Does a typographical error affect the quality of your translations?

In order to answer that question, we need to know the definition of “What’s a typographical error?”

During my experience doing mechanical QA I found out that some typographical errors are often mistakes that occur during the process of copying and pasting the final translation into a template, like an English template in Quark or InDesign.

As you paste the translation into the English template, you may leave behind some English words. In the case of Roman languages it’s more difficult to catch this error, while in languages like Korean, Chinese, Arabic or Hebrew the error is easier to identify.

See examples below:

Example #1: An easy case of identifying a typographical error

In English: Storage and Handling: Control must be stored upright, tightly sealed, at Room Temperature (20-25°C).

Incorrect Korean: S보관 및 취급: 대조 용액은 실온(20-25°C)에서 똑바로 세운 상태로 밀봉하여 보관해야 합니다.

“S” is a typographical error.

In English: Do not use if the package is open or damaged.

Incorrect Portuguese: Do Nunca utilizar ar ou qualquer meio gasoso para encher o balão.

“Do” is a typographical error.

How typographical errors affect the quality of the translation may differ based on the kind of typographical error.

While typographical errors represent poor quality of your translated document it may not affect the meaning of the translation; however, need to avoid having typographical errors in the translated documents.

Human qualities for machine translation – will the dream come true?

The use of machine translation (MT) has become essential in these times, except obviously in literary translation where euphemisms, idioms and metaphors remain in the human translator’s domain. If there is nothing that can replace human understanding, then there is nothing that can replace the machine in terms of productivity. In the translation industry, work volume is constantly increasing, but the number of professional translators remains stable. Productivity is now a crucial key factor. It not only caused the standardization of the whole translation process, but it also led to automation of translation itself.

Let’s go through the main differences between computer-assisted translation (CAT) and MT. The first, CAT, consists of specialized tools dedicated to professional translators that make their work easier mainly by means of a database (TM). The second, MT, is essentially based on the use of the machine. Its existence dates back to the 50’s and its functionality has been continually improved since then. Who hasn’t tried it on a professional or personal level? Online free tools like Systran, Google Translate, Babel Fish, Reverso, and Prompt are not tailored for professional purposes, but they do help to decipher a message in a foreign language or to understand its basic meaning. In any case, they cannot convey any linguistic subtleties: puns, idiomatic expressions, synonyms, etc. Aspects like style, fluency, writing quality are often left out.

However, companies specializing in machine translation look to broaden their target users to multinational firms and translation professionals such as agencies and freelancers. These professionals use programs and server software specially designed for business use, like Systran and its new hybrid machine translation engine or Translator Toolkit for Google Translate. These new tools, which are generally not free, are intended for professionals and are now able to provide an overall satisfactory translation quality.

In order to stay in tune with those companies’ politics and to answer to the needs of some of the biggest multinational firms, SDL, the world leader in Global Information Management specialized in CAT solutions, decided to take a big step in terms of machine translation – it acquired in mid July Language Weaver, a pioneer in statistical machine translation. SDL’s baby, SDL Trados Studio 2009 SP2, combines translation memories with machine translation; in other words, human professional translation and machine translation in order to handle more volume in less time and with a higher quality.

What about the role of translators? Is their future at stake? Will machine translation be the end of freelance translators?

According to experts and market analysts, the translation industry still has a bright future ahead: contents are constantly increasing (they estimate twice the volume for forthcoming years), particularly for online contents, in particular because Internet users search for contents in their native language. SDL forecasts that customers will be able to reduce their costs by about 30% - 50%, accelerating the contents’ time-to-market.

Despite that, translators can be reassured. Machine translation represents only 1% of the whole translation market nowadays.

All this leads to believe that we are on the right track: machine translation is now the professional translators’ best ally as it will allows for combining speed, volume and quality.