iPhone Feature Requests

[character]

Thanks Mike, I've not gotten that deep into iCloud.

There are also issues with primary key assignments - if two different devices each add a new flashcard with ID 28761, we have to compare them, determine that they're different, and change the ID of one of them, which means going through a bunch of other tables and finding / modifying that ID.

One thing done in enterprise software is each system gets its own IDs* so they don't conflict when gathered into some monitoring program. Retrofitting that kind of cure is probably worse than the existing problem.

* I.e. each key is prefixed with some value based on the MAC address and other information tied to a particular system/install.

 

[mikelove]

There are also issues with primary key assignments - if two different devices each add a new flashcard with ID 28761, we have to compare them, determine that they're different, and change the ID of one of them, which means going through a bunch of other tables and finding / modifying that ID.

One thing done in enterprise software is each system gets its own IDs* so they don't conflict when gathered into some monitoring program. Retrofitting that kind of cure is probably worse than the existing problem.

Sadly yes - it would also mean a performance hit and a pretty big database size increase, since any given flashcard's ID is going to have to be stored in something like a dozen different places - we could be talking about several extra MB of data that now need to be synced / backed up / etc.

 

[icebear]

Mike,

Feature request based on a new workflow I've got that with reading articles or books.

My goal is to study key vocabulary for the content before going through it, which will allow more natural reading when I get to it. The way I do this (in the case of a book) is load the book in this tool http://www.zhtoolkit.com/apps/Chinese%2 ... Extractor/ and add the 200 most frequent words which are not already in my list of flashcards. They are sorted by occurrence in the text, and add gradually (a few a day) over the coming weeks. I try to do the same thing with an article each day, although with only 5 new words per article.

The advantage of this method is that reading is done more naturally, and the issue of choosing which vocabulary to add (to a already ballooning SRS schedule!) is simplified. There may be superior ways to go about this, but its certainly better than what I was doing before (just adding whatever words I can across and quickly getting burnt out studying low frequency words).

I was thinking it would be great if Pleco could integrate this type of functionality: I can tell it to take a bit of text (document or from clipboard), parse the words, filter the ones already in my Pleco flashcards, and then X of the remaining most frequent to a new flashcards folder named after the document.

Any thoughts? As I said this already is working for me, but if it were integrated into Pleco I could save a lot of the menial work of copying and pasting and transferring files here, there and so on.

 

[mikelove]

I was thinking it would be great if Pleco could integrate this type of functionality: I can tell it to take a bit of text (document or from clipboard), parse the words, filter the ones already in my Pleco flashcards, and then X of the remaining most frequent to a new flashcards folder named after the document.

That's been on our radar for a while, actually - the big challenge is coming up with a really good text segmentation algorithm, something that can tell us where one word ends and the next begins. There are an enormous number of places in Chinese where that can get ambiguous - three characters in which the first two can make a word but so can the second and third, e.g. With the document reader since you're selecting a single word and can easily change the selection, that's not an issue, but once we're trying to extract text from a document it becomes a very big one.

 

[icebear]

That's been on our radar for a while, actually - the big challenge is coming up with a really good text segmentation algorithm, something that can tell us where one word ends and the next begins...

I suppose a big issue is that the better algorithms already developed/available are proprietary or have licensing issues? Would Pleco need to develop its own from scratch?

Anyway, its good to know the idea is on the radar for the future. Hope it makes it in relatively sooner than later!

 

[mikelove]

I suppose a big issue is that the better algorithms already developed/available are proprietary or have licensing issues? Would Pleco need to develop its own from scratch?

There are some available for licensing, but picking which one and porting it to iOS is a decidedly non-trivial task - it's much easier to evaluate the quality of a dictionary than of a Chinese text segmenter (and even with dictionaries we do occasionally sign some less-than-stellar licenses).

 

[character]

[...] take a bit of text (document or from clipboard), parse the words, filter the ones already in my Pleco flashcards[...]

That's been on our radar for a while, actually - the big challenge is coming up with a really good text segmentation algorithm, something that can tell us where one word ends and the next begins.

How about a partial but useful solution? Scan the text and come up with a list of all possible words (remove duplicates but keep a frequency count for each word). Have an option to eliminate the ones already in flashcards. Present the list (of words/definitions/frequency in text) to the user in the order the words were found in the text so the user can delete words they think are not correct for the document. Then allow the user to add the remaining words to flashcards and/or export the list to a text file.

Pleco already has an OK text segmentation algorithm; it could be supplemented with a simple recursive one to find all possible words which match dictionary entries for a given stretch of characters.

I know there are structures such as resultative verbs this approach wouldn't find, but it seems to me it would still be of real value for learners.

 

[mikelove]

How about a partial but useful solution? Scan the text and come up with a list of all possible words (remove duplicates but keep a frequency count for each word). Have an option to eliminate the ones already in flashcards. Present the list (of words/definitions/frequency in text) to the user in the order the words were found in the text so the user can delete words they think are not correct for the document. Then allow the user to add the remaining words to flashcards and/or export the list to a text file.

It's still quite a lot of work to add that UI - I'd rather get this right the first time. It's not like the text segmentation algorithm challenge is insurmountable; as with a lot of these types of features, we really just need to find time to implement it, and before we do that we have to weigh whether it's likely to have enough impact on enough users to justify doing so sooner rather than later. It seems like there's a pretty good case for it, and it fits naturally with the other things we're improving about flashcards in 2.4, so there's a good chance it will be in that release, but there's not really any way we can dash off a useful version of this in a few hours' coding time, so it's not the sort of thing we can jam into a bug fix update.

 

[Vzzzbx]

All right, this is a very long shot, and as I type it I know it's ridiculous, but if I don't ask I'll never know.

What chance is there of licensing Rick Harbaugh's dictionary for Pleco? I'm asking because I use it more than any other paper dictionary, I'd love to not have to lug it around everywhere, and I think it would enhance Pleco immeasurably.

As far as source material goes, there's zhongwen.com, which is full of GIF images rather than text; and there's also talk of a completed Anki deck here, but I doubt it would include the character breakdowns that make the book so useful.

Assuming there is a sufficient source, I imagine it would require more than a few custom hooks into Pleco, which would take loads of time and effort. But again, if I don't ask I'll never know!

 

[mikelove]

What chance is there of licensing Rick Harbaugh's dictionary for Pleco? I'm asking because I use it more than any other paper dictionary, I'd love to not have to lug it around everywhere, and I think it would enhance Pleco immeasurably.

We wrote him a while ago and he wasn't interested; however, I think it would be possible to get most of the same functionality in it from an improved version of our component list / search feature. What exactly do you use that dictionary for now? Are you interested in the actual character definitions or mainly in the ease of looking up characters based on other characters that you know?

 

[Vzzzbx]

Thanks for the quick reply!

I use Harbaugh's book to remember characters based on their components and derivation. Pleco already has breakdowns in the character information screens, but Harbaugh's work goes a few steps further. If your improved component list feature does the same job in similar detail, that would help immeasurably.

 

[mikelove]

I use Harbaugh's book to remember characters based on their components and derivation. Pleco already has breakdowns in the character information screens, but Harbaugh's work goes a few steps further. If your improved component list feature does the same job in similar detail, that would help immeasurably.

Ah yes, that's high on our priority list - we basically just need to go through systematically and make sure that every common component has a definition.

 

[Vzzzbx]

You're wonderful.

 

[radioman]

@Vzzzbx...

Looking at that link (the book), Does it look like that? or is that just a really bad PDF rendition?

http://goo.gl/snSR4

 

[mikelove]

You're wonderful.

Should probably save those sentiments until we actually release an update with that fix, but thanks!

Looking at that link (the book), Does it look like that? or is that just a really bad PDF rendition?

The text is a lot less fuzzy in the real thing but the fonts / layout / etc look to be correct, yes.

 

[Z-Lo]

Study lists

Does anyone else out there like to study vocab in list form?

I love the Pleco flash card system. I also, however, like to study lists.

Now, I often just view my my flashcard lists (from the "organize cards" screen), but it would be great to have more control when viewing saved vocab in list form.

I was imagining two columns with character and pinyin with the option to hide the contents of one column or the other. And to be able to reveal a single item (in a hidden column) by tapping on it. The definition or full entry could be displayed by tapping on the revealed column - i.e., you would only need two columns total, one for characters and one for pinyin.

To quiz yourself on characters, you could have the character column hidden and tap the blank cell it to write it in in full-screen mode.

I would also like to access such a list a little more directly than it is currently possible to do with the flashcard list (since you have to go all the way to "organize cards"). I wish lists would be the primary screen when you tapped flashcards.

Also there could be a way to star or mark certain items as more important, already learned, 口语, 书面语, etc.

I think there is probably also much more potential with this study format.

What do you think?

(I apologize if this topic already exists - I did look though.)

 

[mikelove]

I was imagining two columns with character and pinyin with the option to hide the contents of one column or the other. And to be able to reveal a single item (in a hidden column) by tapping on it. The definition or full entry could be displayed by tapping on the revealed column - i.e., you would only need two columns total, one for characters and one for pinyin.

Well you can kind of get that now if you configure the Organize Cards screen to have a custom format (Settings / Flashcards / List Views / List layout) - show whatever you want to leave visible all the time, then tap on a card to view the full text. But I can see that having it all in the same list might make this easier. Do you generally go through entries in the list one-at-a-time, or do you scroll around randomly and tap on a particular card to look at it? I'm asking because I'm trying to figure out whether this really ought to be done in the list or whether we simply need a more flexible, quickly-jump-into-it version of the flashcard test screen (something you can access in one button from an Organize Cards list - "quick-review these cards" - but with no test component, just hide/reveal and prev/next buttons).

I would also like to access such a list a little more directly than it is currently possible to do with the flashcard list (since you have to go all the way to "organize cards"). I wish lists would be the primary screen when you tapped flashcards.

That one is a little harder to justify - for most people the New Session screen is the one they access most, so if we were going to change this at all I think we'd probably have to make that the default.

Also there could be a way to star or mark certain items as more important, already learned, 口语, 书面语, etc.

You actually can do that by creating additional categories for them; when you "Copy" instead of "Move" a card into a new category, what you're really doing is creating a link to the same card from another category. Categories are very flexible - you can use them to organize cards into lists from textbook chapters / HSK levels / etc, but you can also use them for flagging cards. We include category options in "Card Filters" for precisely that reason - if you want to, say, review all of your HSK 1/2/3 vocabulary that's also flagged as 口语, you'd choose the HSK 1/2/3 categories in the regular "Categories" screen but then set up a category filter in Card Filters to require that the cards also be in 口语.

 

[Z-Lo]

you can kind of get that now if you configure the Organize Cards screen to have a custom format

Yes, I see! Pleco is constantly reaffirming its superiority. And it is so powerful, that I guess there is some learning required - I wonder what I'll discover next...! Anyway, I did this and got rid of the definitions and card statistics for list view, and it is now much cleaner and closer to what I meant. I guess the only real difference is the ability to hide and reveal characters or pinyin - either all at once or individually - directly from the list screen. Maybe, for example, at the top could be a button to show/hide all pinyin and/or all characters. Then, a list with characters revealed and pinyin hidden, say, would reveal the pinyin with a single tap (and then tap it to make it disappear or have it fade away automatically after 1 or 2 seconds), and two taps would take you to the full entry with the definition. Or something. Just a thought...I don't know if other people would find this kind of thing useful, but I would love it.

Do you generally go through entries in the list one-at-a-time, or do you scroll around randomly and tap on a particular card to look at it?

I like to have it all in front of me. Then jump around as I am inspired to do so. This way I can quickly identify ones I know better and ones that need more review. This is a different kind of study for me than flashcard quizzing - less about memorization than about review and reinforcement.

I use flashcards for my actual textbook lessons. Most of my lists, though, are words I have personally encountered, collected, and categorized. For instance, I find that many words require more contextual reinforcement than others, and that simply identifying a dictionary entry is not sufficient for me to be able to actually use the word confidently. These words, I like to keep as reference, but do not need to "test" myself on.

Maybe that's just me though. But how do other people learn/review vocabulary before testing themselves, if not viewing it first in list form?

We include category options in "Card Filters" for precisely that reason

Again, I see! Amazing and great. This, again, seems to be for the flashcard testing, though, but maybe I'm not understanding it fully (I will consult the manual again). I was really looking for a way to highlight, if not further categorize within a category.

 

[insighter]

If you add a word from most dictionaries Pleco hides the example sentences in unless you go into card info. It would be nice if we could put our own hidden example sentences into the card info.

 

[mikelove]

Yes, I see! Pleco is constantly reaffirming its superiority. And it is so powerful, that I guess there is some learning required - I wonder what I'll discover next...! Anyway, I did this and got rid of the definitions and card statistics for list view, and it is now much cleaner and closer to what I meant. I guess the only real difference is the ability to hide and reveal characters or pinyin - either all at once or individually - directly from the list screen. Maybe, for example, at the top could be a button to show/hide all pinyin and/or all characters. Then, a list with characters revealed and pinyin hidden, say, would reveal the pinyin with a single tap (and then tap it to make it disappear or have it fade away automatically after 1 or 2 seconds), and two taps would take you to the full entry with the definition. Or something. Just a thought...I don't know if other people would find this kind of thing useful, but I would love it.

Thanks! I'm glad to hear that improved matters. Though a button toggle certainly might make sense in a future update (particularly since we've got plenty of toolbar space on that screen anyway).

I like to have it all in front of me. Then jump around as I am inspired to do so. This way I can quickly identify ones I know better and ones that need more review. This is a different kind of study for me than flashcard quizzing - less about memorization than about review and reinforcement.

Makes sense - so the list element to it is important and we couldn't really replicate that with an easier-to-access Review Only test.

Again, I see! Amazing and great. This, again, seems to be for the flashcard testing, though, but maybe I'm not understanding it fully (I will consult the manual again). I was really looking for a way to highlight, if not further categorize within a category.

Well you could do that with Search Cards - "AND" together two categories - but it's not exactly ideal.

If you add a word from most dictionaries Pleco hides the example sentences in unless you go into card info. It would be nice if we could put our own hidden example sentences into the card info.

That should be possible with the custom field support we're planning to add in 2.4, though it wouldn't specifically be about adding example sentences but rather about adding any extra data you feel like.