Output practice is great after a certain amount of input. Without input, output cannot be effective enough.

I must read English more, and then my writing English will be felt more differently by me myself who write it. There will occur a sense of refinement of something in me. Language learners can feel such a sense of development, though, of course, it is not unique to language learning.

Today, thanks to the evolvement of the Internet environment, tools that assist your language learning have also evolved. Probably to keep a diary in English is easier than it was before the Internet.

And the progress of technology is also spectacular... You can marvel (or might have already marveled) at the power of recent striking evolvement of machine translation exploiting neural network and tremendous data. So you may think that due to machine translation, language learning perhaps might become somewhat meaningless in the future.

But no one knows whether the sharp rising curve could continue. But one of machine translation's weak point is when a translated sentence does not make sense. In such an occasion, you must check the original text. But if you lack the knowledge of the language in which the original text was written, how can you know the precise meaning of the sentence? You cannot ask the text like you ask a person talking with you.

The more advanced the content of text is, the more such occasions must arise. Such text requires precision in reading much more. So there remain the necessity of human language learning and probably this situation will not change in the near future.

If there appear an AI that can translate every text in perfect precision, the world must be utterly different in the first place. Such a world would be beyond our imagination. So today's apprehension pertinent to machine translation must likely be pointless, because the whole situation would change drastically at the time when such a powerful AI is realized.