Hacklog: Blogamundo — poking holes in the language barrier since approximately 1 month from now

In a previous post I took Google to task for not indexing content in the Khmer, Burmese, and Ethiopic scripts.

Thanks to Denis for pointing out that Google has fixed them all!

Yay!

I guess they read this blog.

*cough*

Clearly, this is justification for a big pointless graphic.

Random idea:

Igmuwatogla

(by stickywikit)

It would make sense to automatically generate picture dictionary pages like this one, in this case for animal words in the Lakota language, using services such as the Flickr API.

Tim Bray has a post about the utility of using image search as an Online Picture Dictionary. He gives the example of figuring out what a citrouille is with Google Images.

It would be neat to go further: one could build whole lists like the Lakota learning page above — obviously you’d have more luck searching for the English translation of the Lakota words, and there’s no shortage of hits, but a horse is a horse, of course.

See for yourself: here’s a šunkawakan and a igmuwatogla and a šunkmanitu tanka is. (Hmm… I wonder if šunk is some sort of prefix?)

And as long as we’re on the topic of things Siouxan , here’s an interesting article on Scrabble as a learning tool for Dakota. I hope they rebuilt distribution and points on the tiles to correspond to Dakota frequencies!

The fine folks over at one of my favorite blogs, Language Log, occasionally run a language quiz, wherein they invite readers to identify unknown languages from a recording.

I thought it might be fun to have a similar contest here, and I just ran across a blog that would be a perfect test scenario: oldulo

As a bit of inspection will reveal, the blog is in Japanese and Esperanto.

Imagine for a moment that you know nothing of either — most people who can read English can make out a few words of Esperanto, after all. (Japanese is another story, but even if you know some Japanese, pretend you don’t.)

Unlike the Language Log quizzes, there isn’t really an “answer” here, I’m just interested in peoples’ thoughts. Well, there is an answer, but it’s not terribly hard to find — the interesting bit is the approach.

Okay, enough preliminaries, here’s the challenge:

Make the case for whether or not the Japanese and Esperanto texts in each post are in fact translations of each other, and do so without reference to any of the following:

• Transliteration
• Dictionaries
• Google

I’ll probably do up some Python scripts, myself, but use any other tools you’re familiar with, up to and including plain old reason.

If you happen to speak both Esperanto and Japanese… well… pretend you don’t, or wait for the next challenge. ☺

Yes, it’s true, we’re in deep hack.

And it looks like we’re going to stay that way for the next week or so, so updates will be light.

But let me assuage the demand (*cough*) for updates from the world of Blogamundo with a convenient reference to the world’s worst translated menu.

Oh, and as long as we’re on the subject of bad translation, here are some amusingly awful tattoos in “Japanese” . Ouch! (Found that via an old post by Joi Ito.)