online (you'll find the link to Language Cloud Terminology from within SDL Trados Studio or via this portal). Once it's created online in your account, you'll find it as a selectable terminology resource within SDL Trados, or if it was selected by someone else, she can send you the link to the termbase and you connect that way. During translation, the terms from the cloud-based termbase will appear just like they do from MultiTerm, and you can add terms to the termbase in the cloud on the fly.

Daniel also mentioned that it "is still a bit rough around the edges," and that's true as well -- for instance, I couldn't get the import feature to work -- but everything else worked as it should have as far as I could tell.

Here what really interests me about this: Right now this is still in beta, so it's free and will be until the end of the year. After that it will be a paid offering with the price yet to be determined by SDL. This means that SDL is moving into a hybrid role as a SaaS (Software-as-a-Service) and traditional software vendor. The mainstay desktop application acts as the starting point, and other products/services -- whether cloud-based termbases or machine translation -- are sold on top of that.

I asked Massi whether this should or could mean that "SDL [should] lower the threshold (i.e., price) for purchasing the base product and then essentially customize it with other paid offerings."

His response?

"In terms of the pricing for the desktop and LC terminology -- we have not yet made firm plans and we will continue to evaluate the market situation and the value offered by the combination of Desktop with Cloud.

Over the years we have continued to lower the entry cost of Trados -- a fact that is often overlooked.

• The starting price has not been increased for 10 years; in fact, it has been reduced -- by €100 -- and is listed at €695 with many promotions.
• We will be looking at a number of options in terms of bundles which include a combinations of desktop and clouds solutions."

Of course, there were a whole bundle of other improvements in the Service Release. You can read about them on SDL's fancy new website or in more detail right here.

Translate from anywhere with Memsource!

Work online and offline, on Mac and PC platforms, with an intuitive and powerful CAT tool.

Start a free 30-day trial of Memsource for Translators 

As old-fashioned as TMX (Translation Memory eXchange) files seem to be, they still play an important role in many of our workflows. How do I know? I receive a lot of TMX files that are corrupted and need to be fixed from folks and companies whom I help with technical issues.

Here are some tricks I use, and you can use them, too, to fix corrupted TMX files.

A common problem is the assignment of incorrect or incompatible language indicators. For instance, one tool might use this language indicator for English

<tuv xml:lang="EN">

while another tool requires the definition of the subcategory of (UK) English, as in

<tuv xml:lang="en-GB">

Typically you would receive an error notification indicating a problem with languages before the import process of the TMX file starts. The quick and easy way to fix this is to open the TMX file in a text editor (such as Notepad or, even better, a more advanced -- and free -- tool like Notepad++), search for <tuv xml:lang="EN"> and replace it with <tuv xml:lang="en-GB"> (or whatever your tool of choice prefers). If you're unsure of what the correct snippet should be, quickly export an existing translation memory into a TMX file from your tool and see what it uses by opening the TMX file in a text editor. Repeat the snippet replacement for the second language (and any additional languages). When you close and save the file, make sure not to change the codepage (unless there are double-byte languages involved, it most likely has to remain a UTF-8 file) and save it as a copy of the original file with a TMX extension.

That was easy.

It's a little more complicated when your tool of choice doesn't want to import the file because of "invalid characters."

There are a number of things you can do to fix it.

For instance, you could open the TMX file in the free TMX editor Olifant (don't get confused by the messages on that page apparently indicating that this application is not current anymore). Since it adheres very strictly to the TMX specifications, it will open the file only to the place where it finds the first corrupted character. In a little pop-up window it will then tell you where the corrupted character was found:

You could now open the TMX file in a text editor, search for the line number (Ctrl+G) -- in our case, 168948 -- look for the bad guy (in our case, on position 48), delete it, and close and save the file. If that was the only offender in the file, it should now import the TMX file into the translation environment tool of your choice. So, yes, this works great if there is one or maybe just a couple of problems. What if there are a lot more?

Here are two other process that work.

One uses the super-talented Xbench (this works with the free and the paid editions). Open the TMX file in Xbench, select Tools> Export Items, and export the complete TMX file into a new Xbench version of it, with all the invalid characters taken out. This works great, with the only problem being that it also takes out all inline codes (AKA the tags that deal with string-internal formatting, etc.). In some situations this might actually be a good thing (like when you've downloaded data from the web and aligned it for a reference TM); in other cases, though, you'll actually want the inline codes because you wouldn't get 100% matches without them if you went from one version of a manual to the next, for example.

In that case, you can follow this procedure:

I know I've called Xbench a "Swiss army knife" before -- and that's well deserved. It does a lot of different things when it comes to searching for terminology, performing quality assurance, offering an environment that essentially supports any bilingual format, or -- as mentioned above -- cleaning up TMX files. But there's another tool I don't write about as often because it's just a little more on the technical side. It's called Rainbow -- and in many ways you could argue that this truly deserves the Swiss army knife moniker for its potential benefit for our translation projects. You can catch a glimpse of what it does right here, but you'll quickly see that even this long list of features ends with "and a lot more," an appropriate description of the plethora of tasks it covers dealing with the preparation and execution of files in the translation process.

One thing it does -- yes, you guessed it -- is clean up TMX files. There is no specific command to do this, but since TMX files are XML files, the command you'll find at Utilities> XML Utilities> XML Characters Fixing does an amazing job and finishes lightning fast. One difference is that you don't actually select the TMX file to process in the dialog that pops up when you select that command. Instead, you'll have to add it to the "Input List" before selecting to fix the file. You can do that by selecting Input> Add Documents.

Oh, and while we're talking about TMX files, this all applies to other XML-based files as well, including XLIFF and TBX files. 

Need a target-only, live preview while translating? This feature and more are now available in Wordfast Pro 5.1!

What do California, Colorado, Minnesota, Ohio and New York all have in common? They're all confirmed locations for the Wordfast Pro Roadshow to take place this fall in cities across America!

In the May-June edition of the ATA Chronicle I published an article entitled "Remote Interpreting: Feeling Our Way into the Future." (If you haven't read it yet, just click on the link.) The article highlights the challenges and opportunities that remote interpreting presents for multilingual communication, generally, and for individual interpreters. Just as it has permeated every other aspect of modern society, whether wanted or not, the increased use of technology to deliver interpreting services in new ways is inevitable. 

The article was generally well-received. Then came a must-read response in the form of a letter to the editor from Melissa González, an NBCMI-certified medical interpreter in Austin, Texas. Her letter is a real-life example of what happens when we, as a profession, fail to oversee the changes technology is imposing upon us.

Melissa details in firsthand a disastrous implementation of remote interpreting in the workplace. She tells a tale of poorly implemented technology, disrespected interpreters, frustrated healthcare providers and ill-served patients. This letter to the editor (published in the latest edition of the ATA Chronicle) should be required reading for every interpreter, interpreting agency, technology vendor and healthcare provider considering how to implement any form of remote interpreting. I encourage you to click on the link above and read it now.

From start to finish, Melissa's story illustrates a clear example of remote interpreting technology implemented poorly and for the wrong reasons. I'll share one particularly salient paragraph of the letter here: "Technology is a tool. When used as such, it can help us accomplish great things within our profession. However, when technology is employed as a way of replacing human knowledge and expertise to save money, it can only lead to poor and often dangerous results." I completely agree.

New technologies can and are improving multilingual communication and expanding access to professional interpreting services. But Melissa's tale should give any organization pause that is seeking to employ technology as a principal means of saving money and replacing human expertise. This is a story that must be shared as widely as possible so it isn't repeated.

Furthermore, if the transition and expansion to these new ways of working are to be smooth, interpreters, professional associations and other interested parties must get involved now and shape the implementation of these new technologies. Only through our active engagement will they be used appropriately to improve communication and expand availability of interpreting services. Only through our insistence can interpreters be helped to adapt to a changing workplace.  

This is the purpose of InterpretAmerica 6, which will take place in Washington, D.C. and online on October 30, 2017. This one-day summit will bring together interpreters, professional association leaders, language service company executives and technology providers to spark a collaborative, results-oriented conversation between interpreters and those who hire them. We've all seen the disruption the rollout of remote interpreting platforms is having on working interpreters. But we also see the potential for highly positive outcomes for increased language access and expanded work opportunities.

Do you have a question about a specific technology? Or would you like to learn more about a specific interpreting platform, interpreter console or supporting technology? Send us an email at inquiry@interpretamerica.com.

Awhile back I tweeted about using paper and (fountain) pen at the beginning stage of every article, presentation, or other write-worthy utterance (in case you're wondering whether "write-worthy utterance" has ever been "uttered" before, Google and Bing say it hasn't...). Really though, I love the challenge of the clear white page, the ease with which circles and arrows can connect thoughts, and how new ideas seem to -- sometimes -- grow out of the paper.

My response: What!? This is what you want to get away from? This looks like a dream come true! Aside from the very first day 20 years ago when I assembled my current desk, it has never been that clean! Having established that deep divide between the apparently slightly OCD makers of reMarkable and me, I didn't look any further. For some of you, though, it might be interesting. I, in the meanwhile, will still cherish my white, non-synchable paper and try to rub the ink stains from my fingers when I feel the need to look presentable. 

SDL Trados Studio 2017 Service Release 1 (SR1) is now available. Key new enhancements include:

• LookAhead -- immediate results from translation assets, so there is no more waiting for searches.
• upLIFT -- even more leverage from your translation memories.
• New cloud terminology provider -- create, edit and share terminology via the cloud!

Learn more about SR1 »

Seattleite Sarah Williams contacted me via ATA's upcoming president Corinne McKay to talk about Pacemaker, a product she and her sister developed for all kinds of people who need to achieve a certain goal per day measured in any kind of unit -- most commonly word count goals, as it was originally developed for literary professionals and students. In essence, you enter the number of words of a given project as well as the number of words you want to do per day, and the tool then spreads out over how many days you can and should accomplish the project. At the end of the day you enter the amount you actually accomplished, and the tool then recalculates the rest.

Overall the tool has been welcomed by users, as is shown by 10,000 registered and 2,000 active users.

Awhile back Sarah was contacted by a translator from Romania, who gave her some suggestions on how to make this tool more applicable for translators, such as including the ability to work on multiple projects at the same time. Sarah and her sister have started to implement those changes, some of which you can see right here, which also contains a link to joining the beta program. After the current three- to four-month beta period the tool will not be free, with the more advanced multi-project set at $8/month, so Sarah has promised an additional three free months after the end of the beta period if you indeed join the beta testers.

I think there's potential for this tool to be useful. Some features that I've encouraged Sarah to add include

• An integration right into translation environment tools (such as with an add-on into SDL Trados) so there's no need to manually update the number of words accomplished every day
• The ability to set different starting dates aside from "right now"
• The ability to give different kinds of priorities to different clients

These features would be really helpful to my work process. Everyone's different, though, and some of you might find the tool helpful even now. Either way, I'm sure Sarah would be eager to listen to your suggestions.

August Price Melt by Advanced International Translations. Exclusively for Tool Box Journal readers:

http://special.translation3000.com/toolkit50

10 copies of each product available at 50% off. Get yours, become organized and have more time and money left for fun!

This is exactly the kind of question you could find the answer to with Sketch Engine, the corpus tool I very enthusiastically reported about in the last Tool Box Journal. A little too enthusiastically, it turns out.

I wrote this:

Uhm, not so much. I was wrong, and I don't have any excuse (aside from the fact that I was tired).

It is true that you can download many websites containing certain terms to build a corpus. However, you cannot have them automatically align with a translated version of that website through Sketch Engine.

I'm sorry about that. On a positive note, everything else I highly praised about Sketch Engine was accurate.

I'm going to bed.