Feature Suggestions

mikelove

皇帝
Staff member
stephanhodges - the problem with allowing card management while a saved session is present is that in order to get sessions to save / restore quickly we basically just save out a list of all of the cards that are included in that session, rather than querying the database again when the session is resumed - that's nice and fast, since you're just saving out a list of card IDs / scores / etc to a file and reading that list back into memory, but it leaves the system without any way to accommodate significant changes in the database; if a card's score or category assignments change, or for that matter its headword / pronunciation, the system has no way to know about that without going back and re-checking all of that information in the database when the session is resumed, which would make resuming a session take even longer than starting a new one. Adding a way to change the default category with a saved session present might be doable, as I mentioned before, but re-engineering the system to allow full access to Manage Cards / Manage Categories would be a massive undertaking and it's unlikely we'll do that anytime soon.

With the way you're studying I think a better approach might just be to end the session when class starts and start a new one after class. Using Card Filter, you can review cards you answered incorrectly in the pre-class session (last reviewed after XXX, last answer incorrect), then review cards from that same set that you haven't studied yet (last reviewed before XXX). Or just split the chapter into two categories, review one set before class and one after. You could also try making all new cards Uncategorized (or putting them in a separate "New Cards" category) and figuring out where to file them after class, though I can imagine it might be hard to remember where you wanted to file them later on.

The Pinyin keyboard is indeed superfluous now, but there are a number of improvements we're thinking about making to keyboard input that would make it less so in the future. Certainly adding support for some European keyboards would also make sense, particularly for German since right now there isn't really any good way to input an ß for example in a HanDeDict full-text search (you can enter them as "ss" but that's a bit awkward).

And yes, a better duplicate card message would also make sense - we thought about adding a modal alert for those, actually (with the option to override, as in Edit Card) but decided against it since it seemed like it could get really annoying for people who encounter a lot of duplicate categories.
 
The main point with putting the European letters on the keyboard now (before release) would be to empower more non-English only users to start using and entering things. It's a small step, but I think you'd get a big response, out of proportion to the changes made. It would be signaling an interest in adding/supporting more Chinese-xxx dictionaries.

Personally, I'd like to be able to use Pleco with Korean, but that's a different direction :)
 

ipsi

状元
Korean could work in flashcards on Windows Mobile (I'm assuming - have only tested it with Japanese), provided you have some way to enter it. Depends on if Plecos font files support it. Wouldn't work as a dictionary entry (again, assuming. Tried with Japanese and it didn't work).
 

mikelove

皇帝
Staff member
Korean won't work at all, unfortunately; the fonts don't include the necessary characters and there's no way to add them.
 

goog1e

举人
Hi Mike,

Would it be possible to add the sketch box to "Review" mode in flashcards?
I like to review new cards several times before including them in my daily review and being able to write them to practice would definitely help. Currently I just do self-scored on the new categories (I organize my cards by date encountered) several times and then reset their score when I think I'm ready to add them to the big mix. Here's a related question: assuming I did use the Review mode to learn my cards (which doesn't change the score but does increment the review number), would the somewhat high review number but low score throw off any of the srs/frequency-adjust code? Does "# reviewed" factor into any of the equations?

Also, what do you think about having a Notes field for each flashcard that you could easily edit during a flashcard session (but after you reveal the card)? Now this feature, I'm not too sure how many people would use besides me. But personally, I am working on "formalizing" some mnemonics for really tough words I just keep missing and would like to write them somewhere. I'm kind of reluctant to add them to the definition b/c I couldn't test by definition anymore.

And finally, not a feature request but a question:
How does Pleco choose the order of flashcards that are due for review on a certain day (using the repetition-spaced mode)? Sometimes I feel like words with similar pinyin (and maybe other properties?) are grouped near each other... I am wondering if you added in some code to try to minimize "distance" between cards to make things more challenging or highlight similarities in cards.... but, I review enough cards each day to be bound to get some similarities, so perhaps it's just my perception.

Thanks!
 
Hi Mike

I Would Like The Ability/Option To See A Definition
With The Character Presented In The Stroke Order Test.
I Think This Would Really Help With New Characters.
This Could Just Be The 'Char Info' Definition, Or Better Still A Pop-Up Dict One.

Another New Feature Suggestion,
Would Be Some Sort Of Multiple Character Stroke Order Test.

TIA
The Duelist
 
I keep thinking of this, but I don't recall ever posting it...

Add a tab to generate a complete "pleco system" inventory information file that could be later uploaded for tech support, etc. Of course, even better would be direct uploading, etc. but that could be a second option, since many devices won't have direct connection capability (and of course many do).
 

mikelove

皇帝
Staff member
goog1e - it's certainly possible to put a sketch box option in review-only - actually review-only and self-scored seem to be sort-of converging, already a Self-Scored test with Scoring Type set to "None" differs from review-only only because it has that extra Reveal step. The # of times reviewed doesn't factor into the algorithm at all, at least not at the moment - the last reviewed time matters in repetition-spacing but that's about it. If you don't even want that count incremented you could set Scoring Type to "None" - that way even the reviewed count won't be incremented.

Annotations were a feature that was considered but dropped for 2.0, but they should definitely be appearing at some point in the future, both for flashcards and for dictionary entries. The order of flashcards in repetition-spacing is completely random, it shuffles the whole list before it starts the session, so I'm not sure why you'd be seeing words with similar Pinyin grouped together - probably is just a coincidence, yeah.

The Duelist - Good point about showing definitions, you might as well be getting that reinforced while you're staring at the character anyway. Though it might be hard to squeeze that into the interface. Showing all of the characters from a card I'm a little wary of, because you'd end up studying a lot of characters you'd already learned well, but some sort of system that hammered in the patterns for particular radicals / components / etc by showing you a lot of similar characters might be interesting.

stephanhodges - a System Info screen would definitely help with error reporting, yeah, though we probably would need some sort of online component in order to get people using it.
 
My idea was to actually create a small file that would be transferred to the computer on the next hotsync, so it could be emailed, but yes, a screen to see what the info is (and a button to create the file) makes even more sense.

If you do that, add something that could capture the current state (what screen people are on, etc), and also add a "feedback" note writer capability. You'll get some good feedback and bug reports that way.
 

imron

举人
I haven't read through this entire thread, so I'm not sure if this has been brought up already (I did a search for audio though, and my suggestion didn't come up so hopefully I'm not repeating something).

Anyway, currently in order to change the audio speaker between the male and the female, you need to go into preferences and change it. What would be nice is if you could just hold the audio button, and then it came up with a choice of what speaker you want to hear (similar to how most of the other buttons have multiple options when you hold them). That way it would be convenient to switch between multiple speakers.
 

imron

举人
Once again, apologies if this has been brough up already (I did a search for popup definition and my query didn't come up so hopefully this isn't a repeat).

I use the Chinese-Chinese dictionary. Occasionally, when looking up a word, I'll not know one of the words in the Chinese definition. So, I'll highlight and bring it up in the popup definition box (currently I have this mapped to tap-selection) and see what that word means. So far so good. However what also tends to happen is that in the popup definition, there might also be a word or two that I don't know. Currently however there is no way to get the definition of that new word to display in the popup definition box. I have to first copy the word, close the popup definition box, paste/type the word into the search box and then I get the definition.

It would be nice if I could just highlight a word in the popup definition box, and then use tap-selection (or tap-hold, or whatever) and have popup definition box change to display the definition for that character. It would be even better if it kept a mini-history stack so I could go back to the previous term once I understand the word.
 

goog1e

举人
Sometimes I'd like to add a phrase to my flashcards but the pronunciation switches to syllable-by-syllable since the entire phrase cannot be matched to an existing sound file/dictionary entry. I understand the difficulty in putting together some software that tries to parse sentences and pronounce phrases properly -- I think people spend their whole lives in linguistics doing that kind of stuff. But what if Pleco would follow spaces in sentences (manually added by the user so pleco wouldn't do any work at all) and sentences were treated like a space-delimited-list of words? Each word separated from others by two spaces (and adding some case for beginning/end sentence and punctuation) would be looked up and pronounced separately from all other words in the list (sentence). So if some words were recognized by pleco, they would be pronounced with the multi-syllable speech samples even though others may be pronounced syllable-by-syllable. The space (or whatever symbol used) could be stripped out of the final displayed text so it would be transparent when seen by the user. This could even be integrated into the reader and support texts that had been annotated word-by-word by interested people or groups.
 

mikelove

皇帝
Staff member
stephanhodges - well a file could work too, but it's difficult to talk people through digging up a file in their HotSync Backup folders (and harder still to talk them through pulling it off of a Windows Mobile device) so internet sync would probably the best bet, particularly as we're already planning some sort of internet link for downloading flashcard files etc in 2.1 (mainly for the sake of the iPhone version, since there's not really any other good way to get those sorts of files on an iPhone).

imron - I think it might have been mentioned before, but thanks for the reminder - really both volume and gender should probably be selectable from that popup list. Though how much work we put into streamlining speaker-switching will depend on how well we do fixing up the female files - now that noise reduction is looking less and less likely I'm not sure if they'll ever be more than just an optional add-on.

The popup definition screen is probably going to get both a Copy to Input button and its own Input Field in a future version (actually more for the sake of Instant Access than use within Pleco though it would certainly be handy for that too), so cross-referencing should be covered by that.

goog1e - interesting idea, but I'm not sure how accurate that would sound; there are some changes in pitch / intonation that apply based on where a word is in a sentence, and tone sandhi transformations can easily span multiple words. Really if we want to start pronouncing sentences we should just implement a full-fledged text-to-speech system, we'd need some new / halfway-pitch-adjusted recordings and some good algorithms for breaking up words / applying sandhi transformations but I think with a little work we could come up with something that's nearly as accurate as a commercially-licensed one. (or we could just license one of those, but I don't know if we could get it ported to any mobile platform other than WM)
 
I'd really like a feature to be able to toggle on/off the pronunciation, similar to how one can switch to large/small characters(which I use a lot). I find that I "instantly" see the pinyin, and it's a real effort to try and look at the hanzi first, and impossible to not see the pinyin.

Turning it off is great because it really shifts my mental focus into "character mode" when looking at words, but I'm really not that good at it yet, and would still need to have it on a lot of the time. A pop-up would not be useful at all, nor is going to the customization menu every time to switch things...

I'd want to be able to assign to a button too :D
 

imron

举人
Though how much work we put into streamlining speaker-switching will depend on how well we do fixing up the female files
I would be happy to have this option available even with the "unfixed" files from b7. Changing volume is not so much an issue for me.
 

mikelove

皇帝
Staff member
stephanhodges - makes sense, might also be interesting to provide a Zhuyin/Pinyin switch that way (for people trying to get comfortable with both systems).

imron - understood, and I suppose we might have other audio sets eventually that would also be helpful to switch to. (single-syllable sets are pretty easy to record, so if we get some basic tone sandhi transformations working we could add a couple of other speakers with interesting regional accents etc)
 
Hi Mike

Just Wondering What Happened To A Feature That Was In Previous Betas,
That Allowed Highlighted Hanzi To Be 'Read' Using The Audio Function.
I Liked This Rudimentary 'Speech' Function/Option. :wink:
Now Audio Only Works On The Headword. :(

The Duelist
 

mikelove

皇帝
Staff member
There's never been a feature like that, actually - the closest equivalent is to bring up a popup dictionary definition on the highlighted text and then tap on the Audio button in there.
 
The Duelist said:
Hi Mike

Just Wondering What Happened To A 'Feature' That Was In Previous Betas,
That Allowed Highlighted Hanzi To Be 'Read' Using The Audio Function.
I Liked This Rudimentary 'Speech' Function/Option. :wink:
Now Audio Only Works On The Headword. :(

The Duelist
mikelove said:
There's never been a feature like that, actually - the closest equivalent is to bring up a popup dictionary definition on the highlighted text and then tap on the Audio button in there.
Must Have Been A Bug. :lol:
(A Good 'Feature' Though :wink: )
 

Alexis

状元
HI Mike,

I really like the full text search!, But, I thought the "full-text" search wasn't working for the longest time because I was trying to search for an english word in the chinese-english dictionary (I since discovered that it only works in the english-chinese dictionary.). What originally excited me about this feature was the ability to search for English in the Chinese-English definitions (to see sample sentences containing the words, definitions containing the word etc.). Is this implementation planned for a future version (ie. 2.1)? Same wish for Chinese full-text search in the english-chinese dictionaries.

Thanks!
Alexis
 
Top