Hello to everybody. I’m a novice here and this is my first post on this forum. First of all, I must say Readlang is really a great platform. I discovered it via Google AI. Thanks to the creator(s) a lot!
I wonder if it was possible to add a tag called “grammar” for articles focused on a grammatical feature.
Here is one more thing: the generated levels often don’t match the content. For example, a difficult article is often marked as the A1 level and the other way around.
Dificalty level is quite subjective, I believe redlang uses official formulas to calculate that (involves number of words per sentence and a few more parameters, afaik)
Regarding text types, may be custom subtypes can be beneficial, but I believe less is more in this case. In other words grammar is nonfiction, and it’s enough for me.
Disclaimer, I’m just a stranger on the internet, above is my personal opinion :).
Thanks for the kind words.
There’s no “grammar” category at the moment. Texts in the public library have one of five categories, picked by whoever uploads them: Fiction, Non-Fiction, Song, Dialogue and Other. If you want to keep grammar focused texts together, you can put them on a shelf: go to Library, then My Texts, tick the texts you want, then “Arrange on shelves…” and “Add new shelf…”.
Regarding the levels, @r.twelve is right that it’s a formula rather than AI. It combines the Automated Readability Index (based on average word length and sentence length) with the percentage of words in the text which are among the 2000 most frequent words in that language. To be honest it does have a lot of room for improvement. There’s more detail here: How does Readlang know the difficulty of a text?