That's a good list. I think the most important first step is to find what really interests you (whether that be comic books, TV shows, movies, music) and use that as a starting point. For example, a lot of people I know learned Japanese because they were interested in anime. If you are studying something that truly interests you, it is much easier to stay motivated.
In my opinion, there is still a lot of room for innovation in the online language learning space. It is a very tough problem to solve and it is one that I am currently working on with my startup - BiFluent[1]. Our thesis is that one thing that is really missing for people learning a foreign language is metrics. If you are running a startup, I'm sure you could immediately tell me how many new users signed up today, what your current conversion rate is, etc. But can you do the same thing if you are learning a foreign language? Can you tell me what your level was 1 year ago, 6 months ago, 1 month ago? Can you tell me (precisely) what your strengths are, and what your weaknesses are (speaking, pronunciation, vocabulary, idioms, grammar, etc.)? Can you tell me which words you should be studying next, based on frequency analysis and tailored to fit your study goals? I believe that if you truly know your level, and if you are able to measure your improvement over short term periods (every month), it becomes much easier to stay motivated and stay on track.
One tool we are hoping to ship in the next week or so is a vocabulary assessment tool. Historically, the only way people have measured vocabulary is by the number of words you know. For example, "I know 5,000 words". In my opinion, this is a vanity metric. It doesn't really tell us any useful, actionable information. Our new tool reports what we call your Vocabulary Coverage Ratio. This is the percentage of the total vocabulary of a language that you will know or understand (analyzed on a frequency basis). So, for example, if your Vocabulary Coverage Ratio is 85%, this means that you will understand 85% of the vocabulary in any given situation. After we ascertain your level with our tool, we can also break out your coverage ratio across different mediums (TV, newspapers, magazines, fiction) using the data in our corpus. Also, once we know your vocabulary coverage ratio, we can use the frequency analysis in our database to give you tailored study lists. These study lists will show you the next 50 most frequent words you should be studying based on your current level and interest (we have different lists for medical, business, SAT, GRE, TOEIC, etc.). If you choose the right words to target, you can quickly improve your coverage ratio.
So in conclusion, I think the best "hack" is a mix of two things. 1) You need to find what interests you and use that as a jumping point 2) you need to use science and tools to accelerate your progress (just like we use analysis tools in every other aspect of our lives).
A common problem for tech-minded people when approaching human languages is to equate learning words with learning a language but it doesn't work that way.
The two big reasons are differing division of word boundaries and collocations. The second half of this 2 minute video explains both: http://www.youtube.com/watch?v=3cjnP6mogEU
Very true. One way around that is to put more than one word on your flashcard; a phrase of five-ish words enough to study in that way, but long enough to show some grammar and context ("the elephant has a long trunk").
I also agree that the most important first step is to find content that's interesting to you. Coming from France (a country well-known for its crappy language learning methods), I learnt english as a teenager because I wanted to understand comic books I had found on the Internet. Later on I subscribed to english speaking magazines such as the economist.
In my opinion, there is still a lot of room for innovation in the online language learning space. It is a very tough problem to solve and it is one that I am currently working on with my startup - BiFluent[1]. Our thesis is that one thing that is really missing for people learning a foreign language is metrics. If you are running a startup, I'm sure you could immediately tell me how many new users signed up today, what your current conversion rate is, etc. But can you do the same thing if you are learning a foreign language? Can you tell me what your level was 1 year ago, 6 months ago, 1 month ago? Can you tell me (precisely) what your strengths are, and what your weaknesses are (speaking, pronunciation, vocabulary, idioms, grammar, etc.)? Can you tell me which words you should be studying next, based on frequency analysis and tailored to fit your study goals? I believe that if you truly know your level, and if you are able to measure your improvement over short term periods (every month), it becomes much easier to stay motivated and stay on track.
One tool we are hoping to ship in the next week or so is a vocabulary assessment tool. Historically, the only way people have measured vocabulary is by the number of words you know. For example, "I know 5,000 words". In my opinion, this is a vanity metric. It doesn't really tell us any useful, actionable information. Our new tool reports what we call your Vocabulary Coverage Ratio. This is the percentage of the total vocabulary of a language that you will know or understand (analyzed on a frequency basis). So, for example, if your Vocabulary Coverage Ratio is 85%, this means that you will understand 85% of the vocabulary in any given situation. After we ascertain your level with our tool, we can also break out your coverage ratio across different mediums (TV, newspapers, magazines, fiction) using the data in our corpus. Also, once we know your vocabulary coverage ratio, we can use the frequency analysis in our database to give you tailored study lists. These study lists will show you the next 50 most frequent words you should be studying based on your current level and interest (we have different lists for medical, business, SAT, GRE, TOEIC, etc.). If you choose the right words to target, you can quickly improve your coverage ratio.
So in conclusion, I think the best "hack" is a mix of two things. 1) You need to find what interests you and use that as a jumping point 2) you need to use science and tools to accelerate your progress (just like we use analysis tools in every other aspect of our lives).
[1] https://www.bifluent.com