Update October 9th 2019: This blog refers to adaptive/statistical machine translation, which has now been replaced by SDL’s powerful neural machine translation (NMT) solution. To learn more about NMT please click here.
Striving to deliver the highest quality in any industry is always a top objective of any successful organization. In the translation industry, it’s no exception with quality probably one of the most discussed subject areas among translation professionals. Today there are still many debates on how to define what quality means for different stakeholders, there are various standards and quality models available but it can also be a very emotive subject.
In our 2016 Translation Technology Insights (TTI) research, one of the largest industry technology surveys ever conducted, quality was king; considered six times more important than cost and two and a half time more important than speed. Interestingly though, quality was not seen as always easy to get right first time.
When it came to the quality of a translation not being acceptable, terminology was cited as being the number one reason to rework translations. This is where we see the terminology and quality conundrum. What is clear is the pivotal role that terminology plays in helping to improve quality. Providing consistent use of terms and brand across multiple translation projects not only helps improve quality but also increases productivity.
With such an obvious issue to address it was surprising to find that only just over half (56%) of those in our study say that terminology management is a priority.
Making the most of terminology sharing
Why are many translation professionals not prioritizing terminology? The answer could be as simple as a lack of awareness of terminology management tools available. Another explanation is that 55% of translators see terminology tools as seeming too complicated or time-consuming.
It’s not just about what the terminology tools can do, but also how people can use them. Essential to terminology management is the ability to share termbases when required and easily. Something which isn’t being done – only 51% of respondents in the latest SDL survey ‘We need to talk’ use collaboration tools to share their translation assets.
Introducing SDL LC Terminology (Beta)
Clearly there is a need for translators to get started with a terminology tool that is simple to use, easy to share with others and directly integrated into a CAT tool.
Our Product Management team have been busy developing the latest innovation in terminology management, and with the release of SDL Trados Studio 2017 SR1, we are delighted to announce a technical preview of an online, shareable and easy to use terminology management offering, SDL LC Terminology (Beta).
“When building a new product generation and ‘reinventing’ how to address a particular space such as terminology management, product management is always keen to consider the feedback from customers over the years. This is vital to ensure that we can address any gaps that customers identified with existing solutions from day one." – Daniel Brockmann, Product Management Director at SDL.
SDL LC Terminology (Beta) has reinvented working with terminology in three ways:
1. Customer issue – Terminology sharing is difficult for translators without an on-premise server solution. Workarounds include sharing termbases over email or using potentially fragile cloud storage mechanisms, which results in terminology not being centralized or robust and therefore increasing translation quality and turn around issues.
Solution – With SDL LC Terminology (Beta), termbases can now be created online, in the browser, and shared directly with others for collaborative and concurrent use. So what does this mean in practice? Termbases are updated in real time to any terminology changes and everyone working on the same termbase will see the same terminology, even when term amendments are made. The result is consistent terminology use within your translation team and overall improved translation quality.
2. Customer issue – Many translators store their terminology resources in excel spreadsheets, however, native Excel import and export is not supported by existing terminology tools. A translator would need the Glossary Converter app to create termbases or other mechanisms to import and convert terminology from Excel files.
Solution – SDL LC Terminology (Beta) supports direct import/ export functionality for .tbx, .xml, .xlsx, .csv. There is no need to go through the extra step of converting spreadsheets as they can now be directly imported. You can easily map Excel columns to termbase fields in the process to ensure that all relevant data is imported into the online termbase.
3. Customer issue – Traditional character-based fuzzy terminology search has often not been regarded as optimal for languages that expose a certain complexity in how terms can deviate from their base form (such as in Eastern European languages or Finnish, for example). This can then also lead to false positives during terminology verification when fuzzy matching doesn’t find terminology properly.
Solution – When developing the new terminology stack for SDL LC Terminology (Beta), we were keen to build a powerful, sophisticated and error tolerant search. The new linguistic fuzzy search knows how to trace a word back to its root form for each language. What’s more, we are not taking anything away from users who prefer the existing fuzzy search – users have the flexibility to choose between both the ‘classic’ and the new search methods.
Watch the video below to learn how to easily create, manage and share online termbases and how this directly integrates into SDL Trados Studio 2017.
Terminology sharing within SDL LC is core to SDL’s aim of a ‘Best Match Service’ for translators, and we see it as another step towards this vision. The best match principle is where your key translation resources (machine translation, termbases, and translation memories) are readily available in the cloud; yet at the same time seamlessly integrated into SDL Trados Studio. These resources all ‘talk’ to one another, agreeing among themselves the best match to give the translator as a starting point. Through feedback from the translator, they continue to learn, so that ultimately they can jointly deliver results that are greater than the sum of the parts.
Any other information I need to know?
- The SDL LC Terminology (beta) is currently running
- The size of each termbase is unlimited
- There is a limit of five termbases per LC account
- You need an SDL Trados Studio License and LC account to use this new SDL technology preview
- Initial functionality is focused on the translator. More sophisticated functionality, such as creating your own custom termbase definition, will come in the course of 2018.
What’s stopping you? Try it out for free and share any feedback you have in the SDL Community.