Would it be helpful to have options to see word-by-word translations after answering, for example
Los puntos de vida sólo pueden ser restaurados con una manzana dorada.
The | points | of | life | only | can | be | restored | with | a | apple | golden
and/or transliterations (likely just for some languages, for example Russian, Chinese, languages with different writing systems than the base language), for example
Я ему пока не сказал.
a emu poka ne scasal
(also mentioned in Help us improve Clozemaster! What would you most like to see added, changed, or improved? - #63 by Siffi)?
These would be toggle-able the same as the existing translation and pronunciation fields.
(for Russian, Chinese, Japanese, etc.)
As I explained in a comment on the other post, languages like Russian and Hebrew, which use nearly phonetic writing systems, are in a very different category from languages like Chinese, which use ideograms. For Chinese, English transliteration may make sense (though I’m not speaking from experience, since I don’t know Chinese). But for languages like Russian and Hebrew, transliterations using the Latin script are clumsy representations that lose a lot of information and actually are harder to read once you’ve gotten familiar with the writing system (which doesn’t take long, given that the number of letters in the alphabet is roughly the same as in the English/Latin alphabet). So I think the decision would need to be made on a language-by-language basis.
My own experience is just with a single language. I did Duolingo first. Even though I’m still a relative beginner, I know the majority of words in almost every sentence (“she”, “did”, “was”, “not”). So only single-word translations tend to be helpful, and I can and do get those with the Google Translate link. If it’s not a single vocabulary word, it’s usually an idiomatic use of a verb or two - Hindi has a lot of helper verbs - or an idiom. Word-by-word won’t help for that.
Word by word might help a complete beginner, but I’m still not clear on how CM could be useful for a complete beginner, at least for the languages that don’t have FFT and stuff. This goes even more so for alphabet issues. Maybe others have very different experiences than I do, so toggle-able would be nice. But I would still prioritize other features over this one.
Personally, I would find word-by-word translations helpful as an option, especially for languages that use very different syntax rules, like, compared to English, Dutch or Czech. When I think back to the early stages of learning language x, y, or z, I know I often would remember word-by-word translations of different sentences from those languages, to the point where I could apply their syntax rules to any English sentence (to the extent possible). By that point, the syntax rules would be internalized and eventually wouldn’t require extra processing time in the second language. The same applies to idioms: if I can remember them as is in my native language, then I can definitely remember them in the other language. [Edit: Oh wow, I think I just saw the implementation of this for Russian in another post and it is very cool!]
Agglutinative languages and and/or languages that use particles would present some implementation challenges, but, excluding Korean, they should be relatively easily parsable.
Transliteration I’m not a huge fan of; however, for languages with different writing systems and no audio, there you have a good case for it. They would be helpful for any language that uses abjads, too, in my opinion.