Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
The Internet Technology

Covid-19 Is History's Biggest Translation Challenge (wired.com) 45

Posted by msmash from the challenges-in-hand dept.
Services like Google Translate support only 100 languages, give or take. What about the thousands of other languages -- spoken by people just as vulnerable to this crisis? From a report: If we want to avoid a pandemic spreading to all the humans in the world, this information also has to reach all the humans of the world -- and that means translating Covid PSAs into as many languages as possible, in ways that are accurate and culturally appropriate. It's easy to overlook how important language is for health if you're on the English-speaking internet, where "is this headache actually something to worry about?" is only a quick Wikipedia article or WebMD search away. For over half of the world's population, people can't expect to Google their symptoms, nor even necessarily get a pamphlet from their doctor explaining their diagnosis, because it's not available in a language they can understand.

[...] In a pandemic, the challenge isn't just translating one or a handful of primary languages in a single region -- it's on a scale of perhaps thousands of languages, at least 1,000 to 2,000 of the 7,000-plus languages that exist in the world today, according to the pooled estimates of the experts I spoke with, all of whom emphasized that this number was very uncertain but definitely the largest number they'd ever faced at once. Machine translation might be able to help in some circumstances, but it needs to be approached with caution. [...] That's not to say that machine translation isn't helpful for some tasks, where getting the gist quickly is more important than the nuanced translations humans excel at, such as quickly sorting and triaging requests for help as they come in or keeping an eye on whether a new misconception is bubbling up. But humans need to be kept in the loop, and both human and machine language expertise needs to be invested in during calmer times so that it can be used effectively in a crisis.

The bigger issue with machine translation is that it's not even an option for many of the languages involved. Translators Without Borders is translating Covid information into 89 languages, responding to specific requests of on-the-ground organizations, and 25 of them (about a third) aren't in Google Translate at all. Machine translation disproportionately works for languages with lots of resources, with things like news sites and dictionaries that can be used as training data. Sometimes, like with French and Spanish, the well-resourced languages of former colonial powers also work as lingua francas for translation purposes. In other cases, there's a mismatch between what's easy to translate by machine versus what's useful to TWB: The group has been fielding lots of requests for Covid info in Kanuri, Dari, and Tigrinya, none of which are in Google Translate, but hasn't seen any for Dutch or Hebrew (which are in Google Translate but don't need TWB's help -- they have national governments already producing their own materials). Google Translate supports 109 languages, Bing Translate has 71, and even Wikipedia exists in only 309 languages -- figures that pale in comparison to the 500-plus languages on the list from the Endangered Languages Project, all human-created resources.
This discussion has been archived. No new comments can be posted.

Covid-19 Is History's Biggest Translation Challenge More

Covid-19 Is History's Biggest Translation Challenge

Comments Filter:

  • If a person is connected enough to get Covid (Score:3, Funny)

    by dmay34 ( 6770232 ) on Tuesday June 02, 2020 @01:49PM (#60136762)
    ... then they almost certainly speak one or more of the languages that Google translate works* on. *Google translate mostly works sometimes pretty well.
    • It's actually a big communication problem. For example, in San Jose we get text messages in English telling us to quarantine because of coronavirus, and a lot of the messaging is in English. However, there's a huge Spanish speaking community, and if they are watching TV or getting news from Mexico they are hearing news like, "Poor people can't get Coronavirus." The result, which is entirely predictable, is that the Spanish speaking community has been hit harder by Coronavirus than the rest of the region. Al

    • Re: (Score:2)

      by Livius ( 318358 )

      No, but whatever health care they have access to might. The problem is real but not as unmanageable as implied.

  • Weird (Score:1)

    by Anonymous Coward

    I thought Covid was over, what with all those massive gatherings in the cities these days.

    Oh, and weren't black people more vulnerable to Covid or something? Uhh, you guys do know that this won't end well for you, right? Viruses don't know what your politics are.

    • Saying "weird" is the right word. A week ago (for gawd sakes, literally a week ago) it was conservatives declaring victory and having Covid parties and refusing to wear masks, and matching on state capitals (and they were armed with assault rifles). Suddenly the script has completely flipped. It's almost comical. But that's 2020 for ya.

      • And we're only beginning the 6th month of 2020. I hope people refrain from saying things like "At least it can't get any worse", it's like a lightning rod that makes bad stuff happen.

        • This is an election year. If you think this is bad, you haven't seen anything yet. We're not even into November.

          From what I've seen, the Democrats will literally start Civil War II before they allow Trump to get re-elected.

          With any luck, they'll be on the losing side of that... again.

  • For instance, if you translate 'wash your hands' to French, it comes out as 'lave t'es mains'.

    Which is screwed up - basically coming out as "wash you're hands". And this is a common phrase being translated from English to French; one of the best cases for google translate, because it has a ton of source data and they've been working on the translator for over a decade.

    • "Ich bin ein Berliner" translates to "I am a jelly doughnut."

      • Re:Google translate doesn't work consistently well (Score:4, Informative)

        by Miamicanes ( 730264 ) on Tuesday June 02, 2020 @03:48PM (#60137268)

        > "Ich bin ein Berliner" translates to "I am a jelly doughnut."

        Obligatory explanation:

        In German, there are a lot of foods that are named after cities. A "Berliner" is basically a jelly donut. For a perfect English example, think "Wiener" (a/k/a, "hot dog"). It's so common, German disambiguates between "citizen of a city" and "food associated with the city" by omission or use of the definite article (in this case, "ein").

        "I am Wiener" = I am a citizen of Vienna

        "I am a Wiener" = I'm a gelatinous tube filled with mechanically-separated meat waste products and artificial flavorings.

        In English, the definite article "the" is (almost) always required for everything, which is basically the opposite of Russian, which doesn't use definite articles at all.

        In a hypothetical less-hostile political environment, Khrushchev could have probably been tricked by East Germans into making the opposite mistake a week later by giving him a jelly donut, then having a reporter later ask him (with TV cameras running) what he just ate. The same way Kennedy would have been inclined to add a definite article, Khrushchev would have been inclined to omit it and say, "I ate delicious Berliner!" & ended up being shown in political cartoons with Berliners in a cooking pot. Except, obviously, no East German would have ever DARED to pull a stunt like that, because the consequences would have been rather... dire.

    • It does? I just tried it and got "Lavez-vous les mains" which make sense.
      Perhaps a Google dev saw your post and pushed a quick fix in 15 minutes, otherwise you are not using the same Google translate than I am.

    • For me, "Wash your hands" gives me "Lavez-vous les mains", which is perfectly fine.

    • Re: (Score:2)

      by Livius ( 318358 )

      Obviously not human translator quality but that's actually not bad as a first attempt.

  • This is a solution looking for a problem (Score:4, Interesting)

    by ZeroPly ( 881915 ) on Tuesday June 02, 2020 @02:30PM (#60136902)
    Someone living in a rural village and speaking an obscure dialect of Urdu almost certainly knows another language so that they can go into the city and buy jeans.

    You don't need translation into "thousands" of languages. Consider how ridiculous this idea is, in an American context. Do you really need to translate this into Diné, because otherwise Navajo people on reservations would not hear about the pandemic? Of course not. They speak English just fine, and outliers would have existing access to translation.
  • Looks like someone spelled "history's biggest fraud" wrong.

    • Just because you've convinced yourself that the virus is not real does not mean it is a fiction.

      • Did I say it wasn't real? Maybe CV really is a translation challenge after all...
        • You said it's "History's biggest fraud." You are confused in that either A) you don't know history, or B) you think coronavirus is a fraud (or was presented fraudulently), or both. Either way you look ignorant.

          • You said it's "History's biggest fraud." You are confused in that either A) you don't know history, or B) you think coronavirus is a fraud (or was presented fraudulently), or both. Either way you look ignorant.

            I think you are the one confused my friend. Fraud is intentional deception to secure unfair or unlawful gain, or to deprive a victim of a legal right. Fraud can violate civil law, a criminal law, or it may cause no loss of money, property or legal right but still be an element of another civil or criminal wrong. That does not mean that there is not a real virus... it just means that they intentionally lied about a lot of the stories around it in order to get people into a panic, and lock down the world, d

  • I have a friend who works at a big local hospital and awhile back (mid March?) I asked her some questions about Covid-19 and what she has seen first-hand. One of the most surprising things she has noticed is the virus is disproportionately affecting people whose native language is not English. At the time, she wasn't sure if it was the language barrier or if the patients had been working public-facing jobs.

  • The news cycle has moved on, you can stop hyping this virus that doesn't kill 99.6% of people (according to the CDC):

    https://www.cdc.gov/coronaviru... [cdc.gov]

  • If your local Juju-man cannot read about Covid19 (Score:3)

    by ffkom ( 3519199 ) on Tuesday June 02, 2020 @03:26PM (#60137178)
    ... then that is probably the least of your health problems. In comparison to the health risks you are exposed to day in day out in societies so unconnected from the rest of the civilization that they did not even get the message about that new virus, Covid-19 is not really a significant additional risk. I'm also pretty sure the rate of deaths from Covid-19 on North Sentinel Island will remain at exactly zero.

  • I'm willing to bet that even in countries with a government that is producing it's own docs Google or Bing translate are used for drafting. Esp. if it's a country that by itself has multiple national languages.

    I myself surely would do it. I speak and write 4 languages fluently, but when I have a big piece of text in one of them and need it in another, I first try Google to get a quick draft. If the outcome looks reasonable, I'll go over it until all errors are fixed, not hesitating to sometimes make some s

  • I live where there is a lot of construction going on, and I’m not seeing physical distancing among construction workers. Will construction workers end up being a pool of transmission? I’m concerned. I use must-have app for ordering SST card. It is https://www.safetycardtracker.... [safetycardtracker.com] The good news is that the risk of transmission for those employed outside the healthcare sector is low, according to OSHA.

  • How fortunate that COVID-19 is the biggest source of income in a decade and Google has a loss-making service that needs a good drenching in taxpayer's money, When you think about it, everything is translation - isn't tho? I mean really?

  • Machine translation disproportionately works for languages with lots of resources, with things like news sites and dictionaries that can be used as training data.

    Well that's just false. If you use Rule-Based Machine Translation [wikipedia.org], then you don't need any training data. You just need a person who knows the languages and grammars well enough to encode it as rules.

    You can get very reasonable general-purpose machine translation results in 3 months of work - empirically proven by the many Google Summer of Code [withgoogle.com] pro

  • "Services like Google Translate support only 100 languages,"

    Last century, automatic language translation was viewed as an almost impossible task. Now, at the click of a button, we can translate Mandarin Chinese to English; maybe not perfect, but a million times better than I could do by hand (only knowing English, a bit of French, some Latin, and enough to order a beer in German, Spanish and Italian).

    If an academic had built a machine to translate, she'd have won countless prizes and awards. But Google di

Slashdot Top Deals

If all the world's economists were laid end to end, we wouldn't reach a conclusion. -- William Baumol

Close