For the June 2013 issue of Multilingual Magazine, I spoke with PayPal’s Rubén Rodríguez de la Fuente about why companies like PayPal are taking an open approach to machine translation technology. PayPal is just one of the companies that is technology-agnostic when it comes to machine translation (MT). This approach is also shared by some of the most experienced MT practitioners in the industry, including Autodesk, Symantec, Adobe… and, of course, LexWorks!
Companies that are technology agnostic are not tied to a single engine or approach. They obtain high quality MT output by selecting the machine translation engine – Rules-Based, Statistical or Hybrid – that performs best with a particular language pair and with a particular type of content. In our experience, putting the right engine to work on the right project significantly improves the quality of the output. Whether the project in this context is all German documentation or a 30-language customer support site, it helps to have the deep expertise to benchmark a range of engines to make sure you are using the one that delivers superior performance.
According to PayPal’s Rodríguez de la Fuente:
“A good MT strategy should be technology-agnostic and look for the most efficient solution on a case-by-case basis.”
There are a number of good reasons for choosing machine translation. These include to reduce costs, increase translation velocity, decrease rework, improve consistency, standardize terminology, accelerate time to market and so on. All are feasible goals. However, success in meeting any of these goals depends on choosing best-of-breed solutions. That’s why it’s important to be technology agnostic.
At the end of the day, there is some nuance involved in getting quality MT. But most of all it requires an open mind.