All news

Dancing sign language interpreter raps with her hands interpreting for Snoop Dogg

Source: Laughing Squid
Story flagged by: Jared Tabor

Holly Manniaty Snoop Dogg NOLA Jazz Fest

An incredibly animated dancing sign language interpreter brilliantly rapped with her hands, capturing the lyrics and spirit of Snoop Dogg‘s performance at the 2017 New Orleans Jazz Fest. The interpreter, a Portland, Oregon resident named Holly Maniatty has provided her uniquely wonderful skills to a variety of famous acts including Bruce SpringsteenWu Tang Clan and many, many more. In 2014, Holly took part in Jimmy Kimmel’s “Sign Language Rap Battle with Wiz Khalifa“.

See video (explicit language) >>

“Freelance Isn’t Free” Act goes into effect in NYC

By: Susana Magnani

This Act will provide resources for freelancers to claim overdue payment from clients. Although applicable to all freelancers, I thought it would be of interest to our community.

… the law mandates that freelancers be paid in full for work worth $800 or more, either by a date set forward in writing or within 30 days of completing an assigned task.

Wouldn’t it be great to have this kind of protection everywhere else?
(Article by Emma Whitford on Gothamist)

Machines to Replace Human Translators? They Already Have: English Tabloid

By: Peter Gleason

[UPDATE 1530 CET 2017/05/15: The report in question has now been published by Adzuna (link) ]

CRACOW, Poland, May 15 —England’s Daily Mail apparently has an exclusive on the end of the Translation & Localization Industry as we know it. If the British ‘tabloid’ is to be believed, the end is not merely nigh, it’s already here: according to an admittedly ungooglable “study from jobs search engine Adzuna” of “79 million job adverts placed in Britain in the previous two years,” robots are already taking human translators’ jobs on a “grand scale,” and with blame/credit belonging primarily to “Google … among those to have designed automated translation software, which is making human translators increasingly redundant.”

The news also made it around the Commonwealth, being picked up this morning by the Australian, who also failed to link to or otherwise properly reference the ephemeral report. Nevertheless, it ominously quotes UK job site Adzuna co-founder Doug Monro as predicting, “Automation is already replacing jobs and could be set to replace some roles, like translators and travel agents, entirely.”


Increase in research of neural machine translation in the first half of 2017

Source: Slator
Story flagged by: Jared Tabor

Academia continues to ramp up its research into neural machine translation (NMT). Five months into the year, the number of papers published in the open-access science archive,, nearly equals the research output for the entire year 2016. The spike confirms a trend Slator reported in late 2016, when we pointed out how NMT steamrolls SMT.

As of May 7, 2017, the Cornell University-run had a total of 137 papers in its repository, which had NMT either in their titles or abstracts. From only seven documents published in 2014, output went up to 11 in 2015. But the breakthrough year was 2016, with research output hitting 67 contributions.

NMT, or an approach to machine translation based on neural networks, is seen as the next evolution after phrase-based statistical machine translation (SMT) and the previous rules-based approach.

While many studies and comparative evaluations have pointed to NMT’s advantages in achieving more fluent translations, the technology is still in its nascent stage and interesting developments in the research space continue to unfold.

At press time, NMT papers submitted in 2017 were authored by 173 researchers from across the world, majority of them (63 researchers) being affiliated with universities and research institutes in the US.

The most prolific contributor is Kyunghyun Cho, Assistant Professor at the Department of Computer Science, Courant Institute of Mathematical Sciences Center for Data Science, New York University. Cho logged 14 citations last year.

He has, so far, co-authored three papers this year —  “Nematus: a Toolkit for Neural Machine Translation,” “Learning to Parse and Translate Improves Neural Machine Translation,” and “Trainable Greedy Decoding for Neural Machine Translation” — in collaboration with researchers from the University of Edinburgh, Heidelberg University, and the University of Zurich in Europe; the University of Tokyo and the University of Hong Kong in Asia; and the Middle East Technical University in Turkey.

Aside from Cho, 62 other researchers with interest in NMT have published their work on arXiv under the auspices of eight American universities: UC Berkeley, Carnegie Mellon, NYU, MIT Computer Science and Artificial Intelligence Laboratory, Cambridge, Stanford, Georgia Institute of Technology Atlanta, Johns Hopkins University, and Harvard.

Sixty-one researchers from Europe have also substantially contributed to the collection, with authors from the UK (18), Germany (11), Ireland (13), and the Netherlands (7) submitting the most papers.


Best Translated Book Awards 2017 winners announced

Source: Three Percent
Story flagged by: Jared Tabor

The tenth annual Best Translated Book Awards were announced on May 5th, with Lúcio Cardoso’s Chronicle of the Murdered House, translated from the Portuguese by Margaret Jull Costa and Robin Patterson, winning for fiction, and Alejandra Pizarnik’s Extracting the Stone of Madness,translated by Yvette Siegert, winning for poetry.

See more >>

Facebook to open source its neural machine translation

Source: Slator
Story flagged by: Jared Tabor

Facebook is claiming that a new approach to machine translation using convolutional neural networks (CNNs) can help translate languages more accurately (read: increase quality on a BLEU scale) and up to nine times faster than the traditional recurrent neural networks (RNNs). CEO Mark Zuckerberg himself announced the news in his own Facebook page.

The company’s bold claims were anchored on results of a study conducted by five members of Facebook’s Artificial Intelligence Research (FAIR) team and outlined in detail in a paper entitled “Convolutional Sequence to Sequence Learning.” “To help us get there faster, we’re sharing our work publicly so that all researchers can use it to build better translation tools,” Zuckerberg said. Research authors Jonas Gehring, Michael Auli, David Grangier, Denis Yarats, and Yann N. Dauphin shared in an accompanying post on the Facebook developer blog that the FAIR sequence modeling toolkit (fairseq) source code and the trained systems are available under an open source license on GitHub.

Dr. John Tinsley, CEO & Co-Founder, Iconic Translation Machines Ltd., who reviewed the paper, told Slator that the results are impressive. “It’s quite a different approach, using convolutional neural networks (CNNs) as opposed to recurrent neural networks (RNNs). The reason this hasn’t been looked at for translation before is that CNNs typically work well with fixed-length input and RNNs with variable-length input. Obviously, with language, things are very variable so RNNs were the natural starting point,” he explained. His concern, though, is quality. But he observed that some of the shared task data reported are comparable to if perhaps a little better than existing approaches to Neural MT. “However, the single biggest impact of this work is the speed,” he said. “One of the current drawbacks of Neural MT is how long it actually takes to train the models, and this approach by Facebook using CNNs allows them to be trained up to seven times faster. This is because it’s much easier to parallelise the training process of CNNs given how they process different parts of the data (simultaneously as opposed to sequentially). That being said, it still requires powerful hardware.”

On the release of the source code, Dr. Tinsley said he approves of the open source approach. He said, “It’s good to see the tech behemoths taking this approach now and opening up their research to the wider community.”


Translators without Borders improves and expands its translation platform

Source: Translators without Borders
Story flagged by: Jared Tabor
Kató is the improved and expanded translation platform, formerly known as the Translators without Borders (TWB) Workspace, and it is where much of the magic happens. Kató connects over 500 non-profit partners with a diverse community of volunteer translators and many other language services. First established as the TWB Workspace in 2011 following a collaboration between TWB and, the online platform has since helped non-profit partners such as Doctors without Borders, Refugee Aid and Save the Children to share essential information in local languages and to translate over 40 million words. Today, the revamped Kató is more robust than ever with computer-assisted translation tools, functionality for storing common words and taxonomies and even bigger incentives for the community. Translators can now use Kató to interpret for all media, including providing subtitles and voiceovers for videos. The platform is even being used to train fluent speakers of languages that desperately need more translators and interpreters.


40+ million words translated so far

4,000 professional translators

500+ non-profit partners

190 language pairs

See more >>

Could machine translation help slow the decline of endangered languages?

Source: MIT Technology Review
Story flagged by: Jared Tabor

The best guess is that humans currently speak about 6,900 different languages. More than half the global population communicates using just a handful of them—Chinese, English, Hindi, Spanish, and Russian. Indeed, 95 percent of people communicate using just 100 languages.

The other argots are much less common. Indeed, linguists estimate that about a third of the world’s languages are spoken by fewer than 1,000 people and are in danger of dying out in the next 100 years or so. With them will go the unique cultural heritage that they embody—stories, phrases, jokes, herbal remedies, and even unique emotions.

It’s easy to think that machine learning can help. The problem is that machine translation relies on huge annotated data sets to ply its trade. These data sets consist of vast corpora of books, articles, and websites that have been manually translated into other languages. This acts like a Rosetta Stone for machine-learning algorithms, and the bigger the data set, the better they learn.

A map showing how the past tense indicators cluster for 100 of the languages investigated.

But these huge data sets simply do not exist for most languages.  That’s why machine translation works only for a tiny fraction of the most common lingos. Google Translate, for example, only speaks about 90 languages.

So an important challenge for linguists is to find a way to automatically analyze less common languages to better understand them.

Today, Ehsaneddin Asgari and Hinrich Schutze at Ludwig-Maximilian University of Munich in Germany say they have done just that. Their new approach reveals important elements of almost any language that can then be used as a stepping stone for machine translation.

The new technique is based around a single text that has been translated into at least 2,000 different languages. This is the Bible, and linguists have long recognized its importance in their discipline.

Consequently, they have created a database called the Parallel Bible Corpus, which consists of translations of the New Testament in 1,169 languages. This data set is not big enough for the kind of industrial machine learning that Google and others perform. So Asgari and Schutze have come up with another approach based on the way tenses appear in different languages.

Most languages use specific words or letter combinations to signify tenses. So the new trick is to manually identify these signals in several languages and then use data-mining techniques to hunt through other translations looking for words or strings of letters that play the same role.

For example, in English the present tense is signified by the word “is,” the future tense by the word “will,” and the past tense by the word “was.” Of course, there are other signifiers too.

Asgari and Schutze’s idea is to find all these words in the English translation of the Bible along with other examples from a handful other language translations. Then look for words or letters strings that play the same role in other languages. For example, the letter string “-ed” also signifies the past tense in English.

But there is a twist. Asgari and Schutze do not start with English because it is a relatively old language with many exceptions to the rule, which makes it hard to learn.

Instead, they start with a set of Creole languages that have developed from a mixture of other languages. Because they are younger, Creole languages have had less time to develop these linguistic idiosyncrasies. And that means they generally contain better markers of linguistic features such as tense. “Our rationale is that Creole languages are more regular than other languages because they are young and have not accumulated ‘historical baggage’ that may make computational analysis more difficult,” they say.


The language we speak affects our perception of time: research

Source: Times Live (NZ)
Story flagged by: Jared Tabor

New research has revealed another powerful effect that language can have on the brain, finding that the language we speak can influence the way we experience time.

Carried out by Professor Panos Athanasopoulos, a linguist from Lancaster University, UK, and Professor Emanuel Bylund, a linguist from Stellenbosch University and Stockholm University, Sweden, the study is the first to find evidence of cognitive flexibility in people who speak two languages.

Although bilinguals easily and quickly switch between their two languages without even thinking about it (a phenomenon called code-switching), it is not just a matter of switching to a different language for communication — different languages influence the way we think about the world around us.

The study found that people who speak two languages fluently think about time differently depending on the language they are using when estimating the duration of an event.

The research team explain that for example, Swedish and English speakers prefer to mark the duration of events by referring to a physical distance such as a short break, or a long wedding, with the passage of time perceived as distance traveled.

Greek and Spanish speakers on the other hand, prefer to mark the duration of events by referring to physical quantities, such as a small break, a big wedding, with the passage of time perceived as volume.

For the research the team recruited Spanish-Swedish bilinguals, whose two languages look at time differently.

Participants were asked to watch either a line growing across a screen, or a container being filled, and estimate how much time had passed.

At the same time as watching they were either prompted with the word ‘duración’ — Spanish for duration — or ‘tid’ — Swedish for duration.

The results showed a clear difference between the two languages.

When watching containers filling up and prompted in Spanish, participants based their time estimates on how full the containers were, not by the lines growing on screens, suggesting that they perceived time as volume.

However, when given participants were prompted in Swedish, participants based their time estimates on lines growing on screens, suggesting that they perceived time as distance traveled.

“By learning a new language, you suddenly become attuned to perceptual dimensions that you weren’t aware of before,” commented Professor Athanasopoulos.

“The fact that bilinguals go between these different ways of estimating time effortlessly and unconsciously fits in with a growing body of evidence demonstrating the ease with which language can creep into our most basic senses, including our emotions, our visual perception, and now it turns out, our sense of time.”

Slator’s reader poll results on payments, the busy season, and finding freelancers

Source: Slator
Story flagged by: Jared Tabor

Slator has published the results of a recent reader poll, on topics such as payments for translation services in the future, LSPs’ busiest quarters of the year, predictions on LSP stock performance, and where vendor managers go to find freelance translation professionals.

From the section “How to Find Freelancers”:

Professional freelance linguists are the lifeblood of the language services industry. While translation marketplaces may serve their purpose, every LSP worth its salt will see it fit to build and maintain a core pool of freelancers — experts in their field and intimately familiar with specific client requirements.

A number of large buyers have recently opted out of using LSPs entirely, choosing to work instead with freelance talent. The EU Court of Justice, for example, despite having 600 “lawyer-linguists” on staff, issued a EUR 6m freelancer-only tender at the start of the year. Even the European Central Bank awarded 10 freelancers a place in its EUR 2.4m contract.

So where do vendor managers go to find freelance translators? Trade shows, according to 38% of respondents in a recent Slator poll. Others find linguists via ProZ (21%) and referrals (18%). Associations (7%) posted about even numbers as universities (7%).

See the full set of results >>

smartCAT and Lilt announce partnership

Source: Slator
Story flagged by: Jared Tabor

Translation Management Software (TMS) provider, smartCAT and Lilt, an interactive and adaptive Machine Translation (MT) tool, have partnered to combine the latest in MT technology with a robust collaborative translation environment.

smartCAT, whose integrated approach to translation automation offers a broad range of human and machine translation solutions to their customers, are excited to be adding the latest MT technology. Their willingness to rapidly adopt new technology for the success of their customers made them a perfect candidate to take advantage of Lilt’s REST API. The technology enables programmatic integration with Lilt, as well as their Javascript library, lilt.js, which allows the addition of interactive, adaptive machine translation to a CAT editor.

Lilt’s adaptive MT is now available within the smartCAT editor, giving smartCAT’s customers access to this technology in a single activation click.

“Lilt fit just right into the smartCAT ecosystem. We loved the futuristic approach to machine translation it promotes, so delivering the technology to our users instantly became our priority. What makes this integration so special is that it takes smartCAT’s unique real-time multi-user collaboration to the next level. Each time a translator confirms a segment, the engine instantly trains and provides the correct suggestions to all the participants in the project, helping them to maintain consistency across the text. Despite the technology behind the new feature being complex and unfamiliar, we made sure it’s intuitive and easy to use.” said Ivan Smolnikov, CEO at smartCAT.

Lilt’s mission of making high-quality translation widely available led them to offer their REST API and Javascript library to translation solution developers, like smartCAT, in order to incorporate a powerful adaptive MT technology into their existing systems. Partnerships such as this one give customers and translators access to better translation quality, as well as a more ergonomic environment compared to the traditional methods of post-editing.


Top Language Lovers 2017 competition now open for nominations

Source: Lexiophiles
Story flagged by: Jared Tabor and Lexiophiles would like to warmly welcome each and every one of you to this year’s TLL competition. The game is on – the competition has officially started! It’s once again time to find and nominate your favourite blogs/Facebook pages/Twitter accounts and YouTube channels. Last year’s participants will be automatically nominated.

See more and make a nomination >>

“Connections”, a new magazine by, for and about translation professionals

By: Jared Tabor

A new online magazine called Connections has released its first issue. Connections collects interviews, articles and other contributions from translation professionals, and is a product of the members of the Standing Out Mastermind group.

This first issue of Connections contains contributions from, among others, the following members:

Noura Tawil

Pharmacist &Writer,Naturally Translated!

Carolina Garrido

16 years´ experience, MA in education

Anzhelika Kuznetsova

Psychology, Finance and Law

Rea Gutzwiller

Rafa Lombardino

A network of professional translators

You can read the full magazine by clicking on the link or image below:

Read online >>

Terminology introduced for the self-driving vehicle sector

Source: Dealerscope
Story flagged by: Jared Tabor

The Consumer Technology Association (CTA) has announced self-driving vehicle terminology designed to enable a common lexicon among the technology industry and better explain to consumers the terms and concepts of this rapidly innovating sector.

The definitions were developed and approved by CTA’s recently-formed Self-Driving Vehicles Working Group – chaired by Daimler North America and Waymo, and comprised of 34 companies – which also supports driverless vehicle consumer research and policy advocacy.

Among the terms and concepts addressed within the self-driving vehicle terminology:

  • Advanced Driver Assistance Systems (ADAS) or “Driver-Assist” Features: Onboard systems developed to improve safety and performance – examples include lane departure warnings, collision avoidance, adaptive cruise control and automatic braking
  • Aftermarket Technology: Technology services or upgrades provided by companies – unaffiliated with the vehicle manufacturer – added after a vehicle is sold or leased
  • Driving Environment Sensing: The capturing, processing and analysis of sensor data (e.g., cameras, radar, LIDAR) to enhance or replace what a human driver senses
  • MaaS (Mobility as a Service): The shift from personal ownership of transportation modes to shared transportation systems and services
  • Platooning: Synchronous operation of multiple vehicles, often in a convoy, to increase road capacity and efficiency
  • Self-Driving Vehicle: A vehicle capable of fully modeling its environment through an array of sensors, maps and other data in order to navigate and drive without human interaction
  • Urban Mobility: The ability for people in urban and suburban areas to access all modes and forms of transportation.


“The Language of Science and Education: An Expanded Glossary” now available in English and Arabic

Source: University of Arkansas
Story flagged by: Jared Tabor

William McComas’ first doctoral students at the University of Arkansas and at the University of Southern California where he formerly taught were both from Saudi Arabia. So, it’s perhaps no surprise that the idea for one of his recent books was suggested by these Saudi contacts.

“My colleagues in Saudi Arabia wondered if there was a resource that would allow them better access to the literature of science education. They had encountered terms that didn’t translate in the way they were being used in science education. You could look them up in a dictionary but that definition didn’t make sense in the science education context. Essentially, they asked me to write a book to fill this very special need,” said McComas, who holds the Parks Family Professorship in Science Education in the College of Education and Health Professions at the U of A.

This conversation encouraged McComas to produce The Language of Science Education: An Expanded Glossary of Key Terms in Science Teaching and Learning, first published in 2014 by Sense Publishers. That was in English. Now, a new edition has come out from King Saud University Press, which published a new version side-by-side in English and Arabic.

“Every discipline uses words in a context-specific fashion,” McComas said. “For instance, the term ‘informal science learning’ could be confusing because it has a unique meaning in our discipline. Even terms such as ‘laboratory’ and ‘inquiry learning’ could require explanation.”

His Saudi collaborators suggested a list of terms used specifically in science education, and McComas sent these to other science educators for review and to make suggestions for additions. Then, he worked with a team of graduate students and together they researched primary sources to create definitions. Each term in the book has a simple, one- or two-sentence definition followed by a more in-depth discussion of its origin and use in the context of science education.

The original book has been well-received and is cited frequently, McComas said, so he decided to expand on the idea. Now, he and Conra Gist, an assistant professor of curriculum and instruction, are working on one that will be a glossary of the special language of curriculum studies.

See more >>

Elia introduces the Engage membership initiative

Source: Slator
Story flagged by: Jared Tabor

The European Language Industry Association (Elia) has launched a new membership initiative for language service companies and independent professionals to connect and join forces with the ultimate goal of better serving end clients, strengthening the language industry in the process.

Companies and individuals who believe in the power of positive working practices are invited to sign up as Founding Members before 31 May 2017 and help shape the development of Elia Engage and its activities.

Building on the aims of Elia’s annual Together conference, which provides the venue for both parties to come together in person for open discussion and constructive dialogue, Elia Engage is the interactive forum for companies to meet and develop connections with skilled independent language professionals and, together, establish best practices for enduring, long-term, mutually beneficial partnerships.

Elia Engage will be led by committee, with independent professionals and language service companies equally represented. Founding Members will have numerous opportunities to influence the initiative and will be able to join various working groups that will focus on developing key aspects to help both parties achieve success and fulfilment in their businesses. This is an unheard of opportunity to be part of something from the ground up and to contribute directly to creating a legacy for the language industry built on respect and positivity.

All Elia Engage members will get a profile on the Elia Engage website to promote their services, the opportunity to communicate across borders, access to resources and services, and more. In addition, individuals will receive a 10% discount to attend future Together events.

Elia Full Member companies receive access to Elia Engage as part of their Elia membership and simply need to sign up and commit to the aims of Elia Engage and actively support positive working relationships. The membership fee for independent professionals such as translators, interpreters, localisers, consultants and project managers is €110 for 12 months. Register as a Founding Member by 31 May 2017 and receive the prestige of being an early adopter and the chance to influence the industry; independent language professionals who sign up as Founding Members will benefit from extended membership until 31 December 2018 for a special rate of €90.

See more >>

Translation platforms cannot replace humans– but they are still astonishingly useful

Source: The Economist
Story flagged by: Jared Tabor

In the past few months free online translators have suddenly got much better. This may come as a surprise to those who have tried to make use of them in the past. But in November Google unveiled a new version of Translate. The old version, called “phrase-based” machine translation, worked on hunks of a sentence separately, with an output that was usually choppy and often inaccurate.

The new system still makes mistakes, but these are now relatively rare, where once they were ubiquitous. It uses an artificial neural network, linking digital “neurons” in several layers, each one feeding its output to the next layer, in an approach that is loosely modeled on the human brain. Neural-translation systems, like the phrase-based systems before them, are first “trained” by huge volumes of text translated by humans. But the neural version takes each word, and uses the surrounding context to turn it into a kind of abstract digital representation. It then tries to find the closest matching representation in the target language, based on what it has learned before. Neural translation handles long sentences much better than previous versions did.

The new Google Translate began by translating eight languages to and from English, most of them European. It is much easier for machines (and humans) to translate between closely related languages. But Google has also extended its neural engine to languages like Chinese (included in the first batch) and, more recently, to Arabic, Hebrew, Russian and Vietnamese, an exciting leap forward for these languages that are both important and difficult. On April 25th Google extended neural translation to nine Indian languages. Microsoft also has a neural system for several hard languages.

Google Translate does still occasionally garble sentences. The introduction to a Haaretz story in Hebrew had text that Google translated as: “According to the results of the truth in the first round of the presidential elections, Macaron and Le Pen went to the second round on May 7. In third place are Francois Peyon of the Right and Jean-Luc of Lanschon on the far left.” If you don’t know what this is about, it is nigh on useless. But if you know that it is about the French election, you can see that the engine has badly translated “samples of the official results” as “results of the truth”. It has also given odd transliterations for (Emmanuel) Macron and (François) Fillon (P and F can be the same letter in Hebrew). And it has done something particularly funny with Jean-Luc Mélenchon’s surname. “Me-” can mean “of” in Hebrew. The system is “dumb”, having no way of knowing that Mr Mélenchon is a French politician. It has merely been trained on lots of text previously translated from Hebrew to English.

Such fairly predictable errors should gradually be winnowed out as the programmers improve the system. But some “mistakes” from neural-translation systems can seem mysterious. Users have found that typing in random characters in languages such as Thai, for example, results in Google producing oddly surreal “translations” like: “There are six sparks in the sky, each with six spheres. The sphere of the sphere is the sphere of the sphere.”

Although this might put a few postmodern poets out of work, neural-translation systems aren’t ready to replace humans any time soon. Literature requires far too supple an understanding of the author’s intentions and culture for machines to do the job. And for critical work—technical, financial or legal, say—small mistakes (of which even the best systems still produce plenty) are unacceptable; a human will at the very least have to be at the wheel to vet and edit the output of automatic systems.

Online translating is of great benefit to the globally curious. Many people long to see what other cultures are reading and talking about, but have no time to learn the languages. Though still finding its feet, the new generation of translation software dangles the promise of being able to do just that.

Translation News roundup for April

By: Jared Tabor

We’re at the doors of May already, if you can believe it. Here are some of the highlights in Translation News for the month of April 2017:

Translation / Interpreting





From the blog…

The 90s kids who got into translation in order to localize a video game into English

Source: Kotaku
Story flagged by: Jared Tabor

Myria can’t remember exactly when she found out about Final Fantasy’s number problem—it was either 1996 or 1997—but she does recall seeing an advertisement for Final Fantasy VII. “We’re like, ‘Huh, seven?’” she said, echoing the thoughts of RPG fans across the United States. Just a few years earlier, in 1994, Squaresoft had released Final Fantasy III on the Super Nintendo. How’d they get from three to seven?

As it turns out, Square was holding out on North America. The venerable publisher had passed on localizing both Final Fantasy II and Final Fantasy III on the Nintendo Entertainment System, so when it came time to bring Final Fantasy IV to the west, they called it Final Fantasy II. Then, Square decided to skip Final Fantasy V, although they briefly considered releasing it here with a different name, according to their head localizer, Ted Woolsey. When they brought over Final Fantasy VI, they called it Final Fantasy III.

As Myria started to research Square’s weird localization choices, she started thinking about getting involved with unofficial fan projects. She’d always been obsessed with RPGs, and she’d noticed that Final Fantasy IV (II)’s script was particularly messy, full of clunky sentences and awkward word choices. “I wanted to redo that game,” Myria said. “It was a horrible mess in terms of its translation.”

While browsing the internet one day in the late 90s, Myria stumbled upon a group of likeminded geeks that called themselves RPGe. Hanging out in an IRC channel, they’d talk about their favorite Japanese role-playing games and make ambitious plans to write English translations for the ones that never made it west. When she found them, they were talking about localizing Final Fantasy V, which they’d do by cracking open a Japanese version of the game’s ROM file and translating the script to English. Myria was intrigued, putting aside her hopes of redoing FFIVFinal Fantasy V sounded way cooler. (A group called J2E would later retranslate FFIV to subpar results, as documented by Clyde Mandelin on his Legends of Localization website.)

Unlike the two NES games that we’d missed out on, Final Fantasy V was by all accounts excellent. People lucky enough to understand FFV in Japanese reported that it was a blast to play, with a solid story and an elaborate class-changing system that allowed players to customize their party in creative ways. It could be difficult, which was one of the reasons Square hadn’t brought it west, but RPG fans wanted to check it out nonetheless.

Problem was, RPGe’s methods were flawed. Nobody had done anything like this before, so there was no institutional knowledge about how to handle fan translations. The RPGe crew had dug up a Japanese ROM of Final Fantasy V, then cracked it open and started editing the text files, directly translating chunks of the game from Japanese to English. But these files were finicky and tough to handle. When you changed a line of Japanese to English in the ROM, it wouldn’t display neatly in the game, because Japanese characters were rendered so much differently than English ones. Japanese characters are bigger than English letters, and one sentence that takes 12 characters in English (“how are you?”) might just take three characters in Japanese (“元気?”). Final Fantasy V capped each line of dialogue at 16 characters, which looked fine in Japanese but would make an English translation garbled and hard to read.

What they needed to do, Myria realized, was edit not just the text files but also the code that Final Fantasy V used to handle those text files. “I really felt they had the wrong approach,” she said. “That was really my big insight to the ROM hacking community, that you can’t just modify the data of the game to make an effective translation—you have to modify the code as well.”

In order to localize a Japanese game in English and make it readable, Myria decided they would need to reprogram the game. Their version of Final Fantasy V would need to understand that English letters, unlike Japanese characters, have different sizes. They’d need to teach the game that each dialogue box should allow more English characters (including those pesky spaces) than it does Japanese kanji or kana.

Myria (who at the time went by the internet handle Barubary; both names are references to Breath of Fire) started talking with SoM2freak, a Japanese-English translator she met online, about splitting off from the rest of RPGe. By mid-1997, they were making plans to start their own translation of Final Fantasy V, done properly instead of hacked together. “I ignored those people who I felt didn’t know what they were doing,” she said. “We started our own sub group within [RPGe] because I felt they were not able to do this.”

As SoM2freak translated lines of Final Fantasy V’s Japanese dialogue to English, Myria tried to figure out the best way to implement them into the game. She downloaded a disassembler to break down Final Fantasy V’s code, turning it into a file so massive, she needed a special text management program called XTree Gold just to parse it. Then she started changing variables, using trial and error to discern what each line of code actually did. “There were no references on most of this stuff at all,” Myria said. “I just kind of figured out what to do.”

Perhaps the most controversial of the team’s translation decisions was the main character’s name. If you ask Square Enix, they’ll tell you that the star of Final Fantasy V is a man named Bartz. But if you played the fan translation, you’d see a different name: Butz.

It’s a name that’s elicited plenty of snickers over the years, but by all accounts it was the most accurate translation, and Myria stands by it. The alliterative translation of the Japanese name, バッツ, is Battsu, or Butz for short. “There were documents in Japan, for a strategy guide for example, and also these little silver statue things that had Butz the way we’d written it,” she said. “We used those kind of things as reference for intended translation.”

On October 17, 1997, Myria and crew released “v0.96,” the first public version of FFV’s fan translation. It went viral, making its way across IRC channels and message boards as RPG fans began to discover that there was a cool new Final Fantasy that none of them had gotten to play. Although SNES emulators were nascent and rough, it wasn’t too tough to get your hands on one. It was also simple for the average gamer to get a copy of Final Fantasy V and the English patch, which you could apply by following a set of simple instructions in the Readme file. “[The patch] really just spread on its own,” said Myria. “It quickly got news in the emulation community and people started playing it at that point. We didn’t have to market it at all.”

Final Fantasy publisher Squaresoft never contacted RPGe about their translation, according to Myria, even though their U.S. offices were in Costa Mesa, just a few miles away from her parents’ house. But in September of 1999, an official English version of Final Fantasy V finally made its way to North America. This version, bundled with Final Fantasy VI in a PS1 compilation called Final Fantasy Anthology, was a mess.

In the PS1 translation of Final Fantasy V, main character Faris insisted on speaking like a pirate for the entire game.

“We were laughing so hard,” said Myria, “because the translation was absolutely awful. We were like, ‘OK, a couple kids in high school over four months did a better job than Square. It probably took them at least a year. We were just laughing so hard.”

It wasn’t until the 2006 Game Boy Advance port—Final Fantasy V: Advance—that Square would finally release a decently localized version of the mistreated role-playing game, although the main character’s name remained Bartz. “When the Game Boy Advance version came out, I was like, ‘Oh my god, they finally beat us,’” said Myria. “It took them eight years, but they finally did a better translation than ours.”

Read the full article >>

KPMG-Google report on Indian languages, the internet and the role of content localizers

Source: Slator
Story flagged by: Jared Tabor

KPMG and Google published a report on on April 25, 2017 detailing the Indian language ecosystem. The report, “Indian Languages – Defining India’s Internet,” states that nine out of every 10 new internet users in India over the next five years will likely be Indian language users.

Defined as ‘Indian language literates who prefer their primary language over English to read, write and converse with each other,’ this user base is forecast to grow to 536 million by 2021 at 18% CAGR from 234 million in 2016. In comparison, the English internet user base is likely to grow to only 199 million at 3% CAGR for the same forecast period.

With this growth, the report predicts a corresponding rise in user-generated original content online, increase in time spend of Indian language internet users on different internet platforms, and what it calls the “hyper local consumption of local content.”

In turn, this will drive increased investment by businesses to establish digital Indian channels, the rise of digital advertisements in local languages, and spur more Indian languages enablement of digital payment interfaces and platforms such as mobile compatible content for applications and websites.

It is interesting to note that the KPMG-Google report included “content localization companies” and freelance translators in its Indian language ecosystem chart, recognizing their role in the grand scheme of things. A chart shows freelancers and LSPs at the intersection between content creators and content developers. Recognizing the need for human translation and service providers is remarkable in a report co-authored by the world’s most ardent proponent of machine translation.

Overall, the report confirms what Slator reported in January that the world’s second most populous nation may be a tough nut to crack with 22 official languages, but it is arguably the next frontier in language services.


Translation news
Stay informed on what is happening in the industry, by sharing and discussing translation industry news stories.



  • August 2017 (6)
  • July 2017 (23)
  • June 2017 (24)
  • May 2017 (24)
  • April 2017 (55)
  • March 2017 (84)
  • February 2017 (3)
  • January 2017 (3)
  • December 2016 (1)
  • November 2016 (9)

  • I receive the daily digest, some interesting and many times useful articles!

    The translation news daily digest is my daily 'signal' to stop work and find out what's going on in the world of translation before heading back into the world at large! It provides a great overview that I could never get on my own.
    susan rose
    United States Translation News daily digest is an e-mail I always look forward to receiving and enjoy reading!

    Your current localization setting


    Select a language

    All of
    • All of
    • Term search
    • Jobs
    • Forums
    • Multiple search