One thing I did before leaving the US, is that I broke down and spent a rather large sum of money on Rosetta Stone language learning software, for Korean. I had a couple of reasons.
Firstly, of course, there is my desperation to somehow get better with Korean, and therefore a willingness to try new and different things, and spend money doing them.
Secondly, however, was that as linguist, I've been wondering what, exactly, they were doing that allowed them to believe themselves a premium seller of language-learning tools, for that's the way they market themselves. Are they really that good? I wonder. There is so much in the way of really bad materials for language learning, devoid of any apparent familiarity with linguistic theory, often replete with errors and folk-judgements about things like sound change or grammaticality.
I've managed to work through the first 3 lessons of the first unit of my Rosetta Software Korean Level 1. Here are my thoughts.
As software, it's extremely well designed. Attractive, easy to figure out, intuitive, just as they claim. My primary complaint with the interface is the speech-recognition tool… I got lots of answers "wrong" as I worked through it because it simply doesn't seem to "hear" me. And it seems a little bit buggy in the way it handles not being able to "hear" you, leaping along and going "bing," "bing," (error, error) without giving you time to try again. Also, A few times, I became frustrated with a new type of exercise and the lack of instructions on how to do it, but ultimately I recognize that this is part of the "method" being used: they want the user to solve each exercise, each section, as a little puzzle, and be engaged at a more-than-analytical level in using the language.
As far as awareness of linguistic theory, I'm less impressed. They make the same sorts of grammaticality judgements as so many horrible "beginning Korean" texts, and I'm not sure the focus on the highly stilted, fully inflected forms of the nouns is going to lead me to any kind of communicative efficacy, down the road. Actual Koreans speaking actual Korean almost never use the kinds of singular-plural inflections they're teaching here, at least in my experience, for example. I'll try to keep an open mind.
The single most frustrating thing is the speaking exercises. Not just because of the wonky speech-recognition problems I described above, but because they give you little hangeul prompts for words to pronounce, but they aren't really useful at all — because they're not explaining or displaying or in anyway accommodating the extensive and overwhelming processes of sound change and syllable liaison that operate within Korean words and phrases. I can figure out what they want me to do, because I have a degree in linguistcs and several years of effort behind me in sounding out Korean hangeul, but I think that if I was a typical, linguistically naive language learner, my simple, heartfelt reaction would be: WTF!
I'm sure they have a couple lessons in the introductory part, where they're teaching you hangeul (which I didn't work through) — but honestly, exposure to the sound change and liaison rules is not the same as internalizing them — the software needs to hammer these rules home in these speaking exercises. I've seen plenty of beginning Korean texts that will provide two "spellings" for each word: a standard spelling and then a "sounded out" spelling that explicitly reflects the sound change and liaison rules in operation. If the software did this in its presentation of speaking exercises, I think it would be a lot more transparent. As it is, you hear a pronunciation and read the hanguel on the screen and wonder if the person narrating is looking at the same thing you are.
Lastly, I really think I'd have struggled immensely with making heads or tails of this software, if I hadn't brought with me the extensive background and previous effort in trying to learn the language. I've already been exposed to much of the vocabulary, and all of the grammatical concepts being covered, on and off over the last couple of years, and most importantly, I'm comfortable with the Korean writing system and can recognize syllables at a glance. And with all that, I'm getting only 85% correct on most of these units. How frustrated would I be if I was coming at it "cold"? I'd be giving up, is my guess, and muttering "impossible!"
So, that's my review based on about 6 hours of hard work with Rosetta. I'll stick with it, if only because for me, it's very helpful with vocabulary. But was it worth 400 bucks? A tentative NO.