How difficult would it be to add also a unique words count to the stats below at the end of each text? It would be a great way to track one’s progress. If the text had 500 unique words and I looked up 100, that’s a very different comprehension than if it had 1,000 unique words and I looked up 100.
I know there’s complications when it comes counting unique words, but I guess any way of doing it would be helpful. This is especially true when reading longer texts.
If I read a 200,000 words book, whether it have 10,000 unique words, 20,000 or 30,000 makes a big difference.
This text was 1728 words long and while reading you translated 234 words or phrases.
Related to this topic, it would be great at some point to build some sort of dashboard that tracks the words read.
For example, at the end of a text, I get a number of the words I looked up. What if look up a word five times? I assume that counts only once? But it would be very helpful to know how many times I looked up a word. And do this across texts.
I would also recommend extending this to the words not looked up. I use other methods for learnings. I’m going through 3 editions of Assimil and 3 Linguaphone multiple times, so I’m absorbing vocabulary outside of Readlang as well. In such a dashboard, I would include any word read on a page as a known word. If didn’t know it, I would look it up.
If I read a text that has 1,000 unique words and I only look up 200, that means that I know the 800 I didn’t look up and I’m learning 200. If I encounter some of those 200 that I looked up in future texts and I don’t look them up, that’s because I remembered them, so I would add those to the known words list as well. If I look it up, then it could go to the learning list.
I guess what I’m suggesting is a way of tracking know words as well, not just unknown.
And if we have those stats, it would be great to maybe plot them, make a bar chart with the distribution or something like that.
There are all sorts of benefits of having something like this.
For example, if I upload a text and I see that it’s a C1 level, that’s extremely useful. But what does that mean for me? It could be a C1 level with a vocabulary that I know 80%, or it would be a C1 with a vocabulary that I know only 20%. Keeping track of the words I know could allow us to quantify the % of new words in a text. Maybe when looking at the list of texts in my library, next to the difficulty level, like C1, we could have a % stating the number of uknown words as a % of total unique words.
The way this could help tremendously would be to help me select what to read. If I have a list of texts, maybe I start with the ones where I have fewere words I don’t know and work my way through the texts that way. This allows me to read faster and also to acquire new words from context as well.
For those of us who want to do Extensive Reading, this would be a huge beneift. The problem with selecting texts for extensive reading is that one never knows the % of unknown words in a text. What I’ve heard is that for this method to work, there can’t be more than 2% new words or comprehension drops. But the problem is that nobody can know what that number is especially that the number is dynamic. As I’m acquiring more vocabulary, that number drops.
With Readlang we basically have all the data to actually apply this method. And I assume it can’t be that hard. Just count the unique word and keep track of them in a database. To make it very easy, we could assume that every seen word is known. If I turn a page, every word that I didn’t look up on that prior page could be known. Take that list of unique words, compare it with the list of all my known words and all it to that list if it’s not on it. That would simply things considerably.
Of coulrse, we could get fancy with this. If I know the meaning of a word, that doesn’t mean I know all its meanings. One way to take this further would be to count not just the words but the meanings as well. But this would be much harder. Even if we were to keep it very simple, it would still be of tremendous value.