Flashcard Difficult to Create

davidy

秀才
First off, I would like to say that I really like Pleco and the enormous gains it gives me in being portable and feature rich. The reader is amazing and all the dictionaries are superb.

I was trying to make some flashcards with just a list of chinese characters and realised how difficult that was. At least compared to the desktop learning tool I have been using - Pingrid (and Pablo)http://ehaton.blogspot.com/. The structure is very different. All I need to do with Pingrid is open the file with the list I want to learn. The pinyin and definitions are automatically taken from CC-CEdict. Where this would differ from Pleco using just one dictionary is that when there are multiple pinyins / meanings. Pingrid just merges all the different pinyin and meanings together to give the character as full a description as possible. Below is Pingrid training the third 100 most frequent characters. All I did to create this file was to copy characters number 201 to 300 from excel into notepad. Open with Pingrid and away with learning I go - at a desktop.

On Pleco, I could choose 'Use First' but this has given meaningless definitions that do not allow me to know which character is meant. Using Prompt would make me go through labouriously through many definitions which I would rather be merged, as the character means all of those and that is how I want to learn it. To make things worse, the definition is in a small area of the screen even when I use the portrait orientation where there is lots of space.

I wish Pleco could just open a list and allow me to spend the time in learning rather than in making the cards meaningful or making this post. Importing is cumbersome. There should be an option to merge pinyins and definitions from selected dictionaries. Just opening a list of characters is quick and efficient - let Pleco merge things from user selected dictionaries.

To solve the problem of user progress data, Pingrid uses a separate simple txt file that list characters that have been presented before with the date time of when they should appear next, and the number of times they have appeared etc. So it doesn't matter what list I open, if there are characters in it that I already know well, they won't show up until the time is due. I could edit the file to bring characters forward of backwards. eg
!newword 1 最近 r
!newword 1 往前 r
!newword 0 多长
!newword 0 很多
2010.03.16'02:28:50 6 医 r
2010.03.16'06:19:46 6 常常 r
2010.03.16'08:09:01 6 医生 r
2010.03.17'06:09:03 7 男 r
2010.03.17'10:03:43 6 爸爸 r
2010.04.05'21:00:35 2 星期

Am I doint something wrong in flashcard creation? I've read quite a lot of the online manual.

Could the flashcards be made easier? ie just a plain list eg











Thanks.
 

Attachments

mfcb

状元
one of the problems you encountered is, that pleco has for each headword (in chinese) just one pinyin transliteration. if there is a second way to transliterate it (even when same pinyin, but different tone), then there is a second dictionary entry (mostly single characters).

there is another "feature" that you might not have encountered yet. in some dictionary even if the pinyin and tone is same, there are several dictionary entries for a headword. and this is not always to just split otherwise very large dictionary entries...

in the beginning this disturbed me also a little bit. but nowadays i think, that it helped me a lot to get through the massive amount of flashcards i have to review each day, hehe. initially i also started like you, to learn the most common 1000 characters, and still i think that i spent most of my time for learning characters on them, after the first 1000, everything went faster. when there were characters that had more than 1 pinyin, i just created 2, even 3 or 4 cards. that way i saw that character also more often and therefore remembered much more easily that there are more pronunciations for it (because its hard to guess which one is meant, when its just a single character). but at least for ABC (and maybe also other dicts) at the bottom of the entry there is a "see also..." for the other pinyin, so you can check if you got them all right...

therefore, to import flashcard lists automatically, you have to provide both, the headword + pinyin. i never made a list like you did myself, each list, that i learned i downloaded somewhere, and all of them went into pleco without problems, and without me manually assigning entries...
 

davidy

秀才
mfcb said:
one of the problems you encountered is, that pleco has for each headword (in chinese) just one pinyin transliteration. if there is a second way to transliterate it (even when same pinyin, but different tone), then there is a second dictionary entry (mostly single characters).

there is another "feature" that you might not have encountered yet. in some dictionary even if the pinyin and tone is same, there are several dictionary entries for a headword. and this is not always to just split otherwise very large dictionary entries...
Actually, CC-Cedict has one headword with one pinyin for each entry. The same headword could also have the same pinyin with the same of different tone with different meaning. The same headword could also have different pinyin. All these are as separate entries. Yet they are all merged together as the pronunciation, and as definitions automatically. Below is an example CC-CEdict headword entries.
Trad Simp PY Definition
丌 丌 Ji1 /surname Ji/
丌 丌 ji1 /"pedestal" component in Chinese characters/
丌 丌 qi2 /archaic variant of 其|其[qi2]/

This is combined into:
Ji1
ji1
qi2
surname Ji
"pedestal" component in Chinese characters
archaic variant of 其|其[qi2]

or if implemented on Pleco with one line for all the various pinyins for that character before their definitions. This way, on the flashcard, all the pinyin and part of the definition is viewable rather than having to scroll to see all the pinyin which could be 4 or more.

Ji1, ji1, qi2
Ji1 - surname Ji
ji1 - "pedestal" component in Chinese characters
qi2 - archaic variant of 其|其[qi2]

Michael, is there a way to have this simple way of using flashcards - ie just open a txt file with the characters, one on each line. The pinyin and definitions would then be an amalgamation of what's available on the dictionaries the user chooses.

Thanks.

David
 

mikelove

皇帝
Staff member
CC-CEDICT actually has separate entries for each Pinyin, but they appear to have been combined in that Pingrid software.

There's no built-in way in Pleco to create consolidated flashcards for each character - it wouldn't really fit well with how the flashcard system is designed otherwise - but if you don't mind certain flashcard modes (particularly fill-in-the-blanks and multi-choice) being a bit screwed up (wrong #s of syllables, i.e.), you can do it with a customized user dictionary database.

Here's something I whipped up in about 10 minutes using regular expressions - this is a Pleco-format text import file containing every single-character entry from CC-CEDICT, with entries having the same character merged and the definition for each Pinyin put on a separate line:


Entries with the same simplified but different traditional characters were left separated, though it shouldn't take too long to go through and merge them if you want - if anybody feels like doing that / posting an updated file, feel free to do so. (or use it for any other purpose you like)

This uses the not-exactly-kosher-but-probably-OK-for-now private-use formatting characters from this post to separate lines / boldface the definition Pinyin.

To use this in Pleco, copy / upload this text file to your device, create a new user dictionary and import (iPhone / WM) this file into that new dictionary, then in the flashcard Import Cards screen, tap on "Dictionaries" and drag this up to the top of the "Active Dictionaries" list (better still, drag everything else into "Inactive Dictionaries").
 

Attachments

davidy

秀才
Thanks Michael! 10 minutes is really quick to whip up some regular expressions.

mikelove said:
CC-CEDICT actually has separate entries for each Pinyin, but they appear to have been combined in that Pingrid software.
I believe the structure in Pleco dictionaries also has a separate entry per Pinyin and sometimes more for an exact same Pinyin and character.

mikelove said:
There's no built-in way in Pleco to create consolidated flashcards for each character - it wouldn't really fit well with how the flashcard system is designed otherwise - but if you don't mind certain flashcard modes (particularly fill-in-the-blanks and multi-choice) being a bit screwed up (wrong #s of syllables, i.e.), you can do it with a customized user dictionary database.
I was hoping for some undiscovered way of combining the entries. I wouldn't mind having more messy fill-in-the-blanks and multi-choice - those can be done with another lot of flashcards. Actually, multi-choice would be more complete if all the different Pinyin are listed as the choice, rather than just one of the possibility.

The flashcards that I made had some cards that picked up the wrong Pinyin out of the possible choices. This made it hard to remember that character for reproduction. An example was:
ji4 (to tie · to fasten · to button up) was difficult for me to figure out what the character was
xi4 (to connect · to arrest · to worry · to relate to · to tie up · to bind · to be (literary) · system · department · faculty) would have definitely helped to figure out that it was 系

I looked for a way to edit the flashcard during the flashcard session but I found none. (Edit: I found the preference to Allow edits during sessions). It would be great to be able to add information into the flashcard when studying them. Copy and paste from popups, by hand or any other way would be great. Even if a flashcard could be marked for future improvement, that would be good. It would also be prefably done on the pda rather that having to be exported, edited, reimported. The beauty of Pleco is in its portability.

mikelove said:
Here's something I whipped up in about 10 minutes using regular expressions - this is a Pleco-format text import file containing every single-character entry from CC-CEDICT, with entries having the same character merged and the definition for each Pinyin put on a separate line:
Thank you again, that´s fast!

mikelove said:
This uses the not-exactly-kosher-but-probably-OK-for-now private-use formatting characters from this post to separate lines / boldface the definition Pinyin.
I suppose the those codes are invisible unicodes?
EAB1 for newlines
EAB2/3 for bold
EAB8/BB for copy-whatever's-in-this-to-the-Input-Field hyperlinks
EAC1 hhhh hhhh EAC2 for color - lowest-order 15 bits of each character
eg green = RGB code 00FF00 would actually translate to EAC1 8001 FF00
eg blue = 0000FF would be EAC1 8000 00FF
eg red = FF0000 would be EAC1 81FE 8000 I think

How many user dictionaries can there be?

I still would like a way of merging the Pleco dictionaries. That was one of the main reason of getting Pleco - I wanted some more reliable commercial dictionaries. It seems cumbersome to have to flick through them manually when on a computer. Please enable some way of combining all information on a character both in the dictionary as well as in the flashcard.

For the interim, is there any way I can export dictionary entries to recompile into a super dictionary?

Thanks.
 

davidy

秀才
Dear Michael,

I just checked that Pleco allows multiple entries of the same Headword with the same or different Pinyin. I could even have an exact same duplicate card.

Therefore, could there be an option for automatically making cards out of all the different entries of selected dictionaries for the headword. This way, I won't be learning just part of what a character means, but all that it means from all the dictionaries I have onboard. An extra option would be all that was necessary to activate this. Duplicate Entries would be automatically set to 'Allow' when 'Use All' is chosen for Ambiguous Entries.


Please.

Thanks.
 

Attachments

mikelove said:
Here's something I whipped up in about 10 minutes using regular expressions - this is a Pleco-format text import file containing every single-character entry from CC-CEDICT, with entries having the same character merged and the definition for each Pinyin put on a separate line:


I never thought I'd get to post that one, actually.
 

mikelove

皇帝
Staff member
davidy - good idea about "Use All" - we'll see if we can get that working for the next minor update (though it won't be ready for 2.0.4, which is going to be finalized any day now).

As far as picking the right Pinyin, with that 2.0.4 update we've now added the ability for the importer to pick (using Unihan) the most common pronunciation for a single character when no pronunciation is supplied, so you should at least get slightly less obscure / unintelligible single-character imports with that new version. (you can download it now in the beta forum - the latest beta is essentially the same thing we're releasing officially)

As far as marking flashcards for future improvement, you can do that by creating a category for that purpose and assigning it to the category button in sessions via Advanced Settings / Display - with that you'll be able to add a card to that category with a single button tap, no need to go in and edit it.

The codes are invisible ("private use") Unicode, yes - not exactly something you're supposed to use like this in a public file format, but an extremely easy way to implement rich-text formatting internally. There's no real limit on the number of user dictionaries - you can have dozens if you want - though searches that have to check all dictionaries (i.e. searches for which no match is found) might take quite a while in that case.

Merging dictionary entries is something we've never really pushed for because of the difficulty of displaying more than one entry at once on a tiny PDA / smartphone screen; our partial solution to this on iPhone is to display full dictionary definitions in a separate screen (tapped on from a full-screen list), then let you toggle between other dictionaries to display that same word with a single button tap, but we're not sure if we'll ever implement that UI on Windows Mobile (where in general people seem to prefer to have everything on the same screen). Exporting dictionary entries isn't currently supported on any platform but iPhone, and that only for a few dictionaries that it's OK for licensing-wise.

sui.generis - I actually own the t-shirt version of that particular XKCD. And an autographed copy of the book :)
 

davidy

秀才
Dear Michael,

Having gone through the bulk of 'Preferences' manual, it amazes me how configurable Pleco is. This makes it useable in an incredible number of scenarios.

mikelove said:
davidy - good idea about "Use All" - we'll see if we can get that working for the next minor update (though it won't be ready for 2.0.4, which is going to be finalized any day now).
I wonder if I am the only one who wants something like that. I figure, let Pleco do the hard job of creating all the cards, I can easily delete or categorize away cards that are of lesser value during my flashcard trainings.

I'm going to try 2.04b to see how well the Unihan linkage works - sounds great in theory. Thanks for the tip on categories. I actually found editing it ok, but prolongs my training time.

One other feature I would like is a choice of a larger area for the writing entry, and larger fonts for tapping the correct character. On a 3" 800x480, it's not exactly finger friendly nor is it eye-friendly (difficult to see difference in similar complex characters). I've looked through the preferences, and have not found a way to enlarge this area for Free-answer mode.

Another suggestion for new users who may not read the manual, is to make the drop menu in preferences more obvious - maybe in another colour. I didn't realise there were other options available until I followed your link to Importing User dictionaries, and reading the part about Flashcards preferences. Not knowing how to get there menu wise, I read backwards until I found the drop menu. I already know about it now, but it was not obvious despite my normal inclination of clicking through most menus - there are just so many of them in Pleco - which is great! I could yet get it the way I want it.
 

mikelove

皇帝
Staff member
Thanks!

"Use All" actually makes a lot of sense in general - if we're going to have None / You Choose / We Choose options, it's logical to have an All option too. So it's the sort of feature I think a lot of people will use if they have the option - we might even consider making it a default, better to be tested on a word you don't need to learn (which you can easily skip / delete) than to not be tested on an important word.

Finger-friendliness of the WM version is a bit of a rough spot in general, and to be honest not one I'm sure we'll ever fully rectify. With WM in the process of being replaced by Windows Phone 7, which won't run Pleco without a total rewrite, it seems that going forward the vast majority of the installed base of WM devices - which will certainly be in use for years and years to come, but might not be getting manufactured in significant quantities anymore - will have stylus-optimized rather than finger-optimized touchscreens. So optimizing around something that will never be more than a single-digit percentage of the WM installed base seems like it might not be the best use of our time. The flashcard free-answer interface is a particularly bad problem, though, so we might at least be able to do something for that one in a future update.

Good point on Preferences, that's actually been a problem for a long time - we added that capitalized "GO TO PANEL" text to try to help, but we still regularly get reports of people having trouble finding that option. Windows Mobile's tabbed panel interface gets really really annoying when you have more tabs than can fit on the screen, but if we can find a way to boldface / otherwise set off the "GO TO PANEL" more that seems like it might help a bit at least.
 
mikelove said:
sui.generis - I actually own the t-shirt version of that particular XKCD. And an autographed copy of the book :)
It is too bad that the tshirt didn't include the "Perl!" action caption. It really adds that old Batman ambiance. Still, in addition to getting big nerd points for owning it, you get double points for admitting to/bragging about owning it. :)
 

mikelove

皇帝
Staff member
Thanks! I actually think REs are a particularly subject for that short - they're exactly the sort of cool programmer trick that makes you feel like a superhero, C source code is intimidating enough but a line of text like:

Code:
{<tag[^>]*attribute\\s*=\\s*(["'])value\\\\1[^>]*>(.*?)</tag>}
makes most C code look like plain English by comparison.
 

davidy

秀才
mikelove said:
Finger-friendliness of the WM version is a bit of a rough spot in general, and to be honest not one I'm sure we'll ever fully rectify. The flashcard free-answer interface is a particularly bad problem, though, so we might at least be able to do something for that one in a future update.
Dear Michael,

Is it difficult to allow writing on the upper 2/3 of the screen with the whole width available to display the possible characters at a larger font-size, and the buttons below that? I note that almost whole screen handwriting is available for the dictionary screens with an on-off button at the bottom.

Thanks.
 

davidy

秀才
Dear Michael,

Here's a rough rendition with current one side by side. Note the vertical space requirement for the characters and buttons have not changed. The buttons could be enlarged.
 

Attachments

mfcb

状元
there's even room to make the "字", undo and delete buttons bigger, hehe

please cross-post to "finger-friendlyness" thread, i guess its more unlikely to be "forgotten" for a future version over there!
 
Top