TJ Blog
Monday, March 29, 2010
Brave New World of Translation
No one disputes the fact that the translation industry is undergoing fundamental changes, which are affecting what we translate, the way do it, and the compensation we receive for our work.
The basic (interrelated) factors in these changes are
Technology—computers, CAT tools, and the Internet—has dramatically increased our productivity in the past decades, and this trend is expected to continue for the foreseeable future. In a recent conference I attended in Porto Alegre, Brazil, Renato Beninatto, CEO of the consulting firm milengo, predicted that, while the income of translators will remain unchanged or will slightly increase in the coming years, the compensation per translated word will drop substantially. He foresees translators' productivity to rise to 30,000 - 40,000 words a day thanks to CAT tools, which will be free of charge.
Beninatto also foresees that the traditional model of translator-reviser will be replaced by machine translations edited by often monolingual experts.
Since the technologies needed for advanced machine translation can only be afforded by large multinational companies, individual translators and mom-and-pop translation companies will be increasingly marginalized and left with the crumbs of the market and with highly specialized translations such as literature and advertising, which are not susceptible to machine translation. While no one expects Shakespeare to be machine-translated into Quechua in our lifetime, most technical documents (which includes almost all non-literary texts) can already be successfully "gisted" by corpus-based machine translation. In many cases such "gisting" is adequate for the purpose; this is why quality will continue to be trumped by price for all but the most specialized translations where both accuracy and style are critical or where cultural adaptation, rather than straightforward translation, is required.
Do you agree with these predictions? How do you see our industry's future in the next few years and decades? How do you see your own role in this "brave new world of translation"?
The basic (interrelated) factors in these changes are
- Technology,
- Globalization, and
- Concentration of the Industry.
Technology—computers, CAT tools, and the Internet—has dramatically increased our productivity in the past decades, and this trend is expected to continue for the foreseeable future. In a recent conference I attended in Porto Alegre, Brazil, Renato Beninatto, CEO of the consulting firm milengo, predicted that, while the income of translators will remain unchanged or will slightly increase in the coming years, the compensation per translated word will drop substantially. He foresees translators' productivity to rise to 30,000 - 40,000 words a day thanks to CAT tools, which will be free of charge.
Beninatto also foresees that the traditional model of translator-reviser will be replaced by machine translations edited by often monolingual experts.
Since the technologies needed for advanced machine translation can only be afforded by large multinational companies, individual translators and mom-and-pop translation companies will be increasingly marginalized and left with the crumbs of the market and with highly specialized translations such as literature and advertising, which are not susceptible to machine translation. While no one expects Shakespeare to be machine-translated into Quechua in our lifetime, most technical documents (which includes almost all non-literary texts) can already be successfully "gisted" by corpus-based machine translation. In many cases such "gisting" is adequate for the purpose; this is why quality will continue to be trumped by price for all but the most specialized translations where both accuracy and style are critical or where cultural adaptation, rather than straightforward translation, is required.
Do you agree with these predictions? How do you see our industry's future in the next few years and decades? How do you see your own role in this "brave new world of translation"?
posted by Gabe Bokor at 2:45 PM
Table of Contents
Entering the Translation Market
Translation Industry Survey
Communication is Power
Translators and the Recession
Quality
Contracts
Edit or Not to Edit
Specialist or Generalist
Certification
Machine Translation
The Goalie
Liability Insurance for Translators
Translator Licensing
Brave New World of Translation
Globalization
Professor Camayd-Freixas and the Postville, Iowa Raid
Who Pays for Technology?
Test Translations
Globalization
Good Translations from Poor Originals?
Simple comme bonjour
How to Predict Translator Performance
Holidays and Missed Calls
Late Payments and Non-Payments
On-Line Job Mediation Services: Are They Worth It?
The Role of Translation "Agencies"
Native Language
The Internet and the Translation Profession
23 Comments:
I do not agree that that "advanced machine translation can only be afforded by large multinational companies" -- any real intensive use of MT that has a quality focus will require competent human beings to steer and drive quality.
I think it is possible that new kinds of skills will be required in addition to the current skills valued in the TEP model. I think we heading increasingly into a world where global enterprises will need to translate 10X more content than they used to. This is not possible without automation. And as a developer of this technology I can assure you that quality is not possible without continuous, ongoing and sustained human feedback.
I have blogged about this at http://bit.ly/byFEV2 and http://kv-emptypages.blogspot.com/2010/03/why-machine-translation-matters-part-ii.html
Enterprises will seek out companies that help them to get to good enough quality at high productivity levels. MT is another productivity tool not a replacement for humans - computers cannot do this without constant human guidance.
We are just now seeing mostly smaller LSPs beginning to dive in and learn how to use this technology.
I think that many of these LSPs will be tomorrow's leaders.
I do believe that machine translation will change the business, especially for certain types of material, but I do not believe that translators will be rendered jobless. There are still so much a professional translator can do that no machine will ever be able to do.
Instead of blindly accepting the assertion by Mr. Beninatto and other industry lobbyists that high-volume post-edited MT is essential to cope with the explosive growth in content, companies need to become smarter in their purchasing and question whether all this stuff actually needs to be translated.
Perhaps an executive summary would suffice? Ah, but that would require the involvement of a human translator/copywriter who can actually think, rather than simply push buttons.
Remember, this is the same software industry that has convinced practically every small business to buy a fully fledged accounting package to perform simple bookkeeping tasks.
As far as I can see, we're already at the stage where much so-called content wasn't worth writing in the original language, let alone translating.
I agree with both previous comments. I work at Interpreters Unlimited where we provide document translation and interpretation services and in my humble opinion, it is impossible for a machine to do what humans can do when it comes to translation. Although technology will advance allowing for machines to translate better and more accurately, they cannot replace the knowledge a human translator has regarding culture, dialect, word usage, slang, etc. Yes, we have to be realistic to know that machines will get closer and closer, but to phase out human translators. I don't think this could happen.
Tom
I think that what many in the professional industry are missing is this question of value.
What content has the greatest value in terms of building your global business?
Yes, some corporate content must be translated and will continue to be done.
But, in the Web 2.0 world the consensus is increasingly that the customer voice is the one which matters most. Customers discussing products = WOMM Word of Mouth Marketing = User Generated Content = high value content = Facebook, Blogs, Reviews, Twitter. Brands are built and destroyed in the social networks because customers trust other customers more than they do an executive summary from the corporation or "expert". I would argue that this UGC is becoming the most important content and that it is worth the attention of professional translators as they can play an important role together with technology in making it multilingual.
Look at what is happening in global customer support in IT: http://bit.ly/byFEV2
Increasingly this UGC is content that the professional translation industry has little or no role in. Today this is done by Google and Microsoft 10 million times a day. We have already reached a point where machines do 90%+ of the translation done on this planet. Why not learn how to make a boat and flow with the stream?
I think professionals are key to handling this fire hose of new, high value information and that this is both a threat and an opportunity.
Global enterprises will need to digest and communicate in this messy social network dominated world. Professional translators that refuse to learn and use tools like MT, crowdsourcing or web based collaboration and other translation productivity enhancing tools will be marginalized.
I go into this in more detail in my blog if you care at http://kv-emptypages.blogspot.com
However, I do agree that machines will never replace humans for anything that really matters - at least in my lifetime.
1) I think more texts are going to be written with translation in mind, and that greater care will be paid to copywriting, that is writing copy free of grammatical and spelling errors (and that would be a very, very good thing).
2) Machine translation is 'fed' by existing human translations. Language is constantly evolving. Translators will still be needed to stay on top of linguistic changes, if only to update translation memories.
3) Translators' productivity is already on the rise thanks to TMs. We will continue to integrate this kind of help into some of our assignments, but the tool is not going to overcome the craftser.
4) I don't see binding agreements being machine translated and only undergoing proofreading by a monolingual employee any time soon. Large sections of these may undergo machine translation, but they will have to be checked by a translator to make sure they are accurate.
I was asked about this recently; specifically whether translators will go the way of the horse and buggy.
True, there are only a handful of buggies still operating in the modern world. But almost everybody drives, or has access to a vehicle that is driven by someone. And the vehicle manufacturing and service industries employ millions of people.
If the analogy holds up, that means that there will be lots of work available, not only for those who can operate the "translating machinery", but for those who teach languages, and those who design, sell and maintain the "machinery".
Another analogy that I think is useful here is the impact of word-processors on stenographic pools.
Not so long ago documents were sent to a stenographic pool to be typed. IT was expensive and only really important documents got typed up.
As word processors came into being from Wang etc.. it did reduce employment in stenographic pools but also created opportunities for a higher profile office assistant.
With the advent of Office productivity software this role further evolved into executive assistants that could also help make slides, spreadsheets in addition to word processing. The best secretaries were always able to find work because of their value add and new skills.
I think we will see similar kinds of evolution in translation.
Sabine, I don't see any improvement in the quality of the documents that are sent to me for a translation.
Of course, any one of us only sees a tiny fraction of a tip of an iceberg. But most of my documents are technical documents, prepared by one company and read by another.
It is certainly easier these days for a document to be written and released, thanks to word processing and desktop publishing software.
But this has not translated, in my experience, to better written source documents.
I was using the future tense in my comment.
It was a wish for the future, especially based on the fact I had been translating a particularly poorly written source text that day.
Hi!
Technology has changed the way we work and I would like to get your feedback on your experience with the different translation software now available in the market.
http://blog-de-traduccion.trustedtranslations.com/
Best regards,
Amelia
I don't think translators will be replace by machines in the future.
These cannot translate well some of the phrases or expressions only
human beings can understand...
Very interesting blog. I will continue reading it...
Greetings from an argentine teenager, who wants to become a translator in the future.
Nancy
both translator and the machine translator is a good contributions to the industry of communication, though there is nothing to worry about the existence of machine translator, it will not replace human translator, because the two has different functionality and capabilities
English Translator
It's an interesting proposal that we will soon be doing 30k words per day using Translation Memory and Machine translation.
I agree with Kirti that enterprises will need companies that can provide good quality at high productivity levels. Collaboration tools will probably aid that and we have started working on those at our company, TranslateMedia.
There are other approaches though, and as an agency we are increasingly taking a consultative approach to clients with huge needs.
Most recently this involved translating 5.5 million words in just under 8 weeks to coincide with an important launch they were doing with partners. Without Machine Translation.
To do this we provided summary translations according to a brief that we worked on with them.
The translators who worked on this gave us great feedback - with a guide on what was key information they could root through for that and then just write. No need to stick to source text, just write to brief. It was quicker, and therefore way cheaper for the client. Otherwise they never could have afforded it.
Not a machine translation in sight...!!
this blog is wonderful .it help to human with his tools.a world required new tolls to translate like it.
In the twentieth century the figure of the translator has changed from the man of letter interested in traslating foreign texts, into a man commissioned by publishing houses and business organizations.
Quality translation services can be required for literature studies, law enforcement, travelling, medicine and business but sometimes translators are replaced by translator software.
While some organizations still use professional translators to help them translate their documents, many more are turning to translator software, even though they can`t be sure about the quality of translation.
Translation software utilization is a cheap solution for business organizations but professional translations of documents can help business to be more effective and to raise the esteem of translators.
http://www.empowerlingua.co.uk
I agree to a certain degree. As the article says, given the document is heavily technical, it might be feasible for certain language pairs such as English-French, or Korean-Japanese. However, getting a 'gist' by a machine-reliant translation is remotely helpful in any case. After all, how can you be confident? When it comes to language pairs that have totally different language structure such as English-Korean, the argument is almost out of the question. We need the aid of professional translators to do the task.
I definitely agree with Kirti, whatever technical advancement happens in the field of translation, human translators are indispensable in terms of quality and flair.
Technology has changed the way we work and I would like to get your feedback on your experience with the different translation software now available in the market.
Translation Services UK
Wow Nice blog. Thank you for providing the more information about Translation Legal Documents.
There is no urge to translate words, as our language 'English' is growing so vastly. There is really a need of translation to make the language comfortable and more versatile.
I agreed to "it is impossible for a machine to do what humans can do when it comes to translation" but we never neglect that CAT tools help translators to improve consistency, speed and efficiency while doing quality quality French translation or any other language translation. These tools keep a record of the translation done by the translator at the first occurrence then automatically replace these terms/phrases with the translation where ever these occur again in the file.
We can help you with professional translation services from english to arabic and french . please visit:
www.alkhouri-group.com
Post a Comment
<< Home