Extensive CAT tool logging - Big Brother or language technology evaluation panacea?

CAT tools are no longer simple editors that just provide text matching and terminology support. Increasingly, they have become a hub for a range of computational linguistic technologies, including machine translation and automatic speech recognition and various predictive typing technologies.

In my presentation I will describe some of the current machine-translation related use cases for a form of extensive logging of user activity data in CAT tools that we call "instrumentation" based on a Trados Studio plugin and a version of OmegaT. I will focus on real-world productivity tests carried out with two large translation buyers, Welocalize and HP. I will then outline a number of future uses for instrumentation data that extend well beyond machine translation.

Finally, I will discuss how desktop-based CAT tools differ from web-based ones and how valid data privacy concerns from a translator and end-client perspective might be addressed so that some translators may consider it in their interest to volunteer to share logging data from time to time and in a limited way.

John Moran
Owner of Transpiral Ltd.

John Moran is a German to English technical translator, software engineer and publishing researcher specialised in measuring how linguistic technology impacts on translator productivity. In the 1990’s he worked as a lecturer in technical translation in Trinity College Dublin and co-founded Transpiral Ltd., a technical translation agency specialising in German-English based in Dublin, Ireland. He develops software to measure the impact of MT on translator productivity and is a regular conference speaker on the topic of translator productivity.