New discovery: alainravet1’s idea… actually backfired for me. I removed a bunch of the reading counts from the statistics… and LingQ is now telling me I’ve read MORE words, not less…
It’s really unfortunately, but there’s zero chance that I’m renewing my subscription next month with these kinds of bugs. Sorry, but when I’m paying for it, I’ve got to choose software that’s not this level of broken.
Is there a time component as well? It seems like if you just page through, they don’t count, but if you spend more time on a page they do?
@zoran could you clarify and explain the way the read words count is currently updated?
I’m happy/have no problem with it, but I almost exclusively use LingQ to import-and-read short (700-2000 words) articles. They are short enough to be read/processed completely.
It’s possible different workflows trigger problems as mentioned above.
Estimating word counts programatically is always going to have assumptions that make it less than 100% accurate.
Is your goal to count words or learn languages?
LingQ works great for the latter.
I’m inspired and motivated when people tell their story here, about their progress, and how they got there, f.ex by reading 1 or 2 million words/year. That means roughly 3000-6000 words/day. I try to match that number.
We don’t need 100% accuracy, but some accuracy is required though.
It works fine for me and the way I use LingQ. It’s possible some don’t share that experience.
There’s always that person who tries to tell you, “Why are you trying to be accurate? You’re just a horrible perfectionist.” and I really wish they’d just stop attacking people for wanting things to work somewhat as advertised. If the software is supposed to count words read, then it should count what I read. I’m not asking for something unreasonable.
Yeah, I’m not even expecting 100%. I just want it to be accurate enough to be close and flexible enough to allow us to fix things when they get messy. I’ve made suggestions aimed towards that goal.
That’s not what he wrote. 
Where does LingQ advertise their software to know exactly how many words you have read? It keeps a statistics based on the information available. As you don’t tell the software where exactly you stopped reading midst a text, how is it supposed to tell the exact word count?
You haven’t explained why this is such a crucial information to you, just repeatedly stating that it is. And if you go through this thread, you may notice that almost all posts are by you, and that other persons might partially agree, but don’t consider it such a tragedy. So in that regards you are a little bit unreasonable, to be honest.
After I saw this post yesterday, I tested it, and everything appeared to function as I expected. I read a short 412-word lesson, and my LingQ word count matched exactly 412. I don’t see this as a bug
If you spend a minimum amount of time on a page, the words displayed are counted as read. It won’t count words if you’re just flipping through, but it will if you actually spend time on the page. It’s not perfect, but it seems a reasonable proxy to me. You can also manually adjust the count right down to the exact number of words you want.
I also find others’ stories of achievement inspiring and use “words read” as one of my key metrics. Still, if our goal is to actually learn a language, I think the most important thing is maintaining the consistency and volume of effort over time that gets you to the milestone, and not the milestone itself. Hitting each 1 million word milestone is great, but in my mind it doesn’t diminish the effort if, in reality, it were 0.9X million or 1.1X million. Not all words read are equal in impact anyway, and there is always more to learn!
Happy learning!
I asked our team to investigate this further. Thanks.
Thanks, also if possible please add a Word Count of completed materials stat.
This should just show the words in all lessons marked complete (finish lesson). It can serve as min word read.
There is no way to determine if you are actually reading text or not!! Maybe if your computer had an eyeball tracker it would be possible, but otherwise, how can your computer know what you are doing? All it knows is you opened a page.
If you open the page again, there is no way for the computer to know if you read it again or just opened it, so it double counts. There is no other way to do this with current technology other than having you confirm somehow that you read each page.
I don’t see it as a problem. Your “words read” will be a few percent higher than actual, but so what? Any exposure to the language is beneficial in ways we don’t always know. Whether you’ve read 100,000 words or 120,000 words doesn’t really matter. The number is just a number. What matters is that you read as much as you can and read consistently.
Development attention and time is a limited resource, and adding my workflow to the pile: it doesn’t look like many people would see benefit from this manual tracking approach.
I think Lingq doesn’t work like that because, as others have said, the philosophy is that humans don’t acquire languages by consciously “finishing” material.
It’s a point Steve mentions explicitly and repeatedly in his videos: it’s a trap to try and “finish” content, you need to “graze” over it multiple times and absorb a little each time.
edit here it is: https://www.youtube.com/watch?v=-zk_TJYanAI
he says be like a lawnmower (lol).
You’re right! The computer can’t tell if I’m actually reading or not. So, as per the solution I gave, I would like manual control over it. Also, mine was giving me double the words read - 100% added is a lot less than 20%. I wouldn’t find either acceptable, mind you.
No offence, but please don’t try to tell me how to learn languages. I know the process very well after having done it numerous times over the last 20+ years of language learning.
I’m here to see that my word counts are accurate; nothing more. I also tried to come up with a solution which would mean the smallest changes possible and little to no change for those who would keep using the automated system as-is.
@Heshel-C Please check if it’s better now. We improved the reading count process.
Sorry, but I can’t see any changes. Every problem still exists. These automated systems seem to cause more problems than they solve.
I appreciate your attempts to fix the issue, but I’m afraid that I can’t continue using LingQ.
@Heshel-C so sorry to hear that. The reading count is a complicated feature, there is no perfect way to count, as all users reading speed is different and there is no technical solution to understand do user really read it or not. I really appreciate your activity and feedback. Hope you can stay with us and we will improve LingQ together.
The automation causes the problem. Being able to turn that off solves the problem entirely. I can’t pay for software that doesn’t work.