At the 2018 Annual Korea Technical Communication Association (KTCA) conference, HansemEUG presented a session titled “Terminology Management: The Core of Language Quality,” highlighting the importance of defining and managing terminology in technical writing and localization. This blog post is intended to share our presentation’s key messages for those who couldn’t attend the conference.
For example, let’s take coffee, which is an everyday thing for many people. If you wanted to order the coffee pictured below, what would you choose from the menu?
The menu has a lot of options, such as café mocha, cappuccino, café macchiato, café latte, etc.
The right answer is café latte. It is a coffee drink made with espresso, steamed milk, and milk foam. If you were to misunderstand the menu and order cappuccino, it would have less milk and more foam, so you wouldn’t get the taste you wanted.
Sometimes, we think we know the correct meaning of a term, but confusion happens frequently if the meaning of the term is not clearly defined. The lack of clear definition often leads to confusion.
Defining terminology can be challenging. According to ISO 704:2009, an international standard that describes the basic principles and methods for terminology work, a term
How would you transliterate “낙동강,” the longest river in South Korea, in English? Without a set rule, it could be anything from Nakdonggang, the Nakdong River, or the Nakdonggang River.
Even within a single company, the documents produced by different teams can contain the same terms but have different meanings. Such discrepancies create confusion and may incur future costs. This is why it’s important to standardize the terminology used and create style rules.
Back to the “낙동강” example, if a style rule for unique locale names is defined as “Korean local name + semantic suffix,” then the correct translation would be “the Nakdonggang River” and “한강” would be “the Hangang River.”
Standardizing terminology is challenging, yet important. It plays a crucial role in maintaining the consistency of terminology, therefore raising the quality of the content and brand image – not to mention it reduces the multilingual localization costs.
A term does not always need to be a noun. There are many parts of speech, and all of them can be used to define terms.
After extracting various types of terms from the internal documents, you need to select terms to include in your glossary. It would be great if there was a software program that could automatically select terms by using some complex mathematical algorithm. However, the importance of each term may vary, depending on the field, so the selection needs to be made by a human.
The information that makes up a term is called the structure of termbase. The simplest structure of termbase includes the term definition and the term description. You can add multilingual versions and create a global glossary.
You can also add meta information, such as acronyms, parts of speech, categories, and products. Of course, adding extra information requires extra effort.
Creating a glossary doesn’t stop with storing a word in one place. It needs to be updated on a regular basis. A well-made glossary is the backbone of quality content and translation.
“A term is a concept that defines a product or an object. Through terminology, we can understand terms in the same way as the users.”
A well-made glossary can be made into a translation memory; and though this translation memory, the content of user manuals can be reused and localized.
A well-made glossary is useless if it is not used well. It should be used efficiently, not only in the cloud terminology management system but also with the CAT and authoring tools. That way, its value can be fully realized.
No QA tools on the market can always ensure the level of the quality required by customers. Tools detect many errors, but if most of the errors detected are false positives, then it requires more time and effort to determine which ones are the real errors. This actually happens pretty frequently.
Some of the well-known QA tools include QA Checker, Xbench, QA Distiller, ErrorSpy, and Verifika; each one of them has its pros and cons. QA checks can be divide into mechanical QA and linguistic QA. Mechanical QA includes checking missing tags, double spaces, missing punctuation, mismatching number of sentences and units, etc. Linguistic QA includes spellchecking, capitalization, and missing terms. Every project has a different quality goal. Since no QA tool is perfect, it is necessary to set appropriate quality goals with the client.
Linguistic quality can be maintained by checking for the following:
HansemEUG meets the highest standards of quality with its own QA program to meet our clients’ quality expectations.
We looked at quality-related issues by type as follows:
The way to solve quality problems is not difficult. First, make sure that style guides and glossaries are updated regularly, and then go through the QA and review process. This is the most effective method. QA activities cost the service provider time and money, but the results are that the customer receives high-quality work. So it’s necessary for the customer to appropriately reward the service provider for its QA activities.
“Terminology management is the core of language quality.”