Exploring the Future of Post-Editing: Insights from Marijana Mikić, Translator and Workshop Facilitator

Ciklopea 9 months ago 5 min read

Recently, the Department of German Studies in Zadar hosted two workshops, organized in collaboration with Ciklopea. Our colleague Marijana was in charge of them for the third consecutive year.

Learn more about her insights regarding post-editing below.

Exploring the Future of Post-Editing: Insights from Marijana Mikić, Translator and Workshop Facilitator

I prepared a short presentation on machine translation post-editing, followed by practical work on three different texts that were translated by a machine: an operating manual, an EU communication, and the webpage of a tourist agency, both from and into German. So, the participants had to edit the machine translation and eventually give their feedback. I wanted to show them what the challenges of MTPE are and how to solve them. It also helped them learn with what language combinations and what types of texts post-editing does (or doesn’t) work with.

  • How do you see the role of post-editing evolving in the translation industry, especially in the context of advancements in machine translation technology?

After having taken part in the TEF conference in Brussels last November and having heard the experiences of other companies and freelance translators, I believe post-editing has already become an integral part of translation processes in other countries such as Spain, while here in Croatia it is still something new and unwanted, much like the ugly duckling. We will have to adapt and embrace whatever advantages machine translation has to offer and work around the pitfalls of post-editing in order to end up with the best possible translation product. People have always wanted to work with machines and now it has finally become a reality, and I hope that instead of being intimidated by MTPE we will learn how, where, and when to make the best use of it. It will never replace human translators and the experience and context that we bring to the table, especially in the creatively demanding fields, but why not use it as a tool, just like we use CAT tools? Post-editors are what translators and proofreaders used to be, and with MTPE they are taking it to a higher level.

  • Did you notice any challenges or questions that came up frequently during the workshops?

The students are mainly concerned about the rates for post-editing, how much and how is one paid for their post-editing services, having in mind the worries produced by AI and machine translation, and the idea that translators as such are no longer needed.

  • In your experience, what are the main benefits of post-editing, and are there specific types of content or projects where post-editing is particularly effective?

Ideally, when working with good MT, if the text is relatively straightforward with as little word plays as possible, post-editing saves time (and consequently money). Sometimes MT offers words and synonyms you would maybe not think of. Depending on the type of text and audience/readers, you put in as much effort as you think is required. It is particularly effective when translating the textual parts of manuals, instructions for use, privacy policies, and we have seen that when the MT is really good, it performs extremely well even with more contextually demanding types of texts, such as EU communications, directives, regulations, etc.

ciklopea-translation

  • In your opinion, how could the information from the workshops be helpful for those who participated, especially those studying German?

I hope I helped them realize that post-editing is indeed our future, or rather our present. They will not be able to escape it should they choose to pursue a career in translation. But in order to be a good post-editor, they first need to be a good translator and a good proofreader; they must be fluent in both source and target languages, and they must know how to work with various different CAT tools. But it is manageable and achievable. And naturally, they should like/love/enjoy their job.

  • Have you noticed any challenges or concerns among translators regarding post-editing? How can these challenges be addressed or mitigated?

Translators are mainly worried about not being paid fairly for their work, and rightly so. No one enjoys cleaning up other people’s mess, let alone after a machine translation. It is still difficult to separate pure translation from post-editing because it sometimes happens that instead of editing you need to translate a sentence/segment from the beginning and in that case, MT is nothing but a nuisance. However, these situations tend to appear less and less, so be patient.

  • As a translator, how do you balance the use of machine translation and post-editing with maintaining linguistic and cultural nuances in your work?

I usually approach any text that I proofread/edit in the same way. I look at the whole sentence, detect what is wrong or unusual on the grammatical and syntactical level and work my way through to the semantics and check whether it is consistent with the rest of the text, focusing on terminology as well as style. I ask myself how a human would express an idea in the target language. Very often you need to do some research to find the right collocation.

If it works as it is, I tend to edit as little as possible, and if the linguistic and cultural nuances are off, sometimes I change the MT output completely or translate the whole source sentence from scratch.

  • Do you believe that post-editing will continue to be an essential skill for translators in the future? How can translators adapt to the changing landscape of machine-assisted translation?

Most certainly. The simplest way to adapt is to be up to date, follow the trends and what is going on in the industry, and to be willing to adapt.

  • Looking ahead, do you have any advice for fellow translators on how to embrace the evolving landscape of machine-assisted translation and post-editing in their careers?

Marijana Mikic ciklopea

I would say: just dive in! You can’t know if you’re any good with it or if MT is any good if you don’t try. Clients expect and will increasingly expect machine translation post-editing. Whether we like it or not, there is no reversing that. Use your experience to your advantage, and if you don’t have any experience, start gaining it.

Also, hopefully universities will start including MTPE in their curricula soon so that the demands of the market can be met and in a more high-quality manner, by providing know-how for post-editing experts.

Like This Article? Subscribe to Receive More Via Email

  • receive a digest with new articles
  • up to 2 emails a month

Related Articles

Join Our Webinar: How High-Quality Translation Drives Success in the Automotive Market

3 weeks ago

At Ciklopea, we understand that precision and accuracy are vital in the automotive industry. From safety to maintenance and compliance with regulatory standards, translations play a critical role in ensuring successful global operations. That’s why we’re excited to invite you to our upcoming webinar, Fast and Flawless: How High-Quality Translation Drives Success in the Automotive Market.

Continue reading

Maximizing Translation Efficiency with AI-Powered TM Cleanup: The Future of Localization

4 weeks ago

In the fast-paced world of localization, maintaining the accuracy and quality of Translation Memories (TMs) is essential for ensuring seamless, consistent translations. As companies scale globally, the size of their TMs grows exponentially, which can lead to inconsistencies, outdated translations, and even redundant content. That’s where AI-powered TM Cleanup comes into play. This innovative solution combines the strengths of AI with prompt engineering, revolutionizing how organizations manage their TMs.

Continue reading

Understanding Linguistic Annotation: Enhancing Language Data for AI and ML

1 month ago

In the evolving landscape of artificial intelligence (AI) and machine learning (ML), the role of linguistic annotation has become more important than ever. As companies and developers work to build smarter systems, from chatbots to translation tools, the need for high-quality language data is paramount. But what exactly is linguistic annotation, and why is it so important?

Continue reading