Linguists crack Skype encryption, Military Metaphors

Computational linguists have reconstructed an encrypted VOIP conversation that took place on Skype.

What is surprising is that though they are encrypted, the frames that make up a Skype call contain clues about what phonemes are being spoken.

Further, ARPA have decided to build a mountain of metaphors:

Not just American/English metaphors mind you but those of Iranian Farsi, Mexican Spanish and Russian speakers. Why metaphors? ‘Metaphors have been known since Aristotle as poetic or rhetorical devices that are unique, creative instances of language artistry … Over the last 30 years, metaphors have been shown to be pervasive in everyday language and to reveal how people in a culture define and understand the world around them

Translating Ulysses

Terms like digital native are funny in that “what will the kids think of next” way, but is actually driven home as quite appropriate when you see stories like the translation of Joyce’s Ulysses into Twitter’s 140 char context.

As translators are quick to tell you, translation is more than just transcribing words from a 2 way dictionary verbatim – there is nuance, metaphor and culture that need to be taken into account. And that’s where these digital natives are winners – they can think in Twitter:

This is not an attempt to tweet mindlessly the entire contents of Ulysses, word-for-word, 140 characters at a time. That would be dull and impossible. What is proposed here is a recasting or a reimagining of the reading experience of this novel, start to finish, within the confines of a day-long series of tweets from a global volunteer army of Joyce-sodden tweeps.

If you are interested in volunteering for this project, go over to the organiser’s site:

There are no academic or professional credentials required. Ulysses is the world’s book, so if you can read, you can be in The Brave Cast.  But there are a few requirements…

Google Translate API is deprecated

Google has cited economic reasons when announcing that they will be shutting down the service:

Important: The Google Translate API has been officially deprecated as of May 26, 2011. Due to the substantial economic burden caused by extensive abuse, the number of requests you may make per day will be limited and the API will be shut off completely on December 1, 2011. For website translations, we encourage you to use the Google Translate Element.

It should be pointed out that this is not an isolated incident – Google is shutting down a bundle of others as well: “some of our older APIs have been superseded by bigger and better things and others may not be receiving the necessary love“. And for a company that could buy the music industry, economic abuse is a pretty big call – either it’s breaking into other Google realms unexpectedly, or there’s another product on the horizon – maybe something through Google Apps for Enterprise?

For those that are thinking “what the hell is an API? What does that mean for me?”, and API is the programming interface to software – so anytime you used Google Translate in OmegaT (for instance), or you had a website automagically translated on the fly, it was using the Google Translate API. The API is how the OmegaT developers know what to “send” to Google so that Google will respond, and what form that response will come in.

Does this mean that Google Translate is disappearing? I doubt it – the blog has a an interesting piece up at the moment – Define, translate and search for words in Google eBooks – about a new feature in Google Books, where one can get a word or phrase translated on the page:

translate a single word or several sentences of content into dozens of languages, from Afrikaans to Yiddish, by selecting the “Translate” option. As with definitions, you’ll see the translated text displayed in the pop-up window.

And, following up from my previous post of Google Translate’s listen feature being used as a beat box, they have since admitted:

When we built (it) we thought it was a cool tool, but we have to admit we had fairly straightforward ideas about what it would be useful for (lowering language barriers and making more web content available to people around the world). As with many inventions, though, it turns out people have found uses for the tool that we never imagined.

What have people done? Well, there’s Google Translate for Animals, a couple of girls (successfully!) order Indian food via Google Translate, and in Taiwan, nearly half a million people are watching pop songs:

Users input Chinese words or phrases into Google Translate and then use the automated voice to create songs or to spoof music videos, dramatic acting scenes, etc. Some of them are very simple, straight melodies created with the automated translator.

From the look of the video offered, I would suggest that there’s something missing in translation – from my perspective it’s some pleasant c-pop with funny translation errors. I would suggest the view count comes from subtle or clever translations, that are the direct result of change of meaning that adding new Chinese characters to the existing words create. For example, see the translation change form:

I really… -> I do not… -> I am really not good at writing songs… (from 1:55-1:59)

How Language alters Imagery

This wonderful post on the poetry blog Harriet gives a fantastic example of the mash between techno-futures and cultural pasts:

Never before has language had so much materiality — fluidity, plasticity, malleability — begging to be actively managed by the writer. Before digital language, words were almost always found imprisoned on a page. How different it is today, when digitized language can be poured into any conceivable container: text typed into a Microsoft Word document can be parsed into a database, visually morphed in Photoshop, animated in Flash, pumped into online text-mangling engines, spammed to thousands of e-mail addresses, and imported into a sound-editing program and spit out as music — the possibilities are endless.

Make sure you watch the video – it’s the best bit!

Universal Subs announces a Japan Recovery Team

Universal Subtitles have just announced the Japan Recovery Videos team:

For the past several weeks, the most watched videos on Universal Subtitles were both related to the nuclear crisis in Japan.  The first is a documentary about Cherynobl(sic), translated into Japanese.  And the second is a video presenting a theory about the recent nuclear meltdown in Japan.

We are working hard to make Universal Subtitles more useful and responsive in crisis situations around the world.

Today we are launching a Japan Recovery translation team, to translate additional videos that are related to the recent crisis.  Please send us links to important videos and pass along this message to folks that may be able to translate to and from Japanese.

Linguistic studies look for methodologies in other fields

One of the things I’ve always liked about the beginnings of AI was the disparate fields that the first theorists came from – biologists and librarians coming together and finding common ground.

For the same reason, I’m pretty stoked to see that a couple of recent studies in Linguistics have taken methodologies from other fields. A study published in Science magazine that proposes that all languages come from a proto-language, using genetic variation as an example of where it’s worked before.

Another study from Nature is suggesting that common elements in languages have evolved independently and separately.

MKB sums it up well:

I’m not sure whether these two sets of results can be easily compared to one another. The studies were aimed at answering very different questions, so you can’t just line one up against the other. Depending on your point of view, these results may be contradictory … but that’s not necessarily the case. What I do think is interesting about these two studies is the fact that both are based on research methodologies and theories that were born in the fields of evolutionary biology and genetic anthropology.


Only two speakers left, and they’re fighting.

If you have half an eye out for pop cultural references to language, you would have been hard pressed to miss this news item a couple of weeks ago. Making all the channels based on it’s quirky human interest (or disinterest) level, is the somewhat sad story of the last two speakers of “Nuumte Oote” (Ayapaneco to the rest of us) who are refusing to talk to each other. My original source: Gawker