Unify information from authoring, translation to omnichannel delivery with:

Solutions to accelerate performance and transform information into an asset.

STAR Group – your single-source communication partner for products and services

I had a conversation about the new Trados with Daniel Brockmann, the evergreen "Principal Product Manager" at RWS (who despite his evergreenness now sports an all-grey head of hair) and his colleague Andrew Thomas, the Senior Director for Trados Marketing. Before the conversation I felt pretty sure we would be talking about Trados 2022, which was just released at the end of May. It turns out that we did talk about that -- but really only very briefly. The rest of the considerable time we spent together was devoted to the new Trados, as in a "newly recalibrated Trados product offering."

Trados is the market leader among individual translators and non-cloud-based installations at organizations, including translation agencies and translation buyers. This has been the case for a very long time. The fact that there is a qualifier ("non-cloud-based") in here has been somewhat of a stumbling block for RWS's Trados team. Accordingly, we talked a lot about being an underdog vs. being the "top dog" during our conversation -- a topic that never would have come up in any of the previous discussions we've had over the many years that Trados and I have been in conversation.

Let's start by looking at the language technology situation at RWS. After they took an inventory of what they had available they organized the many different parts into one business unit with a more or less coherent strategy and goal.

As you likely know, the Trados brand used to be owned by SDL, which along with Trados and all the other SDL technology puzzle pieces was bought by RWS. (Incidentally, RWS has lost about 40% of its stock value in the past six months.) The other puzzle pieces related to translation technology that came with SDL were either acquisitions that SDL had made over the years (Trados itself was an acquisition, as were Idiom with its WorldServer product and MultiTrans via Donnelley Language Solutions, not to mention a barrage of other non-language-related products) or a number of in-house solutions (TMS, GroupShare). Most of these are products that are directed toward similar customers on the high and high-ish end of the market. Customers that you don't want to frustrate with discontinuing any of those products because, as in the case of SDL/RWS, they might well also be buying language services from you.

On top of that, none of these products is a new-generation cloud-based product. (And when I say "new generation" I'm being kind -- it's really not that new anymore!) All of these products employ a server-client-based infrastructure, the kind of infrastructure that was en vogue in the noughts and early 2010's and that allows for collaboration and centralized data hosting but is not in the less strenuous cloud environment that even the more cautious customers are now asking for.

With the ongoing integration of the MT system Language Weaver (another technology purchase, and incidentally the only one that has not been grouped in the newly formed translation technology unit), a Language Cloud product was introduced in 2014 (see edition 238 of the Tool Box Journal). This Language Cloud product was built upon and eventually became its own full-fledged and stand-alone translation management system in 2019 and is now being marketed as Trados Enterprise, the cloud-based translation management solution that SDL is offering to translation buyers.

While RWS has not end-of-lifed (the fact that you can come up with constructs like that is one thing I love about the English language!) any of its other translation management systems and will still provide updates to customers, it will develop new features more slowly for them and "assumes" that the clients who still use them will eventually switch to Trados Enterprise. (This was -- at least from Daniel's and Andrew's perspective -- the preferable solution to switching to a competing product 😁.)

So, Trados Enterprise is the full-fledged translation management solution that RWS is now actively promoting and selling (and I was promised that in the not-so-distant future, pricing and such things will be much more transparent on RWS's website).

The product for translation teams of all sizes is now called Trados Team. RWS bills it as a "cloud-based translation project management solution." It's very similar to the GroupShare product, except everything is cloud-based. (By the way, GroupShare is the only one of the old guard of products that is still being proactively offered on their website -- but just kind of tucked away between marketing copy for Trados Team.)

This leaves the last of the three products, and the one that is the most familiar to all of you: Trados Studio. Trados Studio obviously is the solution for the individual translator and, if used in relation with one or both of the other products, forms the tail-end of any line in the language business, i.e., the one where translation is taking place (aka "the most important one").

Unlike the two other products that have completed the switch to an online, web-based environment, Studio is still a decidedly hybrid product, with the hybridization process in the 2022 edition mostly complete. What I mean by that is that the Online Editor (which, when I first wrote about it in 2016 in edition 263, really was "only" a secondary environment for maybe proofreading purposes) is now a feature-rich environment that allows for productive translation work. (Of course, this doesn't mean all of you will like it -- I know that the prejudices against a browser-based translation environment run deep for many.) You can switch between the desktop and the online editor seamlessly -- and you'll be able to do it even more seamlessly when the desktop editor in an upcoming service pack is updated with data from the online environment in a more automated fashion.

Also, Trados Studio 2022 gives you access to all "base level of cloud computing," so there shouldn't be any extra cost for any online resource you create and use.

Other changes, aside from the typical bug fixes, etc., include relatively minor features such as multilingual Excel, a handful of binary software file formats (there is still a new version of the localization tool Passolo -- but my sense is that it won't have a particularly illustrious future as a standalone tool), and a new "view" in the desktop's main interface. This view is called the Manage view and essentially combines the Project and Files views. This is still in beta, so you'll have to activate it if you want to use it.

What does all this mean? Overall, I think it's positive that in the midterm view Trados Studio will no longer have to serve all kinds of unconnected and discombobulated translation management systems. It should certainly be easier for the technology teams at RWS to move forward with further development now that everything is all in one business unit. ("The current reorg is the best reorg ever" because "it was centered around technology and not corporate," according to Daniel, and I can testify that he has gone through a lot of those reorgs.)

The Trados Studio user can be pretty certain that the desktop environment is going to stay for the foreseeable future, but they can also know that the Online Editor is actually productively usable.

For larger clients, this looks like another "TSO" moment. That's what I called the (Trados Studio) opportunity when Studio 2009 was first released as a complete redesign of the previous Trados (oh, and was it buggy!). Back then I thought this was a great opportunity for a real market shakeup. Everyone who was using the earlier Trados version, the gorilla in the market, had to decide whether to eventually upgrade and relearn a new tool from the same vendor or switch technologies and learn that technology. Amazingly enough, SDL was able to "squash" the "Trados Studio opportunity" and maintained its top dog position -- partly because competitors did not present attractive enough alternatives. Today, RWS is in a different position when it comes to translation management systems. Between XTM, Memsource, memoQ, and a couple of others there are very strong contenders, some of which are in fact already stronger than RWS in the cloud TMS space. RWS knows that, so it'll be fun to see what happens.

Oh, and with the strong reliance on the Online Editor for Trados Studio, I cannot imagine that RWS will offer perpetual licenses for Trados Studio all that much longer. Subscription-based models have been too successful and make too much business sense, especially when an online environment is frequently updated anywhere. But that's just my guess.

The saying that words can be more powerful than weapons reportedly dates back to the 7th century BC and has since found its way into literature in many forms. And indeed, one could argue that many a written word has led to some incredible changes in the real world -- take Martin Luther's 95 theses, or perhaps a newer example of Elon Musk's tweet causing Bitcoin to crash. But how much of an impact can words really have? Do they really matter that much?

Even in our modern world today, we still see many inequalities between genders, people, ages, or abilities. In recent years, however, people have become more aware of how language, specifically inappropriate terms (such as the frequently used IT terms "blacklist"/"whitelist" or "master"/"slave") further sustain these inequalities and decided that it is time to retire them for better terms to reflect a more inclusive world.

But while replacing outdated terminology with new, more appropriate, and inclusive words seems straightforward enough and can solve concerns around racist, ageist, or ableist terms, using inclusive language to remove inequalities does not stop there. One of the biggest considerations -- and current challenges -- of inclusive language is the topic of gender visibility or gender inclusion.

You may have come across the riddle about a father and son being in an accident. The father is killed, and the son is gravely injured. He is taken to the hospital, where the surgeon on duty says, "I cannot operate on this child, he is my son." This makes us wonder for at least a moment, because all too often (and despite the many seasons of Grey's Anatomy), we still think of a surgeon being male. While this in some respect may give the English language an advantage in terms of gender inclusivity -- as it already includes all genders in the generic term "surgeon" -- things are not as straightforward in many other languages.

Take the German language, for example, where an "engineer" is either explicitly male ("der Ingenieur") or female ("die Ingenieurin").

Classroom experiments with Dutch and German primary school students have shown that girls are more likely to pick a STEM career option if the job titles are presented in pair form (for example, "Ingenieurinnen und Ingenieure," female and male engineers) instead of just the generic masculine form. This strongly supports the idea of "what you can't see, you can't be" and highlights why representation, even "just" in words, is so important.

Here at HubSpot, inclusive writing is one of the key considerations for all of our content, and to achieve this we follow an Inclusive Writing Guide that highlights the things to consider when writing, such as age, gender, ability, or ethnicity.

In the guide, we explain: "Words matter and the language we use should be inclusive and welcoming to our community of readers and end-users who have layered identities."

This guide was initially created in English, but with HubSpot operating in many countries across the world, it soon became apparent that similar guidelines are required for the supported languages such as French, German, Japanese, and Spanish. A working group within the in-house Localization team set out to establish the language-specific counterparts -- Operation Inclusive Language was born.

As we got to work to create these language-specific guides, we realized that each language came with its own set of challenges. One of the biggest was often the rephrasing of gender-specific language.

As mentioned, the English language often uses gender-neutral terms ("the customer"), whereas in German, for example, the generic masculine is used. Here, the subject of gender poses the biggest challenge. There are male and female versions of most terms to describe a person. For example, we have "Kunde" (a male customer) and "Kundin" (a female customer). The plural would be "Kunden" -- the so-called generic masculine. This used to be defined as describing all genders, but as the language has been evolving to reflect culture, this is no longer considered inclusive.

For German, using the pair form (male and female), as in our example "Kundinnen und Kunden," is an improvement over the generic masculine form, but it still leaves out non-binary genders.

There are other options for writing gender-inclusively, such as using an asterisk to write "Kund*innen" or a colon for "Kund:innen"; both options are understood to include all genders. But these forms may pose a challenge for screen readers and therefore potentially exclude visually impaired readers.

Although inclusive writing is gaining momentum, there are no established rules (yet) from the German authority on grammar and spelling, the Duden. So, there is currently a mix of the forms described above -- some more popular than others, and all of them also heavily disputed by purists who consider them an unnecessary interference with the language as it is.

Many other languages face similar challenges (see this wonderful graphic article created to highlight the particular challenges around gender in the various languages around the world).

In most, there isn't a perfect solution out there (yet), but that doesn't mean we shouldn't try to be more inclusive in all of them.

Add to that the complexity of pronouns, which are currently very binary in most languages, with "he" and "she" in English and "er" and "sie" in German, leaving no room for other genders and therefore requiring the invention of new terms such as "ze" or "hir".

While some may still consider inclusive language to be the latest "politically correct" fad, this movement is long overdue and not going to be stopped. Inclusive language is already an obvious choice to many organizations, media outlets, and even government institutions, although many practical questions still remain, not least for the translation and localization trade: How far can translators assume their author supports inclusive language? Do we have the freedom to potentially change a tone of voice by adopting an inclusive style? Should we proactively ask our customers to make this choice upfront, if only to put it on their radar?

An additional challenge is added by the fact that the translation industry increasingly relies on machine translation engines which are not (yet) trained to understand the complexities involved in avoiding potentially problematic terms or to pick up on the more subtle tones of specific words in context in order to be able to filter them out, let alone replace them with more appropriate terms in the specific context or even decide to use "Kundinnen and Kunden" for the simple "customer." Certain types of text (think UI, where space is at a premium, or legal documents that must be very close to the source) also warrant careful considerations.

Bearing all these things in mind, it becomes clear that we still have quite a way to go towards a fully inclusive language.

However, I believe we at HubSpot can play a part in making a positive change, at least within the scope of the content we create. Not only is writing inclusively now doing the right thing, but it is also very important in order to shape the world we want to see -- one of which our grandchildren can be proud. In this day and age, many more people have access to a pen (thanks to social media, among others) than a sword, making it even more important to choose our words wisely.

What if the best editing tool you never knew about was in the palm of your hand?

If you've been thinking about tapping into artificial intelligence (AI) to help you write copy or edit translations, you're in luck.

In this blog post -- based on our techforword insiders webinar on AI writing assistants, we'll explore how to use these tools to:

Ready to improve your writing? Read on!

As language experts, we frequently write in non-native languages -- whether for sending a quote to a client, penning a note to a colleague, or writing blog or social media posts.

Spelling and grammar errors can shoot us in the foot, ruin our image, and determine whether we land a job.

That's why having a solid spelling and grammar checker that covers all your working languages is essential for language professionals.

Although spellcheckers are built into most word processors, we're a big fan of finding a multilingual tool that works across a range of software, including right inside your browser, email, and word processor.

Some better-known tools like Grammarly and ProWritingAid are only available for English, so they may not be the best option for multilanguage professionals.

Instead, try LanguageTool. This freemium tool provides spelling and grammar tips in 30+ languages! With automatic language detection and plug-ins for most web browsers, word processors and operating systems, LanguageTool is our go-to spelling and grammar checker. And at just $4.99 a month or $59.90 a year, the Premium version is an investment that's worth its weight in gold

As its name suggests, GPT-3, or the Generative Pre-trained Transformer, is the third in a series of predictive models that have been trained with a huuuge amount of data across tons of languages.

GPT's massive neural network has actually learned to learn! Give it a simple prompt, and it can generate impressively convincing output across a range of genres -- from nonfiction articles to email and marketing content and even creative texts, copy, and code!

Let's dive into some ways language professionals can use AI-supported writing tools!

Email can be the bane of our existence. What if we had a tool that would learn to complete our emails based on our style?

Enter Smart Compose. Currently available for English, Spanish, Portuguese and French, this built-in Google tool will predict your sentences and complete them for you. Time saving, here we come!

To learn more about how to use write faster and smarter and take the overwhelm out of email, check out our Tame Your Email Monster challenge.

Imagine your AI buddy reading your copy -- and rewriting it for you!

Paraphrasing tools will do just that, offering up new sentences or entire paragraphs.

Interpreting technology researcher Claudio Fantinuoli -- the mastermind behind InterpretBank -- recently released an experimental "neural text improving app." Available in English, French, German, Spanish, and Italian, it will correct grammar errors and rephrase sentences. Why not take it for a spin for free?

Our favorite paraphrasing tool is Wordtune. Draft a sentence, select it, and with a single click, Wordtune will serve up some impressive suggestions to improve your text. Scan through them and pick out the one you like. The free version offers plug-ins for Chrome, Edge and Word and 10 free rewrites per day, while the professional version ($9.99/month) includes unlimited rewrites and lets you choose a formal or casual tone and decide whether you want to expand or shorten your writing!

Although Wordtune only offers writing suggestions in English, you can give it a starting text in Spanish, Russian, Korean, Hebrew, Mandarin, and Arabic and get recommendations in English. Machine translation + AI paraphrasing all bundled into one -- pretty nifty!

Looking for an AI-based tool that can rewrite entire paragraphs for you? Look no further than Writesonic, which will generate copy and rephrase texts in 24 languages!

From articles to blog posts and emails to sales copy, AI copywriters can dish up copy in a vast array of genres and styles.

Simply pick the type of writing, language and tone and let the machine do the rest!

Popular multilingual options include Writesonic (24 languages, $15+/month), Jasper (26 languages, $29+/month), Copy.ai (25+ languages, $35/month), and Rytr (34 languages, 5K characters free, then $9/month).

Our favorite, Rytr, takes the cake for its low price point, generous free tier, and impressive output. Pick your output type (from blog ideas to email and profile bios to longer-form copy) and tone (with 20+ options like "earnest," "casual," "convincing," "enthusiastic" and "formal"), give the machine a few bullet points to get started, and get ready to be amazed.

As multilingual language professionals, we write -- a lot.

Emails. Social media copy. Our websites and professional profiles. Speeches. And of course, our translations!

Our words matter. And we're judged on them -- even in a foreign language.

But coming up with the perfect turn of phrase, an email pitch to a prospective client, or a relevant blog post isn't always easy.

After using AI-based writing assistants for over a year, we're huge fans. They've saved us tons of time and offered up interesting copy ideas -- like the introduction at the beginning of this post. 😅

As with any new technology, always use AI with caution.

These tools can offer up impressive suggestions, but you're the one in charge! Simply mix those robots' ideas with your own for optimal, creative results.

So what are you waiting for? Instead of staring at a blank screen or fretting about your grammar in one of your foreign languages, why not take one of these AI writing assistants for a spin?

Josh Goldsmith is a UN and EU accredited translator and interpreter working from Spanish, French, Italian, Portuguese and Catalan into English. A passionate educator, Josh splits his time between interpreting, researching and teaching through www.techforword.com, which empowers language professionals to make the most of technology.

Do you feel stuck with your interpreting practice? Like you're struggling to find the right tools, speeches or practice partners?

Discover the exact tools I used to graduate interpreting school, expand my language combination, and pass the UN accreditation exam.

The power of stories! I'm not sure who told me that Earl Grey tea's distinctive aroma comes from a substance produced by muskoxen. Yes, I was a very trusting child, embarrassingly enough so that I believed that nonsense deep into the tea-by-candlelight-and-incense-sticks days of my mid-teens. Oh well. Turns out that muskoxen do produce a smell but one that is far less pleasant, and the Earl Grey aroma is derived from the peel of the bergamot fruit.

After this long introduction: Enter the Bergamot project, an EU-funded project between the universities of Edinburgh, Prague (Charles University), Sheffield, and Tartu, and Mozilla (the makers of Firefox, etc.). One of the goals of the Bergamot project is a client-side machine translation engine (i.e., a machine translation engine where the processing takes place on your computer rather than on a cloud-based server).

Now Mozilla has unveiled its first iteration of the engine in a beta version that supports the translation of 12 languages into English. It's a browser plugin that does pretty much exactly what other browser plugins for machine translations of webpages do, only that the data is strictly processed by your machine and not by some server in the cloud. Naturally, Mozilla is very prominently touting the security aspect of the new software. I confess that I often get kind of annoyed with this seemingly hypocritical approach. After all, most of the cloud-based MT engines, among them Google, Microsoft, and DeepL, assure their users that they are not going to use their data if the users use their paid API access -- which most professional translators do within their translation environments. But of course, this is not a product for professional translators -- at least as of yet. It is used to translate webpages as you visit them (websites that typically don't have much private data to start with), but it is also employable for forms. Forms are essentially text boxes used for text input, much like in most online translation environments. So theoretically one might assume that this could be an option for the use of this MT engine in those kinds of translation environment tools -- but since the forms translation feature is really not yet working satisfactorily, it's hard to say whether it ever will be useful for professional translators. So, for whom will the Firefox MT plugin be helpful? Likely for people who browse intranets and other non-public websites with confidential information and in languages they can't read, and eventually for people who want to make sure that their data entry in foreign languages is somewhat correct and not recorded and analyzed anywhere else (as it would be if you used publicly available MT without an API key).

Oh, and the quality of the machine translation data? In the German-to-English combination, it's comparable to Google, Microsoft, and co.

Subscriber to the Premium version of the Tool Box Journal have access to an archive of Premium journals going back to 2007.

You can support the Tool Box Journal and be subscribed to the Premium edition right here.

Thanks!

If you would like to promote this electronic journal by placing a link on your website, I will in turn mention your website in a future edition of the Tool Box Journal. 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.

If you are subscribed to this journal with more than one email address, it would be great if you could unsubscribe redundant addresses through the links Constant Contact offers below.

If you are interested in reprinting one of the articles in this journal for promotional purposes, please contact me for information about pricing.

© 2022 International Writers' Group