Home | About | Archives | Contact

Saturday, October 31, 2009

Facebook I18N: Way More Than A Token Gesture

Tokens (markers that are replaced at run-time by other text or values) in strings can be the bane of a translator’s life if used incorrectly because they frustrate a correct translation being made. However, I love the way that the Facebook translation tool allows you to comment on their use as you translate. 

Looking at the options below for commenting on token use is an education in itself (the tokens concerned are {number} and {chat-service-name}).

This approach allows users to comment as much on the effectiveness of the internationalization (i18n) practice as on the quality of the translation.

Facebook’s internationalization best practices for developers are here.

Facebook: Available in How Many Languages?

I’m sure we’re all very familiar with the vaunted Facebook crowdsourcing translation model by now. It’s been central to Facebook’s phenomenal international growth, and it’s a fantastic innovation even subject to a U.S. patent application. Anyone who supports the global sharing of information can only but admire what Facebook have done, meself included!

But I’m stuck here. Maybe you can help me?

As a user experience (UX) professional, I can see how allowing users to translate their own content can be part of a compelling engagement strategy, and within that context I would have thought the entire user experience should be in the user’s language, not just part of it.

So, then, why is it that when we constantly read that Facebook is available in 65, 70, 80, whatever number of languages, we can find that the Facebook help is available in less than 10? Here is what Irish language (Gaeilge) users see under Help:

Is it because:
a) The Facebook crowdsourcing translation tool doesn’t allow the help strings to be translated?
b) Facebook users don’t want to translate help because they don’t like or need it, or doing so just ain’t cool (or easy) anyway?
or
c) There’s a whole bunch of places out there populated by people way way smarter than others and they don’t need help in their own language?

As a localization professional working according to budget, I was sometimes faced with the prospect of having to preside over a localization plan where help or doc not included and left in English (actually, Facebook doesn’t seem to allow users who switch their language to one where no help translation is available an option to read help in English instead). I wondered: if this approach was acceptable then why the help was written in English in the first place?

For me, partial localization is fine if the market and user experience accepts it, of course, though it’s clear that for some cultures doing so is a negative experience.

But what’s going on with community translation of user assistance like help?

Answers to the organizers of the next localization or UX conference, anywhere, please.

Friday, October 30, 2009

iPhone's Pic Translator translates for you

A new application for the iPhone allows you to take a picture of some text and then translate it. Not only that, with text-to-speech technology, it will pronounce the text as well.

With translation capability from English into 16 languages and pronunciation in FIGS plus Portuguese, the Pic Translator seems aimed at the monolingual American who ends up at a restaurant with indecipherable entrees. An interesting app—and at a cost of 99 cents, it seems worth a shot!

(Thanks to Fast Company for this item!

Friday, October 02, 2009

Facebook and Google: new translation tools free to the masses

Facebook’s recent announcement that it is releasing a new translation tool free of charge to other websites has caught the attention of such media giants as the New York Times. Scooped by the Times once again!

The announcement of a free translation tool based on Facebook’s own methods comes as a bit of a surprise on the heels of the hubbub over Facebook’s patent application over its method of crowdsourced translation (which you can read a bit more about in the upcoming issue of MultiLingual magazine… too bad this latest notice arrived just prior to going to print). To top it all off, Google announced it also is offering a free crowdsourcing method to translate websites. Quick thinking, Google.

What this means in the long run remains to be seen, of course.

Friday, September 18, 2009

Striking a blow for idioms

A book on “novel"-sounding idioms in ten languages will save you the trouble of learning the languages yourself, according to Jag Bhalla, the author of I’m not Hanging Noodles on your Ears and other Intriguing Idioms from around the World, a collection of 1,200 idioms from National Geographic Books, illustrated by New Yorker cartoonist Julia Suits. 

A guest column on the book by the author mentions a few examples, such as to live like a maggot in bacon (to live in luxury: German) and to strike the 400 blows (to sow wild oats: French). The last one may not be so “novel” to fans of French cinema, who will undoubtedly recall a French New Wave flick by the same name (Les quatre cents coups). Apparently, there are two bands that have discovered this delightful idiom as well.

The column also discusses the use of idiom in language (which seems nearly unavoidable, at least for the native speaker) so check it out if you’re interested in anything beyond literal one-to-one language.

Thursday, September 03, 2009

Sand calligraphy

Feast your eyes on these pretty symbols.

No, they don’t necessarily fit together into a meaning message, but who can complain when looking at beautiful calligraphy on a gorgeous beach with the light hitting the symbols “just right”?

(The characters in this piece take their inspiration mainly from the Adinkra symbols of the Akan people of Ghana and the Ivory Coast.)
If looking online is not enough, Andrew van der Merwe also offers prints for sale.

Tuesday, September 01, 2009

Don't sleep, there's a good book

“Don’t Sleep, There Are Snakes” is a fascinating book by Dan Everett detailing living, studying language and observing culture among the Pirahã people of the Amazon jungle.

Everett, who is currently the chair of Languages, Literature and Cultures at Illinois State University, was funded by SIL to study the Pirahã language, and to translate the New Testament into it. Over a span of thirty years, he spent seven living among the people, studying their language and culture. His findings challenged his own (commonly accepted) linguistic theories as well as his belief system. Listen to the language here, and see nonrecursion demonstrated:

The stories in the book are riveting, and the linguistic discussion is approachable even for me, a non-linguist. But the most intriguing part is Everett’s realization that the Pirahã people have a culture completely different from ours. The obvious superficial differences aside, they have no counting system, no color names, no creation story, no history and they are completely content with laughter-filled days. And the title? That’s a common good night phrase – part teasing, part true.

Friday, August 28, 2009

Technology Company in Silicon Valley Applies for a Patent

If you’d been following @localization on Twitter, you’d know it was announced days ago that Facebook has applied for a patent (in the U.S.) called “Community Translation On A Social Network” (this was covered on the Baltimore Sun’s Tech Blog before anywhere else.)

The details, filed with the U.S. Patent and Trademark Office, aim to patent the much-vaunted social network translation process we hear so much about: where Facebook “volunteers” contribute freely-provided translations, which are then “voted” on as being appropriate or not. The filing states:

Embodiments of the invention provide techniques for translating text in a social network. In one embodiment translations of text phrases are received from members of the social network. These text phrases include content displayed in a social networking system, such as content from social networking objects. A particular member is provided with content including a text phrase in a first language, and the member requests translation into another language. Responsive to this request, a translation of the text phrase is selected from a set of available translations. The selection is based on actions by friends of the member in the social network, the actions being associated with the set of available translations. These actions can the viewing of or approval of translations by the friends, for example. The selected translation is then presented to the member requesting the translation.

(I admit being familiar with some of this kind of language, although lawyers were paid to explain it.)

This development would seem to offer a great deal of potential to not only extend the debate about crowdsourced translation, to whom the benefits really accrue, at what cost to others, and so on, but to ask the question “where will this all end up?” Except that a linkedinfail-style debate hasn’t ignited (yet.) Facebook, although wrong-footing the pundits very badly on this, is really doing what anyone else in the business (and it is a business) would attempt to do, so such a patent application isn’t surprising. It’s what tech companies do all the time, and why wouldn’t Facebook want to try and “own” a core global user engagement process like this? Amazon, after all, made a science out of it (and nice profits after a bit.)

However, Facebook is not alone in the social networking space (and the filing is specifically for “social networks") using the approach for translation, but that’ll be a “prior art” issue that the U.S. Patent and Trademark Office will need to examine.

All I can say - as the pursuer of a patent application with the U.S. Patent and Trademark Office myself - is these things can take years to conclude - even if the patent is to be awarded eventually.

Personally, I think this patent application will eventually fail, but yet again Facebook has made its translation process a headline event in even trying (nothing wrong with that.) You may have an opinion on all this, so until we hear from the U.S. Patent and Trademark Office, we’d like to read it ...

Monday, August 24, 2009

Chinese input on the new Mac OS

Perhaps it is because the new Mac operating system is nicknamed “snow leopard.” (What’s with the cat names? Personally, I am holding out for “margay"-- should be fast and small. But I digress.) The snow leopard’s habitat is mostly in China.

And the new Macintosh OS (more mundanely known as OS X 10.6) has a fun input method for Chinese characters. According to Apple’s release, “You draw them right on the Multi-Touch trackpad in your Mac notebook. They’ll appear on the screen in a new input window, which recommends characters based on what you drew and lets you choose the right one. The input window even offers suggestions for subsequent characters based on what you chose.” Fun! I can’t wait to see it!

The OS will sell for $29. Plus the cost of a Mac notebook, for Chinese input. 

Saturday, August 08, 2009

Translation Party

Bored? Try Translation Party.

Yes, if you’ve nothing to do because Twitter and Facebook are broken again, then amuse yourself for hours this way. Translate phrases backwards and forwards between English and Japanese using Google Translate until you reach a state of “equilibrium.” Or your boss appears.

More on Translation Party from Techcrunch.

If you think that’s pointless, then consider some people have no problem paying good money for back-translations.

Friday, August 07, 2009

Whiffling away

Adam Jacot de Boinod brings us a blog and tweets based on his book.

The Wonder of Whiffling by Adam Jocot de Boinod explores “extraordinary words in the English language.” He has now started a blog on the topic, and also tweets regularly. The blog website also has several pages of pages from the book.

Check it out! It’s good for at least a snirtle. 

Wednesday, July 29, 2009

Localized Spam

Are people six times less likely to be irritated by localized spam?

Perhaps you are familiar with Common Sense Advisory’s report “Can’t Read, Won’t Buy: Why Language Matters on Global Websites.” Well, it seems the spammers are on to this fact, as well. Slashdot reports that some Europeans are being targeted with automatically translated spam.

Danch Danchev has a great blog entry on this phenomenon.

Members: Login | Register

About

Blogos covers language through multilinguality and translation, localization and global markets, individual skills and emerging technologies, enablers and barriers, knowledge and speculation.

Categories

• Language in Business
• Language in the News
• Language Industry News and Events
• Localization Culture
• Personalization and Design
• Translation Technology

Archives

• Complete Archives
• Category Archives

Most recent entries

• Facebook I18N: Way More Than A Token Gesture
• Facebook: Available in How Many Languages?
• iPhone's Pic Translator translates for you
• Facebook and Google: new translation tools free to the masses
• Striking a blow for idioms
• Sand calligraphy
• Don't sleep, there's a good book
• Technology Company in Silicon Valley Applies for a Patent
• Chinese input on the new Mac OS
• Translation Party

Syndicate

• RSS 1.0
• RSS 2.0
• Atom

Search:

Advanced Search

MultiLingual Subscription Center

About MultiLingual About MultiLingual digital Subscribe to the print or digital magazine Free sample digital issue

Other Blogs

• Common Sense Advisory Global Watchtower
• Global by Design
• Inttranews: Daily News for the Language Industry
• Lingformant: The Science of Linguistics in the News
• Language Log
• Gilbane Globalization Blog
• The Global Small Business Blog
• Localization Project Management Log
• Web-Translations
• L'Observatoire de la traduction
• Beyond Words
• Medical Translation Blog
• Tomedes Blog
• Translation and Software Localization Blog
• Translation Current
• Localization, Localisation