it happens when i’m memrising between various other tasks, that i pick up the wrong Hanzi when typing Chinese
does not matter, it seems, if i type 提 instead of 替 (both ti-s), Memrise says is correct. Quite upsetting. This happens usually for individual characters, but today it happen even for the wrong “bus” …
plus speed review does not decrease the number of items left for review in some 50% of the cases
thanks!
I had something similar in Hindi were it accepts wrong answers as long as the consonants are correct (Hindi is a language where the vowels are added to consonants to form one glyph). It seems that in either case Memrise is simply comparing two Unicode strings but adding some logic to it (probably for a good reason, like accepting answer that have different Unicode values, but are the same when you read it).
yeah, I know about that, that is why I never took courses where one has to type in hindi (btw, many misundestandings would be avoided if you’d take a look at the respective profiles of the people you are replying to… 
; for example, maxine is a long-term user of memrise, she knows all the things you mentioned in your reply to her). I don’t think however that the these two problems are one problem in fact. In the case of mandarin, recognition of characters functions perfectly for everything ≥ 2. I think is has sth to do with “interference” from the pinyin column (present by default in any English to Mandarin course)