Technical glossary of translation terms

Automatic or machine translation (MT) refers to the translation of text by a computer program. Direct machine translation is the oldest and simplest MT method: Here, the words of the source text are translated word by word, and then the sentence order and inflection are adapted according to the rules of the target language.

Then there is the transfer method, in which the program analyses the grammatical structures in the source text and then generates text in the target language based on the language’s rules.

Despite the hype surrounding machine translation, unedited results from an MT should only be used in well-founded, isolated cases. Ideally, the text should be reviewed by a (human) post-editor.

See also: CAT tool, Machine translation post-editing (MTPE)

Post-editing (PE) is an Essential Part of Machine Translation (MT). Post-editing is when a human translator reviews, corrects and improves a machine-generated translation.

It’s an essential process because, while machine translations are fast, they’re usually lacking in quality, accuracy, conviction and style.

Learn more about: Post-editing

See also: Machine translation (MT)

When translating a piece of text, any earlier translations written for the same client are automatically suggested to a translator by the CAT tool. This ensures that a company’s specific terms are always used consistently. Even when different translators are involved. Plus, repetitions and matches generally cost less for the customer, as they require less manual intervention by the translator.

A 100% match is when exactly the same segment has already occurred in a previous text.

A context match is when a 100% match where the context (i.e. surrounding content) is also the same.

A fuzzy match is when segments don’t exactly match a previously translated segment. Fuzzy matches are automatically assigned a match rate, expressed as percentages, depending on the degree of similarity.

See also: Analysis, CAT tool, Fuzzy matches

Meta descriptions are short sentences that summarise the content of HTML documents for search engines. For this to work, they are stored in the upper area of an HTML document. In the browser search, the meta description is displayed together with the title tag. A good meta description has a positive effect on the search engine ranking and attracts users to the website.

See also: HTML, Search Engine Optimisation

Metadata is structured data that describes other data such as tables, columns, keys and indexes. It provides a means of indexing, accessing, assessing and discovering digital resources. Metadata uses comprehensive information in order to make it machine-readable and so it can be evaluated.

In the case of a book, for example, metadata is the name of the author, the publication date, the publisher and the ISBN. In the case of a technical drawing, metadata would include the name of the designer, the component number or the version number.

In HTML website programming, metadata is used to deliver information to servers, browsers and automated programs (such as robots, spiders and crawlers). That’s how search engines receive information about the page content. Users see the meta description when they search. If no meta description is given, the search engine displays text from the main body of your page instead.

See also: HTML