Ever since machine translation has come to the fore in the translation sphere, another “new” area of activity has also started to show up: post-editing.
While I’d heard about this for a while now, I only had a vague idea of what it actually entails. So when my regional chapter of the BDÜ announced an evening of discussion on this topic, I made sure I was there.
Two colleagues who had attended a seminar on post-editing first gave a brief overview of machine translation and the different approaches to it and then explained the difference between “light” and “full” post-editing, the former involving only the most minimal amount of intervention, basically making sure that the text is understandable, and the latter taking care of correct spelling, grammar and punctuation, as well as ensuring a uniform style and the feel that the translation was done by an actual human being.
The way I understand it, full post-editing is a lot like proofreading, only you have to be faster, as an average of 1000 to 1500 words per hour are the expected output. But at the same time, doing this for more than four hours a day is apparently pretty much impossible.
It also seems to largely depend on one’s area of expertise
if any inquiries for post-editing are coming in at all – medicine, for
example, is an area where MT and therefore the need for post-editing is
virtually non-existent.
The consensus in the group was that post-editing – particularly the light version – is only something for a certain type of person, someone who has no problem with NOT imprinting one’s own style onto the text and sometimes letting “good enough” be. Someone called it “not looking for errors”, and that is indeed not everyone’s cup of tea.
I left the meeting with quite a bit of new information as well as the notion that perhaps I should give post-editing a try. Being also a teacher, I am constantly correcting and improving both written and oral translations, so maybe this really is something for me.
I’ll keep you posted if and when I have some hands-on experience.
