[Unofficial] Feature Request / Suggestion List

jurgen85

探花
MOE's citations are nice and hidable, but for some reason has its own examples inline. Occasionally they are very verbose which can make things harder to navigate, especially on a small screen. E.g. 想 (as it flows on my device):


  1. 思索、思考。如:「想辦
    法」、「想不出所以然」。
  2. 欲、要、打算、希望。如:
    「想結婚」、「想出國」。
  3. 推測、猜度。如:「料
    想」、「推想」、「猜
    想」。
  4. 認為、覺得。如:「你想這
    樣對不對?」、「我想你應
    該回家一趟。」
  5. 思念、懷念。如:「想
    念」。
  6. 似、像。

念頭、意念。如:「夢
想」、「理想」、「不作此
想」。
while the core defintion is much shorter and simpler to read:

1 思索、思考。
2 欲、要、打算、希望。
3 推測、猜度。
4 認為、覺得。
5 思念、懷念。
6 似、像。

念頭、意念。
so it would be fantastic to turn these into examples (and possibly also making citations and examples independently hidable).
 

Knil

秀才
It would be great to have an upcoming review calendar showing the number of cards that are due on upcoming days. My number of due cards often fluctuates wildly (e.g. between 70 and 200+ cards), especially depending on the times of day I do the review, so without being able to check how many cards are really due on the day or in upcoming days, it is very difficult to gauge whether now is a good or bad time to add new cards. (I've found the number of unlearned cards can be an uninformative metric as that can also vary wildly from day to day, especially since I add a lot of new cards so there are a lot that are just in short-term memory). Even having a simple list of the current and upcoming 7 days of number of expected reviews would be helpful!

I saw that there was a feature suggestion for this a while back, but I wasn't able to find information about it since. Is this something that's in the works for the next version of Pleco?
 

mikelove

皇帝
Staff member
It is something you could theoretically get with our new statistics screen - plot due dates on a bar graph - but at the same time, we are actually strongly pushing against the idea of specific due dates for any card that you expect to be in long-term memory (i.e. one that's not recently introduced or recently forgotten) in our default profile. So you can still get it - with the "Pleco Legacy" profile which mimics our old behavior, or, if you import an Anki deck, then with the Anki emulator settings profile we offer for people migrating from Anki, or with a custom profile that uses whatever crazy system you want - but the system we'll be steering people towards pretty much ignores that.

The default behavior now is that all long-term memory cards are placed in a big 'review pool', without any particular attempt to calculate algorithmically optimal intervals for them - we just draw cards from that pool (sorted by approximate date of last review) in whatever number will fill out the number of reviews you want to do right now. Basically the priority is:

- forgotten
- recently started learning
- brand new
- review

So if you're reviewing 100 cards and you have 100 cards in the first two categories, you would not see any new cards or any review cards in this session. If you're reviewing 100 cards and you have 95 forgotten + recent + new, you'd only see 5 review cards. On the other hands, if you have hardly any cards in the first few categories then you might have a test consisting mostly of review cards. The main thing is that reviews are no longer considered high-priority, since with long-term memory (and, moreover, long-term memory of something that you're probably also encountering outside of our app from time to time) there isn't really a clear exponential decay and hence there isn't really a good reason to worry about reviewing something on a particular due date.
 

Knil

秀才
Interesting, I never thought about this kind of system for cards in long-term memory, but that makes a lot of sense. Looking forward to seeing it!
 
Top