Slashdot Mirror
Startup Hopes To Crowd-Source the Developing World
GalaticGrub writes "Technology Review has an article about a startup that wants to build a business out of crowd-sourcing the developing world. The company, called txteagle, seems to be interested mainly in using local knowledge to translate information into less common languages. The Finnish cell-phone company Nokia is a partner in the project, and CEO Nathan Eagle says that it provides a good example of a Western company that could benefit from txteagle workers. Eagle explains that Nokia is interested in 'software localization,' or translating its software for specific regions of a country. 'In Kenya, there are over 60 unique, fundamentally different languages,' he says. 'You're lucky to get a phone with a Swahili interface, but even that might be somebody's third language. Nokia would love to have phones for everyone's mother tongues, but it has no idea how to translate words like "address book" into all of these languages.'"
What could possibly go wrong with this plan?
Have you read my blog lately?
They may not have any one particular translation for words either. Remember, a lot of these small languages have relatively few people who are not well connected with each other, which is necessary to expedite the definition of new words for new technologies.
Forget computers, these people may never have used a paper address book before..
It wouldn't surprise me to find out that something like "address book" might not even have a direct translation in some of these other tongues. Thankfully, this sort of crowd-sourcing should yield the closest usable match, which, of course, is the entire point. (Pat, I'd like to buy a 'comma'.)
This guy's the limit!
Duh... use Babelfish.
http://alternatives.rzero.com/
Kenya has a population of about 37M people. That means that most of those languages are bit players in the country. It's questionable as to how many of those languages would actually have a large enough population that cares about using their primary language on their phone to exclusion of cheaper translations for much larger languages, and how many people using those languages could even afford a phone.
Learn English, you're gonna have to one day anyway, so start now. Problem solved.
It will fill up with vandalism and bullshit. However, it won't have the OCD-inflicted monkeys on Wikipedia who get to the vandalism within a month or two. The vandalism will be there forever, on your company's cell phone. I can see the startup menu now:
WELCOME TO NOKIA
1) Send Call
2) Check Email
3) Fuck Your Mother
...that we all speak different languages. These people are welcome to try and make a profit off these inefficiencies. But the fact that this market exists (or, perhaps, the fact that these txteagle people might be able to convince some VCs it does) says to me that we should be trying to teach these people a more global language, so they can participate on equal footing rather than being marginalized.
If your theory is different from practice, then your theory is wrong.
Nokia is exactly the sort of company who could, very easily, hire 60 different people (full time no less), who all had English (or whatever) as a second language and also had writing skills, each of whom could be in charge of the localization for their particular "first language". The additional manpower cost would be truly insignificant to their bottom line, and they'd end up with well-translated manuals, support documentation, et cetera.
This has a far greater relevance for someone with a low- or un-funded project than a major multinational corporation.
You're special forces then? That's great! I just love your olympics!
P.S.: Arrest, try, convict, and sentence the world's most dangerous person.
P.S.: Arrest, try, convict, and sentence the world's most dangerous person.
Fixed that for ya.
"Wise men talk because they have something to say; fools, because they have to say something" - Plato
What would Microsoft do if, due to the economy, individual users forgo upgrading to Windows 7, and decide to keep what they have. Will Microsoft go and say, "XP is old and we won't be supporting it anymore. Also, by the way, VISTA will only be supported for another year."
At that point, Microsoft would alienate users of previous versions of Windows, the users will decide that they need to choose something else, and a large corporation ends up loosing a large chunk of the marketplace.
I would say that Microsoft needs to test the waters prior to dropping support for XP, otherwise people that could decide to upgrade after the ecomony improves would end up getting comfortable with whatever alternate they end up choosing and tell Microsoft that they're not coming back!
I'm kind of thinking that xlating cellphone menus into dialects that may not even have a single book written in them is... silly. Why don't they have books? Because illiteracy is rampant in the developing world. Will cellphones now become the driver for people to learn to read? Oh wait, in the poorest countries most of people don't have cellphones: http://www.web4dev.org/images/8/8d/Africa_Mobile_Fact_Book_2008.pdf
Sometimes translations are even barriers to usability- my Swedish wife can't use the Swedish-localized version of Excel because the menus don't make sense to her after having used it in English so much.
I actually agree, he should be prosecuted for being a mental risk for all internet users.
Blah, blah, blah...
"Wienerschnitzel"
"Sauerkraut"
"Leberwurst"
"Blitzkrieg".
"...und der Ullstein mit der Wurst!"
happy kraut-sourcing ....
A colleague of mine provided a "translation" of HP BASIC for april fools one year (35 years ago - HP2000). PRINT became SCRIBBLE, IF became SHOULD, GOTO became LEAP, LET became MAKE
10 I = 0
20 SHOULD I = 10 LEAP 50
30 SCRIBBLE I
35 MAKE I = I + 1
40 LEAP 20
Interestingly, no one had any trouble with their assignments (programs were stored tokenized, so you saw the new keywords in the "editor").
...hmm let's see... gettext anyone? I mean, we recently uploaded our strings to one of those sites (https://launchpad.net/rosetta) where open source translators work together... for free we now have 12-something languages and a lot of corrections to even the original strings.
https://launchpad.net/rosetta: Launchpad Translations (codenamed "Rosetta") is a platform for open source application translation on the internet. It lets anybody help translate their favorite open source application into their favourite spoken language.
Launchpad supports most localizable open-source applications.
Custom electronics and digital signage for your business: www.evcircuits.com
So, they won't understand the desktop/file metaphor then, I guess. Hmmm... How about the dirt/rock/twig metaphor?
Exactly! Localization should start from concepts. They should use a high level description language to map the concepts into functionality. Forget about addressbook, welcome the (address|location|person|connection|relationship|friend|whatever)(table|list|storage|bin|jar|cabin|whatever). Processes shoud be given the same treatment.
Not that I would prefer to use whateverwhatever myself.
Then we could send them letters about our poor dead Uncle who was killed in a bloody coup and needing to transfer millions of dollars out of our county.
"The ferrets, they're every where I tell you!"
I would suggest they put this on the back burner and concentrate on fixing some of the obvious faults in their bloody handsets, such as getting the 'cancel' button to actually cancel the web browsing which has been triggered by the accidental pressing of the 'web' key (which is placed in the corner of the keypad).
They whose government reduces their essential liberties for temporary security, receive neither liberty nor security.
Teach everyone Globish http://en.wikipedia.org/wiki/Globish, a language already spoken, and you are done. Doesn't take much to speak from a 1500 word dictionary instead of the 175,000+ word Oxford dictionary.
No need to delay projects for the required 6912 translations (languages in existence: http://www.ethnologue.com/ because one word was changed in the UI. You need at worst 6912 translaters and at best 1 (who speaks all 6912 languages in existence today).
Globish is already out there, the lingua franca of globalization. Hi Y'All doesn't quite work when buying lumber in Papua New Guinea. A one-time translation of the 1500 words will last humankind for ever, unless physicists make up their mind how many dimensions there are in string theory and bring about contact with other civilizations.
Her lips were softer than a duck's bill, but her quacks
Am I the only one who got a little turned on by the phrase "everyone's mother tongues" ?
To be fair, neither have I...
People List?
Surely most cultures make lists and have words for people. So People List could work.
Acceptable substitutes:
*Friend List
*Community List
*Home/House List
*Contact List
s/list/$OTHER_ACCEPTABLE_WORD that is similarly descriptive.
If a culture doesn't have a word for potato, I'm sure they could go with small brown thing.
(Aw crap, I hope I didn't just inspire another poop-related troll post.
Random Thoughts From A Diseased Mind (Not For Dummies)
They may not have any one particular translation for words either. Remember, a lot of these small languages have relatively few people who are not well connected with each other, which is necessary to expedite the definition of new words for new technologies.
Forget computers, these people may never have used a paper address book before..
I think that's the entire point of this project. Everyone takes for granted that there is usually a correct canonical term for a certain type of thing, but it's obvious that there aren't, and sometimes there aren't even terms for things. Some words that have been canonized are even completely incorrect like "Kleenex" (tissue) or (coincidentally) "Rolodex"
Natural language translation is an incredibly subtle task. What's the difference between translating between one language and another, and translating between one dialect and another? Can there be a statement which simply cannot be translated for a particular language? Culture? Individual? People disagree about whether female authority figures should be addressed as "sir" or "m'am." Think about that.
It may be tempting to criticize this because of the obvious potential for catastrophic failures, or because these languages/regions simply don't have terms for certain ideas we (Nokia, apparently) want to give (or market) to them. Consider that perhaps many expressions you use today began life differently and changed; perhaps some phrase started out more sarcastic, or officious, but is more or less casual and direct when you say it. Maybe a humorously inappropriate translation that gets picked by the most people (ie. crowdsourcing) is simply the most memorable and will do the job of making sense to the user best.
I'm not saying this is the end-all killer localization app, but I think it's more than a little silly to assume these obvious pitfalls are fatal ones, or even that they aren't beneficial diversions off the beaten path; after all, when is the last time you saw Corporate America make use of the most poignant slang terminology in any way that wasn't transparent and condescending? What would all the names of skateboard tricks be called if they had all been invented by the marketing staff of a skateboard manufacturer?
For better or worse, the public should participate in the advancement of their language.
I heard about this concept a few years ago in an academic research paper by Craig Gentry, Zulfikar Ramzan, and Stuart Stubblebine:
http://portal.acm.org/citation.cfm?id=1064009.1064026
I must admit that it seemed a bit too science-fictiony, but maybe they were on to something.
Abstract: This paper is a preliminary exploration of secure distributed human computation. We consider the general paradigm of using large-scale distributed computation to solve difficult problems, but where humans can act as agents and provide candidate solutions. We are especially motivated by problem classes that appear to be difficult for computers to solve effectively, but are easier for humans; e.g., image analysis, speech recognition, and natural language processing. This paradigm already seems to be employed in several real-world scenarios, but we are unaware of any formal and unified attempt to study it. Nonetheless, this concept spawns interesting research questions in algorithm design, human computer interfaces, and programming language / API design, distributed systems, and cryptography, among other fields. There are also interesting implications for Internet commerce and the B24b model. We describe this research area and suggest a basic framework for the design of such systems. We analyze security and reliability against malicious parties using standard probability theory tools. We then derive design principles using standard decision-theory concepts. Finally, we list extensions and open problems.
I thought the third world was an excellent source for crowds?
Let's be honest here folks. Crowd sourcing is just a substitute for getting other people to do jobs white people are too cool to do themselves.
And if anyone gets the Brujeria reference, we salute you, my heavy metal brother or sister from Switzerland!