Feature request (various things)

[shaluig]

Hi,

Just another suggestion (android) : unless I miss something, it seems to me that the simplified/traditionnal switch is not
immediately effective, especially when reviewing flashcards.

I will try to explain more precisely : this morning I was reviewing an ancient flashcards list, but this time I chose to do it in traditional characters.
On the first card I gave a wrong answer ; at this precise moment I changed from trad to simplified to help me recognize the char, but on the screen the display had not changed, in fact it needs at least one finger action to actually switch.

I know this might not be simple, but would there be any way to change the display...immediately, right after the time you switch ?

 

[mikelove]

Possibly - may turn out to be a complicated thing we have to save for 4.0 e.g. if it impacts fill-in-the-blanks scoring.

 

[Chris 詹艺豪]

A feature to "pair" characters...

E.g. I often confuse, say, 令 and 今. If one of those comes up in my flashcard test and I guess wrong, then in a way I have actually gotten both of them wrong. If there was a way to reset the other card to 100 (or apply some other kind of penalty) at the same time on the occasion of a wrong answer then it should ensure that the two cards are always shown at around the same kind of interval.

 

[pdwalker]

Chris,

Is it really necessary? If you get one wrong and the other right, why do you need to see the correct one again so soon? And if you do then later get it wrong, it's frequency will be reduced so you'll see it again sooner the next time.

 

[Chris 詹艺豪]

Chris,

Is it really necessary? If you get one wrong and the other right, why do you need to see the correct one again so soon? And if you do then later get it wrong, it's frequency will be reduced so you'll see it again sooner the next time.

If there's a 50-50 chance of me being right, then it's not outside the realms of probability that I could get A right 3 times and B wrong 3 times. This could give me the false impression that I know A where in actual fact I have just been lucky.

 

[mikelove]

I'm a little wary of this too - wouldn't actually be that hard to implement, but it's obscure / subtle enough that I'm skeptical many people would use it.

Wouldn't it perhaps be more useful if we did something specifically for easily confused characters? For example, we're considering an option to explicitly attach a list of incorrect choices to a card for multi-choice testing, so you could attach 令 as an incorrect choice to your card for 今 and then when you ran a multi-choice test drawing its incorrect choices from there you'd see just the definition for 令 and the definition for 今 to choose from. (or, if you ran a test prompting for characters, you'd see the definition for 今 and then 令 and 今 as your two choices)

Easily confused characters are one of the few places even I'd admit there are clear pedagogical benefits to multi-choice testing, since, just like tone practice, it's a very efficient way to hammer on the specific thing you need help with, and this way we're attacking the problem head on rather than just vaguely factoring in the fact that two characters look alike.

 

[Chris 詹艺豪]

I'm a little wary of this too - wouldn't actually be that hard to implement, but it's obscure / subtle enough that I'm skeptical many people would use it.

Wouldn't it perhaps be more useful if we did something specifically for easily confused characters? For example, we're considering an option to explicitly attach a list of incorrect choices to a card for multi-choice testing, so you could attach 令 as an incorrect choice to your card for 今 and then when you ran a multi-choice test drawing its incorrect choices from there you'd see just the definition for 令 and the definition for 今 to choose from. (or, if you ran a test prompting for characters, you'd see the definition for 今 and then 令 and 今 as your two choices)

Easily confused characters are one of the few places even I'd admit there are clear pedagogical benefits to multi-choice testing, since, just like tone practice, it's a very efficient way to hammer on the specific thing you need help with, and this way we're attacking the problem head on rather than just vaguely factoring in the fact that two characters look alike.

That is actually more along the lines of what I was thinking, but figured it was too niche/out of current scope to suggest!

 

[FrancoisTaipei]

A feature to "pair" characters...

E.g. I often confuse, say, 令 and 今. If one of those comes up in my flashcard test and I guess wrong, then in a way I have actually gotten both of them wrong. If there was a way to reset the other card to 100 (or apply some other kind of penalty) at the same time on the occasion of a wrong answer then it should ensure that the two cards are always shown at around the same kind of interval.

Actually I came up with a solution for this: I create a new flashcard with both characters next to each other. So I would have 今冷令 for example. Seeing them together makes me totally remember all of them, as I can clearly see their small differences.
In the end, I don't reset to 100 the single characters, but I reset to 100 that one. I would also reset to 100 the single character if it is prompted and I fail. But I wouldn't reset the other ones.

 

[pdwalker]

I'm late to the party here.

I still think the problem is inevitable self correcting over time.

Francois is suggesting the same idea I just had this morning. Make your custom flashcard with the characters in question (in my case it's 欠...久). Fill in the card details (or use the batch operations to fill in the rest. Then make sure your definitions are correct. You may also consider making a second card with 久...欠. I separate my characters with three ellipses after I've filled in the pronunciation fields to make sure I don't actually match a dictionary entry. My definitions are then "definition A / definition B", and "definition B / definition A". Thus I can test myself even more on those two characters that I mix up (4 cards now, not 2).

I think this is easier and more workable (and faster to implement) than an obscure feature request.

Full disclaimer: I want to see a v4.0 release before Mike is given even more interesting features to implement first.

 

[FrancoisTaipei]

 

[pdwalker]

Yeah, similar to that.

 

[Chris 詹艺豪]

View attachment 2257

Yeah, that's a good idea, combined with pdwalker's idea of reversing each pair (without that, there's a danger of just memorising the sequence without actually learning to distinguish the difference). I currently have a list of these in a CSV file. Is it possible to do a mass import?

 

[pdwalker]

Absolutely. Read the pleco documentation in in importing from text files at http://iphone.pleco.com/manual/30200/flash.html#importexportfileformat

Short version:

//MySimilarWords
欠...久<tab>qian4 jiu3<tab>owe ... duration of time, long time
久...欠<tab>jiu3 qian4<tab>duration of time, long time ... owe

Import this file (replace <tab> with a real tab character) and Pleco will create a new category called "MySimilarWords" with the two cards for you to study.

When you import, or after you import, you can use the Pleco batch command to add the Cantonese pronunciations and traditional characters.

I suggest you don't add them to your user dictionary on import.

/edit/ also, if you have them as a CSV file, you can use Excel to open the file, then create some formulas to manipulate the data and you can then export the data in exactly the format that Pleco likes.

If you're not an excel expert, you can send me your csv file and I'll set it up for you.

 

[FrancoisTaipei]

I think everyone has different ways of learning, so what I'm going to say only applies for me. But I could almost always remember without a need to reverse the pair.
I would say it's more about getting a really high score on the cards. If the card score is less than 2000, it's still too fresh in the memory to be really memorized.
When the score is reaching a high value, for example 8000 or 15000, the flashcard will only show up again after 80-150 days. After that time, I really don't remember what was the order and I just have to answer based on my memory of those single characters.
Today I failed a flashcard which was 均勻勺叼 (jun yun shao diao) - it had like 5000 and I already forgot 3/4 of them. Now it's 100 again, time to study again

 

[xuesheng_01]

I apologize if this has already been suggested. I am a noob and don't have time right now to dig through the forum. Anyway, it would be convenient to have left- and right-arrow buttons on the main dictionary screen to allow rapid scrolling back and forth through recent search history items.

 

[mikelove]

Sorry, you mean you'd like an option on the search screen to scroll through recent searches you've typed in? You can already scroll through recently looked up words (= words you actually tapped on) in the first tab of the History screen - tap on the first item there and you can use the up/down arrow items to scroll back through other items.

 

[Bvo]

Is it possible to add colorization for chars in examples?
Colors are great help to remember tones.

 

[mikelove]

Already supported, assuming the examples have pinyin attached - option is in Settings / Colors.

 

[Bvo]

True, thanks!
And it will be even better if it will highlight examples without pinyin (hello, NEC).

 

[Peter]

Stoke order page: for multi-char headwords, add an 'auto-next' play button that pause momentarily after the final stroke, then draws the next headword character, and so on. Stop when all the headword chars for the selected fanti/jianti mode have been drawn.