The reason for this is that any edit to the source word risks breaking the connection with one of the contexts and I didn’t want to deal with the complexity of designing a UI for that for the first version of this new edit view. I’m not happy with the current situation though and agree that there should be a way to do this.
Thanks! But TBH I’m not sure it’s sustainable to respond to every comment and I worry I’m setting an unrealistic precedent. I did read the suggestion and used the heart button to “like” the comment. I may have to do more of that going forward since I think most of my time should be devoted to directly improving the product.
Is there a way to change AI context explanations to another language in practice mode? I prefer monolingual explanations and don’t see a way to change that.
This keyboard shortcut only applies within the Classic Flashcards for now. See list of shortcuts here: List of keyboard shortcuts in Readlang. I’ll consider adding it to the new practice modes too, perhaps as Ctrl+D and with an ability to undo in case you hit it accidentally.
That option in the settings applies to the Classic Flashcards only. If you want no typing with the new practice session, you want to select “Blitz” mode.
Good idea! The thing you’re remembering is an experiment I ran briefly which interspersed mini practice sessions throughout the texts you read. These included a match challenge as you’re describing plus a comprehension challenge. They were generated by AI and I killed it partly because the multiple choice comprehension challenge was often of low quality. Bringing back the match challenge either as a new more, or just as an additional challenge type within the current practice sessions, could be a good idea. However one thing I like about the new practice session is the integration of the edit/explain mode to help ensure that you are learning the correct thing, and I’d need to give thought to how the UI for that should work when there are multiple words within the same challenge.
I like the idea, and have thought about it, and it’s obviously doable as you say, but not trivial. I opted instead for the different but related and easier to implement auto-highlighting feature: Auto Highlight Words As You Read - Readlang Blog. Two of the things that make what you’re asking for tricky are:
The reading UI is the oldest and messiest code within Readlang so any change risks breaking stuff. I hope to gradually improve this over time and may even do a rewrite at some point! Adding more options which non-trivially change how the page works is particularly risky because it makes it easy for bugs to go unnoticed as number of configurations that need testing increases.
Readlang actually doesn’t record the positions of your saved words right now, only the contexts. The reason that recording positions is tricky is that you are allowed to edit the text after saving a bunch of words, which could then invalidate the positions of the words you’ve already saved. There are potential solutions: e.g. I could use the saved contexts to infer exactly which words/phrases you clicked on but this would need extra work to figure out how to do it in a computationally efficient way.
Also, to keep the discussion civil: please avoid shouting with ALL-CAPS, and saying things like “playing the idiot game with me, idiotically”.
You can actually access a Delete button by clicking on “Explain / Edit” and then on the “Delete” button which appears. I agree this isn’t obvious at the moment and will think about improving the UI.
Yeah, this isn’t supported I’m afraid. You have to pick one answer to type, if you try to type multiple synonyms in the same answer it will grade you as incorrect.
Thanks for the suggestion! I’ll consider this. One of my goals with this new practice session is to keep the UI as simple as possible so I want to really consider how to add new features without cluttering it and making it feel complicated to new users. The reason for showing explanations in your first language (only English for now) is:
When I defaulted to showing the sidebar explanations in English instead of your learning language, usage went up a lot, indicating that people find this more useful. (This applies to me too! I initially thought that I wanted them in Spanish for more practice, but in the end I much preferred being able to scan them very quickly in English saving my mental energy for understanding the Spanish text I’m reading)
The practice sessions, for better or worse, are very focussed on translation to/from English right now, and one of the main functions of the explanation is to check that you have a correct translation. An explain in your learning language would be far less likely to include an English translation so it’s not as useful for that purpose. And with the English explanations it’s super-quick to add English translations to the card by clicking the little inline buttons that appear.
All that said, I’m not totally against adding the option for monolingual explanations sometime in the future.
Can you change how clicking on the edit button adds an explanation without me asking for one? What happens is that I realize while practicing that the translation is wrong in the first place, so then the AI explanation ends up adding another wrong thing lol
I’ve found the explanations to be very good in Spanish, but I’m sure it’s worse for some languages. In which language(s) have you noticed that the explanation quality is bad?
I meant: I’m impressed with the speed new features are implemented and bugs are resolved. Overwork and burnout are a very real danger on a project like this so I expect that you’ll have to slow down for the long term, yet I don’t think it’s a reason for worry – “the product” is near perfect and, speaking for myself, I’d be as satistied a customer even if you went ahead at a fraction of the current speed.
I just fixed a bug where if you clicked the “Skip typing for …” checkbox during a Mastery session it wouldn’t actually take effect during the current session, only for subsequent sessions. I think this is the bug that you were reporting but when writing my earlier response I didn’t fully understand.
I started using the new practice session (blitz mode) again yesterday and I have two suggestions regarding a couple of features I miss from the traditional flashcards.
Display the context sentence in a box to make it easier to spot and separate it from the question and clues. (in a box as shown in the traditional flashcards or in some way to make the context salient).
Retain the ability to click/highlight words/phrases in the context sentence and get inline translations, similar to the read UI or the browser extension.
I considered whether this is just me resisting to change or to adapt to these changes. However, I believe these two features are genuinely important, and I will try to explain why.
Saving the context is one of Readlang’s key features, and it is well supported by scientific research. Keeping the context salient during practice is, IMHO, important. It make it easier for the user to focus on the context and helps the user to attempt to retrieve the information from memory before looking at clues. Similarly, when a user reads the context but cannot remember a few key words, it may prevent them from recalling the word being practiced. Allowing the user to quickly refresh those words can help form the necessary connections in the brain to retrieve the missing word before resorting to the clues.
Like Eduardo, I find the ability to click on words in the context missing in the two new modes. Another thing I miss is the ability to edit the whole flashcard, including the keyword (at least regarding capitalization) and the context(s).
It’s a bit clunky having the two different ways of editing a word (since there’s still the EXPLAIN/EDIT button at the bottom to access the other mode), but at least now it’s possible to edit the word and the contexts from the new practice session.
Instead of choosing numbers, how about using the words themselves, like duolingo? i mean when i want to answer i need to press 1, 2, 3 or 4. but it would be useful when i press first letter, then second letter of the word/sentence
