GIGO (Garbage In, Garbage Out) As with software, if the data fed into the computer program (read source text) is bad, the program's results (read translation) will also be bad. I thought of this when reading ultan's recent post "Information Quality, MT and UX" on Multilingual Computing's Blogos blog. ultan notes that quality information not only makes machine translation easier, but simply is better information that is more easily understood by both humans and machines. So what is quality information? I think quality information consistent and concise, but well-written text with an audience-appropriate level of technical terminology.
In this context, well-written refers to grammatically correct, clear structures free of spelling and punctuation errors. Clearly the amount and complexity of subject-specific terminology used depends on the text's end users. However, such poorly written source text not only hampers the flow of reading, it often also adds ambiguity to the text. The Coming Disintermediation and Disruption in the Translation Industry.
I spent a few days in a very hot and dry Las Vegas last week at the IMTT Vendor Management Seminar, which I would wholeheartedly recommend to anybody who wants to understand how to better manage the relationship with translators to mutual benefit and also get a sense for changing production models in the industry. IMTT does a better job than anybody else I know of including translator perspectives into broader LSP and enterprise level localization discussions. They do this in a smaller and more engaged and interactive format than larger events. There were many great presentations on the nuts and bolts of vendor management, but two high level industry presentations stood out for me.
In economics, disintermediation is the removal of intermediaries in a supply chain: "cutting out the middleman". One point that RB made was that real innovation usually comes from outside the industry e.g. “Those who do not learn from history are doomed to repeat it” What is Holding the Wider Adoption of Machine Translation (MT) back?
Facebook: From 1 to 100 languages in two years | Global by Desig. Social Translation : Using the WWL API To Build Multilingual Sit. Language is one of the few remaining barriers on the Internet. The web has rendered time and distance largely irrelevant, but much of it remains fragmented by language. The Worldwide Lexicon, an open source project I have worked on for some time, aims to lower those barriers by combining people and computers to participate in translating and curating web content.
We're releasing a new web services API, hosted on Google's cloud computing platform, that makes it easy to embed social translation features in virtually any website or web app. This tutorial explains how to build a variety of translation tools using straightforward web services development techniques. If you're interested in a review of Google's App Engine platform, see my companion post, Why I Started Coding Again (Thanks Guido!). The Worldwide Lexicon is what's known as a translation memory.
A Polyglot WSAPI Sending data in to the Worldwide Lexicon API is simple. A Typical Interaction That's just one type of query. Or. Wait till I come! » Blog Archive » Translating or localising doc. Friday, December 19th, 2008 at 6:02 pm We just had an interesting meeting here discussing plans of how to provide translations of our documentations in different languages. I am a big fan of documentation and have given several presentations talking about the why and how of good docs. In essence the “good code explains itself” talk is a big fat arrogant lie. You should never expect the people using your code to be on the same level as you – that would defeat the purpose of writing code.
Fact is that far too much software has no proper documentation at all let alone translations into different languages. Price – translation is expensive, which is why companies started crowdsourcing it. All of these are things to consider and not that easy to get around. Local experts – this could be someone working for you or a group of volunteers that see the value of your product for their markets. What do you think? NLA Machine translation (MT) Urdu software. NLA Machine translation (MT) Urdu software NLA launches Urdu software to break language barriers Islamabad, July 19: English is no more a barrier in learning computer and internet technology as National Language Authority (NLA) on Saturday released and Urdu software carrying computing and processing systems, developed by its Centre of Excellence for Urdu Informatics.
Machine translation (MT) software is a big step and now Urdu Data House is ready to replace other National languages on cyber space, NLA Chairman Iftikhar Arif announced in a ceremony. "Microsoft applications like Windows, Office are now converted into Urdu operating systems. One font and one keyboard for Urdu and other Pakistani languages is also released by NLA," Arif said.
Arif said in a rapidly growing integrated technology, it was inevitable to make Urdu language parallel to developed contemporary languages in usage of computer, internet and informatics. Post your comments. MT on the Slope of Enlightenment? | Technology. Evaluating Machine Translation:The Present and Future of Multili.
A recent study conducted by researchers at The University of Granada’s School of Translation and Interpretation attempts to analyze and evaluate the results of machine translations done with popular online tools such as Google Translator, Promt, and WorldLingo. The study was published in this month’s issue of Translation Journal, and it raised interesting questions for me about the possible uses for online machine translation. Looking at the findings, it should come as no surprise that all of the machine translation tools produced poor results in terms of the number of errors, or that after the translations passed through a round of human editing, the number of errors were drastically reduced.
What is interesting, though, is that certain online tools performed better than others, and specific language combinations produced varying results. The graph below shows results from German into Spanish (the researchers used EvalTrans Software). An open-source shallow-transfer machine translation engine and t. Untitled. Do You Speak International? - Gill Corkindale - Harvard Business. By Gill Corkindale | 3:58 PM June 28, 2007 I may be in the minority here, but has anyone else noticed the strange things that have been happening to English lately? I don’t mean business speak or management and technical jargon, but the way we are all starting to speak “international.”
The lingua franca of the business world has gone global. More people are speaking English today — but are we all speaking the same language? I ask this question because recently I have found myself struggling to follow some conversations between executives from different countries. This month I have coached two teams: one comprising Scandinavians, Italians, and Japanese and the other Africans, Dutch, and Lithuanians. They were all capable businesspeople, speaking in a second or third language.
But on a few occasions it all went wrong. This is an interesting question. Exactly my thoughts. Can you translate 34,501 words in 10 hours? | GTS Blog. An SDL advertisement in 2009 claimed that one person translated over 34,000 words in a 10 hour period using the new SDL Trados product release. The logic in the ad: buying a Trados license is a no-brainer and the expense can be recouped quickly. Many people, however, were upset by that ad and called it false advertising. You can read the discussion here. Close to 100 comments were written until the discussion was closed for further comments.
In another discussion about SDL Trados on LinkedIn (read it here) over 35 comments were received, while more comments keep coming in daily. The following are some selected responses from these two discussions: I have been using Trados and/or SDLX since 1996 and nothing comes close. Franck Abate A few years ago I was misled by SDL’s sales pitch which was phrased in a way which suggested that Concordance search in Workbench does more or less the same thing as EBMT (example based machine translation) in Dejavu. Piotr Bienkowski, Poland No problem, folks. Welcome back, Yeeyan. If I had to pick a project that most excited me in 2009, it would be Yeeyan, a distributed translation project focused on making influential English-language media accessible to a Chinese-speaking audience.
Yeeyan’s founders built a community that included thousands of translators and struck partnerships with content providers like The Guardian, giving them permission to publish translated content. I was particularly struck by the talk Yeeyan cofounder Zhang Lei gave at the 2009 China Internet Research Conference at UPenn Annenberg – he made it clear that the motivation behind Yeeyan was a desire to use translation as a bridge between cultures, letting Chinese and English-speakers see the world from each other’s perspective. I was singing the project’s praises to a journalist last week when he pointed out that Yeeyan’s website was down.
Danwei’s article on Yeeyan’s closure gives a sense for how abrupt the move was. So what’s next for Yeeyan? What every mass marketer needs to learn from Grouch. Perhaps the most plaintive complaint I hear from organizations goes something like this, "We worked really hard to get very good at xyz. We're well regarded, we're talented and now, all the market cares about is price. How can we get large groups of people to value our craft and buy from us again? " Apparently, the bulk of your market no longer wants to buy your top of the line furniture, lawn care services, accounting services, tailoring services, consulting... all they want is the cheapest. The masses don't want a better PC laptop. They just want the one with the right specs at the right price. It's not because people are selfish (though they are) or shortsighted (though they are). Fixing this is almost always a losing battle. The Marx Brothers were great at vaudeville.
Then the market for movies like the Marx Brothers were making dried up. It's extremely difficult to repair the market. It's a lot easier to find a market that will respect and pay for the work you can do. The middlemen: how translators are boosting India's writers - Th. "Good enough" machine translation - Windows Live. At Localization World 2008 in Madison I heard many remarks on how machine translation cannot compete with human translation, MT with human post-editing is not increasing productivity and how in general raw MT is only appropriate for "not so important" content like support articles. Many good points, but I couldn’t shake the feeling that it could not be that bad. Returning from Madison I ate Chinese food at the Detroit airport and my chopsticks came in this wrapper: I had a good laugh.
I do suspect that this is a human translation, but certainly not one that would meet the quality criteria of LSPs or their customers. When measured against a high-quality human translation this translation would probably get a very low BLEU score. But the translation gets its point across and is also quite funny. This got me thinking about what quality expectations we as humans have towards machine translation. The answer does not lie in some automated measure like BLEU, TER or METEOR.
Why do I think that? Videos uploaded by asiaonline. Untitled. Ideas. Better Translating Service the problem of making an intelligible automated translation -- much more one that's actually reasonable-sounding -- is not easy, and it's not exactly solved anywhere. the proposal is to make translations more direct and decipherable by not "sugar-coating" them trying to put them into regular English (or any other recipient language obviously) grammar. For example, Mandarin grammar is *completely* different from English grammar.
When translating from Mandarin, notation (brackets, etc.) could be used instead of normal English grammatical structures to relay the grammatical structure of the Mandarin phrase. Most people aren't too stupid to get the hang of such a system after a couple of translations; the bigger problem is simply not knowing the words of another language. Another example of this feature is how to handle agglutination. Tolingo secures Series A funding for fast, cheap translation ser. [Germany] Tolingo, the online translation platform, has secured a Series A round of investment. The investment comes from Neuhaus Partners in Hamburg which is using a local start-up fund run by a public programme in conjunction with the local KfW bank (hey, this is Europe remember). The investment will be used to expand further in Europe and internationally. Terms were undisclosed but sources say it is in the €1 million to €2 million range.
Tolingo was founded in 2008 and now has over 2,500 certified translators online at any one time. Matthias Grychta, managing partner at Neuhaus Partners says Tolingo’s platform “significantly increases translation speed.” Tolingo has competitors in the shape of Livetranslation.com and mygengo as a comparible system. By combining expert knowledge and innovative vision, tolingo have created a unique online system that marks a new era in translation.
The Italian Job | Microtask. January 14th, 2010 by Tommaso De Benetti If you’ve ever holidayed in Italy, you might have noticed that many of the locals, even in popular tourist spots, struggle with English. There are probably a number of causes for this phenomenon, such as a rusty education system and the all southern conviction that “you don’t really need English anyway”, but perhaps the major reason why the Italians and other Southern European countries struggle is the habit of dubbing popular TV shows and movies into their native tongue. Lost in Translation Take for example the situation in Italy, which I am personally familiar with. The Italian Dubbers’ Union might pride itself as being the best in the world, but it seems clear that their expertise contributes to the linguistic shortcomings of their countrymen.
While many Italians love their dubbed TV, a large number prefer subtitles in order to maintain a show’s original flavor and to help develop their English skills. Machine Translation – articles for reading « Communicate Digital. Casacuberta, F., Civera, J., Cubel, E., Lagarda, A., Lapalme, G., Macklovitch, E., et al. (2009). Human Interaction For High-Quality Machine Translation.
Communications of the ACM, 52(10), 135-138. Retrieved from Academic Search Complete database. This article provides with deep understanding the specific problems of machine-based translation of legal texts. Savoy, J., & Dolamic, L. (2009). The authors (Switzerland) of the article evaluate current systems of translation for using in search queries by monolingual users in bi- or multilingual countries (Canada, EU). Steding, S. (2009). The article examines the situation of machine translation usage by students in education in Germany. ————- Waibel A., & Fugen C. (2008). In this article the authors review state-of-the-art speech translation systems. Wilks, Y. (1989). This article reviews the book “Machine translation: Theoretical and methodological issues” and provides with the short overview of history of machine translation (MT) in USA. Untitled. Collaboration. Unprofessional Translation. Swedish translation system gets EU funding. Synaptic Web: Realtime is Just the Beginning.
Crowdsourcing Tools. Untitled. Fanlation ~ English to French translations. Welcome to UNTERM. Darpa Looks to Build Real-Life C3P0 | Danger Room. Machine Translation Terminology - SDL Blog. A Translatours Catechisme (II): A New Drill : : Translation Guy. Translation As a Force of Change. Ushahidi & The Unprecedented Role of SMS in Disaster Response – Rule-based MT vs. Statistical MT: Does it Matter?
Malaproprisms From The Bully Pulpit - 13 Translation. About MT & smartphones, speech recognition and hybrids « Pangean. Collaborative Translation by Monolinguals with Machine Translato. Amazing Pictures | Chinese restaurant menu translation - oh oh.. Interview with Edith Grossman at The Boston Globe « By The Firel. Putting Google to the Test in Translation - Graphic - NYTimes.co. Google’s Translator Toolkit Is Helping The Machine Become A Bett. Consider the Luddites. The Ongoing Quest for “Best” MT Translation Quality.
Blog: Offering Your Content in 100 Languages. Machine Translation « Communication Desideratum. Some things are better lost in translation | Natalie Haynes - Ti. The state of OmegaT. Launchpad Translations. Blogos. Machine and Human Translation 2: Recycling COMAL. When machine translation and volunteer translators collide: A Yo. Web Innovation Project » Is Machine Translation Coming of Age? About Us. Notes and reflections from the Open Trans. Global Voices Online » Global Voices Author Demographics 2010. The tradeoff between open and closed cdixon.org – chris dixon's. Falling Translation Prices and Implications for Translation Prof. Posts. The incredible shrinking delivery times | Translator's Shack. Freelance Translators Talk about Machine Translation « CETRAblog.
Adobe launches translation crowdsourcing in China | Global by De. No Monkeys! Lost in Translation.