Thursday, January 15, 2015

Google translate muscles into the personal conversation business

Luckerson, V. (2015, Jan. 14). Hands-on: Google Translate Is Now a Way Better Travel Companion. Time. Hands-on: Google Translate Is Now a Way Better Travel Companion.

Dougherty, C. (2015, Jan. 14). Google Translate App gets an upgrade. New York Times blog.

May, P. (2015, Jan. 14). Google's new Translate app shines in a crowded field. San Jose Mercury News.

The new app makes a lot of promises. Life will be much easier, they say, for every schmuck like me who is about to sojourn into a place where the local language is different from mine. I'll simply point my phone at the "atenciao piranha" sign (see picture, third article) and I'll know that there are piranhas in the river.

The first things that are wrong with this picture are 1) I can't turn on my cell phone outside of the country, and 2) I can't seem to keep it charged. There's also the fact that 3) it's pretty obvious by the fish picture that the sign is about fish, and, 4) I wasn't about to go skinny-dipping in the Amazon anyway, but let's disregard those. Here are some more serious ones. 5) Translation programs can make things worse, 6) Not every language is as easy to translate as Portuguese, 7) Siri can add an unexpected dimension to a personal situation, 8) phones don't work as well after you smash them into the concrete barriers that hold up hotels.

Here are some interesting points brought up by the articles. First, the fact that Google has moved in on this business has increased the success of machine translation so that it is definitely on a track toward better in virtually every language. It promises 38 (out of what, several thousand?), ok, but it's improving on those 38, and it's got the money to hire people that are making those better. So we're living in a time when machine translation, instant, by phone, is in its infancy, but nevertheless, capable of overturning everything we know, if only because it can only get better. The last time I checked Google Translate, Spanish and French were already pretty good (the first article mentions this), a number of languages were miserable, but there was a wide variety and clearly an arc toward better in each one. If they are getting better, we are dealing with a dynamic system which as one article says "might put high school Spanish teachers out of business."

Here's one thing that's scary. The geeks at Google are proud of themselves in that they made Google Translate better by turning Google's massive calculation apparatus on the languages themselves. May quotes Cattau as saying, "We base translation on machine learning, by looking at billions of Web pages that have been translated into other languages," says Cattiau. "We find 'dog' has been translated millions of times into 'chien,' for example, so the computer now knows the two mean the same thing." Ah, but the computer is doing exactly what people do, basing its experience entirely on learning and reality. Thus, we can trick the computer, or we can change the meaning of things simply by changing the way they appear. How is the computer supposed to know? It's pretty easy with a word like 'chien' which is pretty uncontroversially, universally recognized to be 'dog'. Their claim of course is that roping in thousands of translation statistics helps them strengthen their assertion that 'chien' actually means 'dog' which is not complicated, in this situation, by other factors. In the world of translation it's usually complicated by other factors. Translation is dogged by these problems, in fact.

More later. I'm going to see what my phone does with "y'all cain't".

Labels: ,

Monday, January 12, 2015

War on Passive, bib.

Gopen, G. (2014, Winter). "Why the Passive Voice Should Be Used and Appreciated" Litigation, (ABA Journal), Vol. 40 #2, pp.16-17.

Leverett, T. (2014, Oct.). Documenting the war on passives. thomas leverett weblog.

Sheffield, N. (2011). Passive Voice in Scientific Writing. Duke Graduate School Scientific Writing Resource. Accessed 1-15.

Pullum, G. K. (2014, Jan. 22). Fear and Loathing of the English Passive. Language and Communication. Accessed 1-15.

Rhodes, S. (1997). The active and passive are equally comprehensible in scientific writing. Doctoral dissertation, Univ. of Washington. Accessed 1-15.

Neurobonkers. (2015). The Passive War: "The blind warning the blind about a nonexistent danger". The Big Think. Accessed 1-15.


Watson, A. (2012, Feb. 20). Navigating ‘the pit of doom’: Affective responses to teaching ‘grammar’. English in Education. Accessed 1-15.

"...Interviews with 31 (secondary English) teachers reveal two discourses which frame the ways in which teachers express their feelings (about grammar teaching): a dominant discourse of grammar as threatening, reactionary and dull, and an oppositional discourse which positions grammar as inspiring, fascinating, and empowering." (from the abstract).

Jean, G. & Simard, D. (2011, Aug. 10). Grammar Teaching and Learning in L2: Necessary, but Boring? Foreign Language Annals. Accessed 1-15.

"...Results showed only slight discrepancies between students’ and teachers’ beliefs and perceptions, and very few differences according to the target language and students’ gender or age. The main findings suggest that grammar instruction is perceived by both students and teachers as necessary and effective, but not as something they enjoy doing." (from the abstract)

Myhill, D., Jones, S., Watson, A. & Lines, H. (2012, Nov. 27). Playful explicitness with grammar: a pedagogy for writing. Literacy. Accessed 1-15.

Truscott, J. (2006, Oct. 27). The Case Against Grammar Correction in L2 Writing Classes. Language Learning 46(2), 327-369. June 1996.

Labels: ,