Highlighting error with Kanji

I think there’s a problem with the highlighting. Target language is Japanese.

For example, today I was looking at A Very Short Autumn (With Kanji) from the store.

First sentence: 今朝、我が家で初氷を観測しました。

I highlight “今朝” and make a lingq telling myself that it means “this morning”. The word gets highlighted.

Then I look at “我が家”. It’s one word. I can select it and create a lingq, which appears in the vocabulary section.

But it’s not highlighted in the text. It appears that the highlighter will only recognize words separated by spaces or punctuation as words, and will not highlight “我が家” as a word.

Or am I supposed to store " 我が家で初氷を観測しました" as a word?

You have identified the main issue that exists in Japanese. We hope to get time to look into it in the new year. Essentially, because there are no spaces between words in Japanese, the system is unable to properly split the text into words. Only properly split words that are LingQed will be highlighted in the text. As you say, sometimes the way the words are split works out but mostly it doesn’t right now. So for now you will only be able to review most of your vocabulary in the Vocabulary section. Some of our Japanese learners are importing text and putting spaces between words in the text to try and get around this problem.

This is the reason why we haven’t introduced other Asian languages yet. However, there are many members who are still enjoying LingQ for Japanese despite its limitations.

I wouldn’t recommend putting spaces between words, if “real” japanese isn’t written with spaces.

Other than that, this is a nice site, especially the large amounts of stuff I see in the store. With audio, too.

I suppose this goes for Chinese as well. If the dictionary/translation tool doesn’t interpret the highlighted words correctly, one can always copy the words from content you are importing (e.g. web site or word document) and add them “manually” to the vocabulary section. This way, the imported content is “space-less” (as it should be).

Another thought just struck me - how does the tool count words? Is it a word only if/when there are spaces? If so, perhaps the “best” thing is to import content with spaces (for the tool to calculate the proper number of unknown words).

Right now the system counts words wherever it inputs the spacing which, in Chinese and Japanese is incorrect. As you say, if you do import the text with spaces, the system will calculate the word count correctly. Otherwise, the statistics will not be correct. For many reasons we need to figure out how to split Asian texts into words. We hope to get some time in a few months to work on this.

Until then, I think spacing the text is a good compromise and shouldn’t be affected by any future changes we make.