I’m now on a sixty-five day streak with my Chinese learning (woohoo!) and I’ve been using Google Sheets to closely track the amount of LingQs I’ve made compared to the amount of words that are marked as “known” in LingQ.
I am very religious about moving LingQs that I’ve made closer to the sacred “known” checkmark. I only check known word if it comes to me with the proper pronunciation and I know what it means.
I never click level four on a word because I know that marking a word as level four marks it as known (and I don’t want a word I don’t actually know fully to show up in known words).
Over the last couple of days, I have done more or less the same amount of LingQ’ing that I’ve been doing for the last sixty-five days (although I have increased my listening, but I don’t really change words’ statuses when listening).
Yet for some unknown reason, my known word count has increased dramatically. I am sure that I have not been marking more than five to ten words as known per day, and yet, somehow LingQ is saying that I’ve learnt thirty three words since yesterday and thirty five words the day before that…?
Now, here’s something interesting: if I filter my vocabulary by known words only, it displays that I have only four hundred seventy nine known words. How could that be?
I guess I am super confused as to how the known words count works in LingQ. Do I not have to tap/click “known” to mark a word as known? Where are these mystery words coming from?
Have you read/listened to new lessons during this time? If so, every time you turn a page (proceed to the next screen in the lesson), LingQ assumes that you “know” all the words on the page you have just read and/or listened to unless you specifically made them into Lingqs (yellow). Any previous blue words (which the program considered unknown according to your history on LingQ), will now become white (known) if you do not specifically turning them yellow.
If you look/listen to material away from LingQ or knew words before you subscribed to LingQ, this may account for the additional known words that you did not deliberately create from previously LingQs but which the program is now tracking. Keep in mind that the program also tracks words in the title of a lesson and proper names which you may not think about making into LingQs. (Some people make LingQs of proper names, others don’t.)
If none of this applies, then I too would be mystified.
I turned off the “Paging adds to known words” setting on both the web client and iOS client, so that can’t be the cause of it.
I had learnt some Chinese before I came to LingQ, but I make sure to mark those words I already knew as known right away. Anyway, I stopped seeing words I’d already learnt about a week ago, as I’m further ahead in LingQ now than I ever was beforehand.
Most of my lesson titles are in English, as I mass-imported over three thousand Chinese articles using the LingQ API but kept their English titles. I’ve never heard that LingQ tracks words in the titles of lessons, how would it do this? I have no way of interacting with the contents of a title, so that doesn’t make sense.
Oddly enough, I just went to the vocabulary section and filtered for known words only. What I found surprised me: there were only four hundred seventy nine entries in the list. I have attached an image to my original post displaying this.
I don’t have an answer but I’ve wondered about the mechanics myself. Would be very curious to see an official reply on this, especially as the OP has gone to the trouble of including a detailed description of the problem including screenshots.
I would also like an official explanation as to how the total of known words posted next to the language flag at the top of the page is calculated. For Norwegian I did a select of all vocabulary records tagged as 4 (learned) and check mark (known) and the total is 3,554. But the total next to the Norwegian flag is only 3,381. Why is there a shortfall of 173?
The “Vocabulary” section only shows words you have previously lingqed (i.e. they were among the yellow words at some point) and later moved to known. It doesn’t show words that were marked as known without creating a lingq.
I guess those are longer phrases. When you create a lingq by selecting several words, they do not count in the word count (because they are not really “words”), but they show up in the vocabulary section.
While true, that does not explain this type of scenario:
Sometimes, I open a lesson that has a few blue words. I make LingQs from those words (not marking any as known), and then complete the lesson once I am done with it. I never marked a single word in the lesson as known. Not one!
Yet, after completing the lesson, I suddenly have more known words than I did before…?
This kind of scenario has happened to me on several occasions (even today). I am going to try and get a screen recording of it happening soon.
What I meant in my reply by saying “I make LingQs from those words” is that I make LingQs from those words, the blue ones. So that BEFORE I click on the complete lesson button, there is not a single blue word (and yes, I am sure, my lessons are not immensely long yet).
I think there is a bug that allows a lesson to contain “invisible” words (probably due to some formatting marks that weren’t processed correctly when the content provider imported the lesson). If that happens, you should still see a non-zero number in the blue circle next to the progress bar at the top.
Pretty sure I would be in legal trouble with one company or another if I shared my code, sorry.
However, if you want a fun project idea, I’ll tell you what I did:
I signed up for a subscription with The Chairman’s Bao
I built a script in Python using the Scrapy framework to scrape data off of it (lesson titles, lesson contents, image links, audio links)
I then built a script in node.js to run through all of these articles (which I had in JSON format) and download the ones which had valid audio, a valid image, and valid contents (some articles had broken audio or their contents couldn’t be scraped properly)
All of that data was compiled into a massive folder on my Mac
I wrote another node.js script to upload all of these articles to my LingQ account.
The result: around 3100 articles covering society, education, animals, technology, and more. Was easily worth the two days or so that I put into it. I basically never run out of content now.
Haha, thanks anyway I can follow (doesn’t mean I am in any way near being able to recreate) everything but step 0. Bao??
I have hobbyist knowledge of Python, JS, and Java. I had a look at what the API has to offer but so far I couldn’t imagine doing anything worth the effort with it that wouldn’t be possible using the site as it is.