SAP’s terminology blog

Mark Childress is a member of the Terminology@SAP team responsible for SAP’s terminology management and for the SAPterm terminology database, and I recently found his blog. I think you will like his posts on practical cases on how they do terminology work.

He started his blog a few years ago and left it inactive, but last September 2017 he restarted writing. These are some of the titles of his posts:

  • Is it a term? “It’s complicated”
  • Getting down to the nitty-gritty terms
  • The cost of one term – how to figure it out for yourself

Click here to visit.

Connect with him in LinkedIn.

I have added it to my page of terminology blogs.

Localization Terminology Management – Top Tips

I am always happy to have guest posts by companies who are fully engaged in terminology management. Karl Pfeiffer works as Senior Language Lead at Argos Multilingual (see bio below) and has written this post on terminology and localization. I must confess I have rarely referred to localization in this blog, although it is one of the fields that manage terminology the most. So, I couldn’t pass up the opportunity to have them publish a post here. Big thanks to Argos Multilingual and Karl Pfeiffer for this opportunity to share their terminology love!

Businesses with plans to grow globally need to pay attention to terminology management. Consistent terminology helps to enable high quality translations, saves costs and time, and helps to build your brand worldwide. This article provides insights into Terminology Management in a localization project.

Why should you be bothered?

Easy: Your business needs to build and protect its brand and corporate image – whichever markets you choose. By establishing business-specific terminology, you avoid misrepresentation of your brand caused by mistakes in the translation of industry and product-critical words and phrases. By having a consistent global message, you can ensure the accuracy of your translated content and your communications will hit the mark consistently, globally!

What’s more, a precise and unambiguous terminology database (termbase) made for your business will help to save both time and costs during the translation process. Linguists work more efficiently when they can refer to existing approved terms and phrases, because they spend less time researching vocabulary and can instead focus on translating. When everyone working in a language uses the same termbase, inconsistencies can be avoided and a high-quality translation can be assured. Read More

Six document search engines you should use

The following is a list of document search engines that you can add to Google Scholar and Google Books and that have allowed me to discover interesting documentation.

  1. Academic Index

Its creator, Dr. Michael Bell, explains “As a meta-search engine, the Academic Index integrates into its search results only the first 1-2 pages returned from each site it searches. Because most sites rank search results as to relevance, this ensures that only the best (most relevant) information is returned to users.” [2]

  1. Base

The Bielefeld Academic Search Engine searches for academic web resources: journals, institutional repositories, digital collections etc.

  1. Directory of open-access journals (DOAJ)

The Directory of open-access journals gathers documentation on science, technology, medicine, social science and humanities (approximately 10.000 journals). The aim of the DOAJ is to increase the visibility and ease of use of the journals to promote their use and impact.

  1. DocHound

DocHound is the EU Interinstitutional Document Search tool by the Terminology Coordination Unit (TermCoord) of the European Parliament and it updates its content regularly, so you are sure to get up-to-date documentation. You will find basic documents, legislative drafting, procedures, documents from the EP and other institutions and bodies.

  1. CORE (COnnecting REpositoires)

CORE gathers content from repositories and journals around the world. CORE harvests all metadata records in a repository. For now, they only offer PDF files but hope to expand the service to include HTML, webpages, etc.

  1. RefSeek

This great site is like the Google for academics, science, and research. It strips results to show pages such as .edu or .org and includes more than 1 billion publications, such as web pages, books, encyclopedias, journals, and newspapers. In a test done by IT journalist, Stan Schroeder, when he searched for “flower”, RefSeek showed him documents from botany (as compared to Google that returned a list of florists!) [1]

For a comprehensive list and by topic, I recommend checking these pages.

  1. Top 11 Trusted (And Free) Search Engines for Scientific and Academic Research
  2. 100 Time-Saving Search Engines for Serious Scholars (Revised)

Share your favorite engine in the comments or send me a note to add it here.


[1]          Schroeder, Stan. RefSeek is Google for Students and Scientists. 2008 [consulted on 2/1/2018].

[2]          Bell, Michael. Academic Index. 2003 [consulted on 2/1/2018].


Terminology in Translation Quality Assurance

Terminology is without a doubt one of the key components in a translation quality assurance (QA) process. How we deal with terminology from the beginning is a make or break situation. In a study, Sharon O’Brian reported that “ten out of eleven models specifically refer to terminology as an error category”.

Indeed, terminology is usually one of the major issues when dealing with translation errors. Arango-Keeth and Koby (2003, 119) indicated that one of the problems associated with translation evaluation was indeed terminology or, in their words, “lack of a standardized terminology”.

The first requirement is to use the terminology standards available. In 2006, UNESCO created the “Guidelines for Designing Terminology Polices”, then the European Association for Terminology (EAFT) held a seminar on Minority Languages and Terminology Policies, and the year 2010 saw the birth of new ISO standard to guide the development and implementation of terminology policies: (ISO 29383:2010), later revised in 2016. (see links below).

Also, if you are a freelancer, you may also want to take some important factors to take into consideration. These might seem obvious, but there is a reason why terminology is still a major issue. Read More

Sign up for TermNet’s Terminology Management certification

You still have time to sign up for one of TermNet’s Terminology Management training courses.

Sign up for the Basic course by clicking here.

Sign up for the Advanced course by clicking here.

Sign up for the Engineering course by clicking here.

Read my blog post “Should I sign up” in which you will find useful information about these training courses.

Happy learning!

Terminology webinars and study cases by SDL Trados

You probably remember the webinars on terminology that SDL Trados has organized with renowned terminologists. Well, it turns out there is a dedicated page to look for specific resources in SDL’s webpage, so I did a specific search just for terminology and found other interesting material.

Click here to go the page.

Click here to go to their Translation Software Resource Hub.

Also, you might want to check out my summaries of some of these webinars in my tag cloud VIDEOS. I hope SDL Trados keeps organizing this type of terminology training, so share this post to let them know that we want more!

Database of Cross-Linguistic Colexifications (CLICS)

I recently found this database that I was unaware of. Here is a short extract copied from their site. You can also play with this tool here:

“CLICS is an online database of synchronic lexical associations (“colexifications”, see here for more information) in currently 221 language varieties of the world. Large databases offering lexical information on the world’s languages are already readily available for research in different online sources. However, the information on tendencies of meaning associations they enshrine is not easily extractable from these sources themselves. This is why CLICS was created. It is designed to serve as a data source for work in lexical typology, diachronic semantics, and research in cognitive science that focuses on natural language semantics from the viewpoint of cross-linguistic diversity. Furthermore, CLICS can be used as a helpful tool to assess the plausibility of semantic connections between possible cognates in the establishment of genetic relations between languages.”

To generate a concept map, I picked the concept “tree” and ran a graph , this was the result:

Happy searching!

Compiling terminology in a new field of knowledge

A couple of years ago, I wrote a post titled “Tackling Terminology in a New Field” and while reviewing COTSOES’ Recommendations for Terminology Work, I thought it would be interesting to summarize their recommendations related to the compilation of terminology, particularly when you know that you will be doing a lot of translation in that new field. You will find the details on page 62 (section 5.6) of COTSOE’s document.

Although COTSOES’ recommendations are for large organizations, I think this could easily be transferred to freelancers.

Collecting documentation: Your primary source will be the person who is requesting the job. He/she will probably have texts and documentation in which you can also find other specialized documentation (journals, reports, etc.) This might take up to six months.

Interest specialized bodies: You can identify key players (organizations, individuals) from the documentation collected and they can point to other reliable documentation and, possibly, be interested in collaborating with you. Read More

Playing with language in a virtual world: The MUSE Project

I was recently taking a look at this project by the University of Leuven in Belgium called the Machine Understanding for Interactive Storytelling (MUSE) project, in which written text is converted into 3D virtual worlds, “a way of bringing text to life”. The research team has created this tool to teach computers how to understand human language. So far, they have been testing children’s stories and medical applications which the machine translates into images. MUSE’s project could give rise to very promising applications. For example, they would enable people to understand complex instructions or medical treatments.

You can read more in the sources below, but you can start “playing” with this technology and create a 3D demo. Very easy to use here:

Sources and further reading:

  1. MUSE official page.
  2. Would you like to draw by just using words? This project is part of the FETFX, or FET research. You can take a look at their other AI projects by clicking here.
  3. Watch this presentation recorded during the 2016 Translating Europe Forum given by Marie Francie Moens about Deep Learning for Machine Translation and Machine Understanding. (start at 3 h, 15 min to skip other presentations).

Have you used DeepL, Linguee’s offspring?

I think we have all used Linguee, haven’t we? Well, the company changed its name last year to DeepL. Linguee is one of the most widely used computer translation engines: 1 billion users have made 10 billion consultations. In 2016, based on all the knowledge gathered through Linguee, the developers started working on a neural network system by training their neural networks with billions of translations gathered by the Linguee crawlers. In less than 10 years (Linguee was launched in 2009) the company has taken great strides into machine translation using deep learning and neural machine translation.

DeepL has claimed that it is the “most accurate and natural-sounding machine translation tool based mainly on the assessments they have made using the BLEU method (Bilingual Evaluation Understudy), which compares a human translation against machine translation measured on a 1 to 10 scale. It is indeed promising, but still a lot of work needs to be done (terminology is still a big challenge) and some experts agree that DeepL’s claim is a bit far-fetched. For example, Kirti Vashee, a technology consultant, commented: “I think they are using well understood and public test sets and claiming that they have better results. To my view they are slightly better, but far from revolutionary.” Read more about this in this post by Slator.

As a final note, I thought it is interesting to mention that DeepL’s supercomputer is based in Iceland, to take full advantage of that country’s renewable energy which makes operations cheaper. This supercomputer can translate 1 million words in less than a second. One thing is for sure, this company will probably give us a lot of good products, even beyond translation, as they say that their “neural networks have developed a level of text understanding that opens several exciting possibilities”.

So, if you haven’t given it a try, here is the link to DeepL:, where, by the way, you will find a tab for Linguee too. 2 in 1!


Arthrex looking for terminology intern (Naples, Florida)

Language and terminology lovers take note! I just read this on Uwe Muegge’s Facebook page.

“Any aspiring linguist interested in working as a Terminology Intern as part of my team here at Arthrex (one of Forbes 100 best companies to work for) in sunny Naples, Florida? This is a paid, full-time position where you will be working with the absolute latest and greatest in enterprise language management tools and processes, and you’ll be learning from some of the best in the industry. Oh, and did I mention free lunches and our free on-site health clinic?” Please apply directly online:

Good luck!

Magic Search tool for terminology searches

Thanks to Spiros Doikas, Translatum Founder, Greek translator, and CAT tool trainer for sharing Magic Tool. I copy below part of his presentation about the tool, but visit this page to read the details.

It has been 4 years since Magic Search was launched, providing one-page search results of multiple dictionaries for a limited number of language pairs (up to 28).

The code of Magic Search has been refactored and massively enhanced to support a number of new features and well over 10,000 language pairs. Now it has its own dedicated domain—

All you need to do is select your favourite language pairsubmit a search, and you will get search results from a number of different sources (dictionaries, corpora, MT engines).

You can either have one page results ON or OFFON will allow you to scroll through the sources but it will be a bit jumpy until all dictionaries load, whereas if you select OFF you will have to click on each dictionary button in order to display it.

The next time you visit, the site will remember your choice of language pair. But this is not all. If you click on the gear icon, you can reorder as well as exclude/include the available sources. You can even add monolingual sources (excluded by default) in a bilingual search. For example, if you search from English to Greek, by clicking on the gear icon you can add English dictionaries from the Monolingual section of the Excluded column.

To make it easier for you, a dedicated Chrome extension has been developed (use ALT+double click to look up a word); language-pair-specific browser search buttons that work in most browsers (select a word and click the button to search); a language-pair-specific Word macro (which will lookup any word if the insertion point is within its boundaries); as well as a search form which you can embed in your site.