iPhone Feature Requests

[numble]

That's part of the issue right there, actually - web-based dictionaries can't be integrated into other parts of the software like the document reader, nor we can we really use them as an automated fallback. All we'd really be doing is adding a command to automate the process of copying out your search query and forwarding it to something via a web browser... I suppose that might be worth doing with customizable toolbars, though; let someone define a URL and then assign a toolbar command to insert the current search query into a particular place in that URL and pop up a web browser window with the results. But I really wouldn't want to integrate it into the UI much more tightly than that.

Here's a thought--what about a completely separate product that does something like that. As much as feature-creep is making Pleco more of a "Chinese Learning Tool" than a "Chinese Dictionary," I think something with encyclopedia look-up could be useful, and might even be attractive to users that can already speak/read Chinese (when coupled with Chinese web encyclopedias). It could also perhaps be coupled with some different version of OCR (since the current implementation is geared towards dictionary lookup). When reading a text and I come across a name of a famous person, I often want to Google/Wikipedia the person's name, especially if it's someone I don't recognize. People could also use it to look up drug ingredients, etc. And it's possible that these web encyclopedias have better entries for restaurant food entries, brand names/products, historical events, and places, something which dictionaries seem to be lacking in (or only provide very sparse info).

 

[radioman]

I would like to have the capability to cut to the pasteboard the entered characters off of the Dictionary screen.

So, if i have the dictionary and enter text, I want to be able to submit it to the dictionary. But I also want to be able to be able to copy the text to paste to another application, like SMS.

Once a string is submitted to the dictionary, the string is still at the top, but there is no way to copy that string. You can copy the definition, but not the input string (best as I can tell). Maybe a link at the bottom of the definition page to copy the input string, similar to the copy to pasteboard.

In this way, I can just have Pleco open in iOS 4.x at the dictionary screen and use that for writing input rather than some other application using Apple's input method.

 

[mikelove]

Here's a thought--what about a completely separate product that does something like that.

That's a possibility, though people aren't going to be happy about effectively shelling out for OCR twice - if we did this at all it probably would have to be integrated with the main Pleco app, though as you suggest it would require a different interface; can't return "live" results if we're waiting on a response from a server.

I would like to have the capability to cut to the pasteboard the entered characters off of the Dictionary screen.

That should be working in the 2.2 beta - you can tap-hold select/copy text there just like in a standard iPhone text edit field.

 

[radioman]

I just tried using the copy method, and its just not convenient (at least for what I'm tryin' to do). Getting the area highlighted properly, then selecting all, then saying copy... My idea is a bit more automated.

Specifically, if I'm in the dictionary, I can use the dictionary per normal. But, if I input a bit of extra text, I would like Pleco to optionally write that text to the pasteboard automatically. This gives the benefit of using Pleco as an input method (and while not off off the standard iPhone keyboard selector, still WELL worth multitask switching effort to use Pleco HWR as an ad hoc input method).

So in the dictionary module, I scroll a bunch of HWR text, maybe 8 characters, maybe 40, and it is all written to the pasteboard, no need to copy, cut, highlight, press selector buttons. I can then swap over to the SMS app (or any other app) using multitasking and paste.

That should be working in the 2.2 beta - you can tap-hold select/copy text there just like in a standard iPhone text edit field.

 

[mikelove]

How about using the built-in text editing function to compose the text you want to copy over - does that help matters any? I'm somewhat wary of this automatic pasteboard overwrite, since it would be easy to forget you'd turned it on and lose some important bit of material in the pasteboard - I think in general this would be better done by a separate / quick-input-optimized version of that text editor screen with an easy button to copy the results to the pasteboard.

 

[radioman]

I would respectfully offer that it should be:

1) automatic,
2) optional (you have to consciously set a switch), and
3) when you set the switch, you get a warning of the perils of using the function.

The beauty of it is that, if Pleco is camping in Dictionary mode while you use the iPhone to do other things, you immediately (just by quickly task switching) can get in and start writing - no nonsense - where you can then:

1) look up words and/or
2) start immediately writing sentences (who would want to use the built in HWR if you have Pleco's?? its just easier and I get to two finger tap to move to the next character or delete... )
3) never have to leave dictionary mode (which is good for the next time you need to run to Pleco when using another program).
4) and (I think), it has the advantage of allowing someone to just write into the Dictionary (e.g., throw them the phone and let them write) and if you really want to look at the sentence in more detail, you can flip to the reader without having to do anything. It will already be in the pasteboard, ready to be reviewed.

I do no not want to have to switch to Dictionary mode, then over to some editor mode, then hit the wrong button, then yada yada, and when all is said and done, go back to dictionary mode, etc. Done the way I propose, I do not have to do any of that.

I would agree that there is some risk. But the risk in the case of the pasteboard is there anyway. If someone switches out of Pleco to some other program, that information is already at risk (email, web browsing, and any other program where you might cut and paste). I run into this from time to time now with the way that I use Pleco's Reader and do other stuff on the iPhone, and its not that bad. In fact, when you recut and switch back Pleco, Pleco immediately resets back to the reader, re-enters pasteboard, and I'm back up and running.

Maybe Pleco would need to so some check to see if the Pasteboard has changed since writing (and if not, it does not throw the user to the Reader in Pasteboard mode) upon next re-entry into Pleco via application switching.

Anyway, long story short, I do think there is value with the approach.

EDIT: Related - Maybe this is some new input mode... Along the same lines, using HWR, I can stack up Hanzi for input in this way. However, if I switch to Pinyin Input, It does not let me enter in continuous converted Hanzi. So I get to Pleco Dictionary, and I start typing in Pinyin, see the right word, then want to type the next utilizing Pinyin (but selecting Hanzi. This is different than the Pinyin input where NO definition is provided. In this case the entry mode would allow you to type in the Pinyin and it would show the definitions underneath like the current input method in Pleco for the dictionary using Pinyin. However, similar to HWR, characters would be brought up underneath (with their definitions). Similar to HWR, the user could two-finger tap to accept the first one, could press a second. But rather than bringing up the dictionary for that entry, it would just put the Hanzi in the entry line. In this way, you could build sentences.
EDIT: Unrelated - If someone has a jailbroken iPhone, I found a program that works really well, allowing the user NOT to have a to use the home key for switching apps - called "quickdo" (used to be called "mquickdo"). So I never have to use the home button to switch applications. Saves the home button and allows you do some other quick-launch stuff.

How about using the built-in text editing function to compose the text you want to copy over - does that help matters any? I'm somewhat wary of this automatic pasteboard overwrite, since it would be easy to forget you'd turned it on and lose some important bit of material in the pasteboard - I think in general this would be better done by a separate / quick-input-optimized version of that text editor screen with an easy button to copy the results to the pasteboard.

 

[character]

I've become convinced that handwriting practice is helpful for advancing beyond a certain point in studying Chinese (feel free to disagree). Given the HWR module and the Stroke Order diagram/test, I wonder if there's a relatively easy way for Mike to create a handwriting practice screen with the HWR screen, and optionally have that overlaid on top of a large image of a character. It's a little different than a flashcard in that one would want to be able to repeatedly write the character and not automatically move on to the next.

 

[tennonn8]

I have a question about the navigation in the Flashcards system. Could the new version of Pleco give us a possibility to customize the buttons in the Flashcards screen? Like configure the layout, move the buttons between the "tabs," etc.

I understand that everybody has their own habits when using the Flashcards system. As for me, I would like to have the "Back" button (and, optionally, the "Forward" button) always on the main "tab," i.e. side by side with the buttons like "next card" (in the "Review Only" mode), "reveal entire card," "mark correct" and "mark incorrect," etc. According to the manual, it used to be like this in the WM version of Pleco (http://www.pleco.com/manual/images/wmscreens/reviewonly.gif) and I'm not quite sure that I understand why that behavior was changed. Probably because of the "touch" interface, to give more room for the finger of a user? Well, even in that case the "Back" button could have stayed. But if most Pleco users should prefer not to have this button on the main "tab" then could it be made configurable, perhaps?

I found a very short discussion about the "Back" button on this forum, here is a reply from Mike to a comment from another user:

#2 is a good point - going back to a previous card is sufficiently nondestructive that we don't really have to worry too much about people hitting it by accident, so we might as well trigger it automatically when you select the back tab. Forward turning into skip is one of those cases where I think the logical thing didn't turn out to be the right thing - it seems natural that you'd have one button to advance cards regardless of where you are in the session, but it would probably be better if we swapped it with the (always-disabled-when-you're-on-the-current-card) undo button so that once you got back to the current card you'd have to move over to start skipping. Though FWIW if you just tap on the lightbulb icon you'll go right back to the current card.

So, Mike, what is the current situation here? (I.e. in regard to the new version of Pleco.)

BTW, another problem is mentioned here that I encounter all the time. After hitting the "Back" button several times, I want to go forward (but not directly to the current card!) so I hit the "Forward" button several times and then all of a sudden it turns into the "Skip Card" button! Very dangerous!

Well, there's probably a possibility to have a solution without going into further customization. You already mentioned, Mike (a citation above), that it makes sense to implement it in such a way that one starts going back as soon as they hit the "Back" tab (that is, from the main tab.) This is certainly a very good idea, to save us the second tap here. So if it were the case, then I hit the "Back" tab, I go immediately one card back and the whole "Back" tab is brought to the front so that I can continue clicking the "Back" button. Then I reverse and start clicking the "Forward" button. After I got back to the "current" card, should the tab, perhaps, be changed automatically into the main tab again? (I.e. I click the "Forward" button for the last time just before the current card is about to show and then the tab changes automatically into the main tab, right?) But if you don't want to change the tabs automatically, then at least the "Forward" button should just disappear, so that clicking the same place on the screen again by mistake wouldn't do any damage like skipping the current card. Well, these are just my thoughts. Maybe I'm reinventing the wheel and the behavior of the buttons on the Flashcards screen is already planned to be something like this for the next version...

I'd be very grateful for the answer, thanks!

 

[Entropy]

I've become convinced that handwriting practice is helpful for advancing beyond a certain point in studying Chinese (feel free to disagree).

I certainly can't disagree. I started learning to decipher menus in March, and can already recognize a fair percentage of the characters in question. I can't imagine trying to memorize these characters without the constant writing practice.

Given the HWR module and the Stroke Order diagram/test, I wonder if there's a relatively easy way for Mike to create a handwriting practice screen with the HWR screen, and optionally have that overlaid on top of a large image of a character.

This makes me think that once you write a character, you can turn it into a picture and then run the OCR module on it. Could this be turned into a test of how good your writing actually is?

eStroke's HW practice screen gives you outlines and optionally stroke number and direction arrows. But, it uses calligraphic outlines, so what you write looks sloppy compared to the outline. This makes it entirely unclear to *me* whether the character actually looks like it should.

As for this being in the iPhone feature request thread, I have to say that using either of these tools on iPad is much more enjoyable than on iPhone even with a stylus for HW input (since I can't actually fit whole characters into Apple's HW input field with my finger.) If the rumors of a smaller, lighter iPad are true, that would make a really nice compromise platform for use with Pleco, if we could just figure out how to get picture data into it for OCR. I'm still thinking that using the phone's camera to send live data to the pad would be the easiest and best option.

~ Kiran <entropy@io.com>

 

[teab]

The only request I really have is to find a way to work in a Cantonese dictionary, especially one with Cantonese pronunciation.

I know this isn't easy to do because it's hard to source an electronic version of a Cantonese dictionary. But please keep trying!

Thanks!

 

[mikelove]

I would respectfully offer that it should be:

1) automatic,
2) optional (you have to consciously set a switch), and
3) when you set the switch, you get a warning of the perils of using the function.

Well I suppose we could have an "input field mirrors clipboard" option or somesuch, but I really don't feel like the search input field is a good place to enter more than a few characters at a time - we might be able to make it one (make it expand when you type > 1 line worth of text, say) but that's a whole different discussion. The whole UI is designed now around entering just one word at a time, not a whole sentence.

Maybe it's just because we're in the middle of redesigning our Settings screens, but I really feel like the threshold for adding a new option now ought to be that it's a) useful to a large number of people, b) implemented well enough that they're likely to actually want use it, and c) easy enough to understand that they're actually able to find / enable it, and I'm not sure about b) and c) in this case - I think for this option to really work we'd need to do some more fundamental re-thinking of how people get text into Pleco and how they look up strings of text longer than a few characters. Burying an option in "Advanced" that, if enabled, causes the existing search input field to act as a sort of pasteboard "scratch pad" is too much of a half-measure; there are already too many experimental loose ends in Pleco as is (e.g. document editing, Live Mode, flashcard stroke order testing, and Zhuyin input, just to name a few) for us to keep creating new ones without having a clear plan for how to eventually flesh them out into real / widely-used features.

So in other words, I think the idea has merit but I think it needs to be done as more than just a hastily-inserted new option backed by a dozen lines of code.

Your Pinyin-input-with-definitions idea is also interesting - I think it would actually have more use for general document composition, though, people can type in text and get definitions if they're not sure about the characters. (I seem to remember seeing an IME that did this on Windows a few years ago... anyone know the name?)

I've become convinced that handwriting practice is helpful for advancing beyond a certain point in studying Chinese (feel free to disagree). Given the HWR module and the Stroke Order diagram/test, I wonder if there's a relatively easy way for Mike to create a handwriting practice screen with the HWR screen, and optionally have that overlaid on top of a large image of a character. It's a little different than a flashcard in that one would want to be able to repeatedly write the character and not automatically move on to the next.

We've been playing with a bunch of ideas like that, actually - at the very least we really need some sort of freeform stroke order test, KingKanji was doing that on Palm OS like 10 years ago and it's embarrassing that all we have now is a tap-on-the-outlines thing.

I have a question about the navigation in the Flashcards system. Could the new version of Pleco give us a possibility to customize the buttons in the Flashcards screen? Like configure the layout, move the buttons between the "tabs," etc.

I understand that everybody has their own habits when using the Flashcards system. As for me, I would like to have the "Back" button (and, optionally, the "Forward" button) always on the main "tab," i.e. side by side with the buttons like "next card" (in the "Review Only" mode), "reveal entire card," "mark correct" and "mark incorrect," etc. According to the manual, it used to be like this in the WM version of Pleco (http://www.pleco.com/manual/images/wmsc ... ewonly.gif) and I'm not quite sure that I understand why that behavior was changed. Probably because of the "touch" interface, to give more room for the finger of a user? Well, even in that case the "Back" button could have stayed. But if most Pleco users should prefer not to have this button on the main "tab" then could it be made configurable, perhaps?

It could be, but these sorts of things take longer than you might think - there's a LOT of debugging needed, all sorts of conflicts possible in specific combinations of buttons that might not occur naturally to us but would be very obvious to some of our users.

But it's certainly possible, and given the number of opinions on flashcard button layouts it seems highly desirable too - I think we're going to try to get some of the kinks in custom toolbar designs straightened out in the dictionary / reader screens first, though, since in general there seems to be a lot more demand for it in those.

So, Mike, what is the current situation here? (I.e. in regard to the new version of Pleco.)

The buttons have now been swapped for 2.2 - that seemed relatively non-controversial.

Then I reverse and start clicking the "Forward" button. After I got back to the "current" card, should the tab, perhaps, be changed automatically into the main tab again? (I.e. I click the "Forward" button for the last time just before the current card is about to show and then the tab changes automatically into the main tab, right?)

I think the solution of replacing Forward with an inactive button once you get to the current card makes more sense - if we simply switch back to the answer buttons, people might absentmindedly reveal a card they still needed to be tested on, which would be almost as bad as skipping it altogether.

This makes me think that once you write a character, you can turn it into a picture and then run the OCR module on it. Could this be turned into a test of how good your writing actually is?

Unfortunately no, because the things the OCR module tends to have problems with are things totally unrelated to good Chinese writing - joined strokes, for example, which are perfectly fine calligraphic form (and are very easy to do accidentally on an iPhone screen) but which don't occur in printed Chinese text. But measuring whether the strokes you draw are in roughly the correct order / position relative to each other is quite doable.

Generally speaking, for a stroke order test would you find a system like KingKanji (where you draw the whole character, then see strokes marked green if correct or red if incorrect) or like Skritter / PopupChinese (where you're corrected after each stroke) preferable? I think I tend to prefer the former - that allows for areas in between "right" and "wrong" (so we can eventually refine / build on the algorithm to start looking at subtle things like stroke lengths and pointing out minor problems to work on), and discourages guessing / cheating by drawing the first stroke of the thing that you think is supposed to go there and seeing it reinforced if correct.

The only request I really have is to find a way to work in a Cantonese dictionary, especially one with Cantonese pronunciation.

I know this isn't easy to do because it's hard to source an electronic version of a Cantonese dictionary. But please keep trying!

Working on it, hopefully we'll have some good news on this front soon.

 

[mikeo]

Re: iPhone Feature Requests : Frequently confused characters

Some way of associating one character with one or more other arbitrary characters would be nice for reviewing frequently-confused characters.

One way would be (please let me know if there's another way to do this now !):

Any flashcard can link to one other flashcard (one with which it might be confused or associated). So you can build a chain of these, all pointing to each other.

Of course there'd need to be some easy way to, when looking at one card, choose one to which to link it to --- you'd need to have HWR/Pinyin dictionary access available at flashcard review or "management" level in order to make these associations.

And there'd also need to be a way of having a flashcard review session which traverses these links.

 

[Entropy]

Generally speaking, for a stroke order test would you find a system like KingKanji (where you draw the whole character, then see strokes marked green if correct or red if incorrect) or like Skritter / PopupChinese (where you're corrected after each stroke) preferable? I think I tend to prefer the former

I think I concur, though I'd like to be alerted if the first stroke is wrong. There are things like 成 that don't seem to follow the rule about each component being written as a whole.

~ Kiran <entropy@io.com>

 

[tennonn8]

So, Mike, what is the current situation here? (I.e. in regard to the new version of Pleco.)

The buttons have now been swapped for 2.2 - that seemed relatively non-controversial.

Mike, did I understand you correctly that you meant that the "Undo" button and the "Skip" buttons will be swapped in 2.2? That is, if one goes back first, then is clicking the "Forward" button, the "Forward" button will turn into the disabled "Undo" button as soon as she reached the current card again, right? Well, if this is so in 2.2, then it's a great improvement already, thanks!

And what about the possibility to go one card back immediately as one clicks the "Back" tab? (I.e. to save us one additional click on the "Back" button.) Or did you also refer to this feature when you said that it would require a lot of testing and is therefore not implemented yet?

By the way, sorry if maybe I wasn't paying attention, but is there a current list of new features / bug fixes that have been already scheduled for the coming version 2.2, like, published somewhere on this forum or the official site of Pleco? It would certainly helped us to avoid asking for the same new feature more than once...

 

[mikeo]

Re: iPhone Feature Requests : Frequently confused characters

Some way of associating one character with one or more other arbitrary characters would be nice for reviewing frequently-confused characters.

This kind of feature would also be useful
for 多音字which occur in phrases. The dictionary will link to the alternate pronunciation of, for example, 率， but what you often want is to learn the difference in how the character is pronounced in the context of the phrase in which it occurs, and the individual character's dictionary entry link to its alternate pronounciation doesn't help you in those case (because you never see the individual character's entry - only the entry of the phrase in which it occurs.

For example:

比率
表率
粗率
概率

 

[mikelove]

Some way of associating one character with one or more other arbitrary characters would be nice for reviewing frequently-confused characters.

Interesting... so not just a hyperlink, in other words, but something that could actually be used as the basis of a flashcard testing mode / brought up through various screens / etc... can you think of any other significant uses for this beyond frequently-confused characters? Because if it's just for those, there might be a more elegant way to do this for you based on data we already have.

I think I concur, though I'd like to be alerted if the first stroke is wrong. There are things like 成 that don't seem to follow the rule about each component being written as a whole.

That makes sense, though at the same time I worry that it'd be a bit of a kludge if we alerted you on an incorrect first stroke but not any other; how about some sort of an "accuracy meter" that gave you a rough idea of how well you were doing as you drew the character? If you saw it way down in the red after your first stroke you'd know you had it wrong, but this would also give you some of that fine-grained feedback that's the major advantage of this system over an auto-improving one.

Mike, did I understand you correctly that you meant that the "Undo" button and the "Skip" buttons will be swapped in 2.2? That is, if one goes back first, then is clicking the "Forward" button, the "Forward" button will turn into the disabled "Undo" button as soon as she reached the current card again, right? Well, if this is so in 2.2, then it's a great improvement already, thanks!

Yep, that's already in there.

And what about the possibility to go one card back immediately as one clicks the "Back" tab? (I.e. to save us one additional click on the "Back" button.) Or did you also refer to this feature when you said that it would require a lot of testing and is therefore not implemented yet?

That one's not in yet but it might be - my major worry is that people will hit it by accident when exploring tabs to find out / remember what the other buttons are, plus it breaks the whole "tab" abstraction in general.

By the way, sorry if maybe I wasn't paying attention, but is there a current list of new features / bug fixes that have been already scheduled for the coming version 2.2, like, published somewhere on this forum or the official site of Pleco? It would certainly helped us to avoid asking for the same new feature more than once...

No, in fact 2.2 has been heavily scaled-back because some non-OCR things are taking longer than expected and the interest in OCR is insanely high - we're basically trying to get a few more essential changes in there, release one more beta to make sure we haven't screwed anything up too badly and then submit the finished 2.2 to Apple (saving other still-uncompleted changes for 2.2.x or 2.3) - getting a little sick of half a dozen daily "when will OCR be ready?" emails

This kind of feature would also be useful
for 多音字which occur in phrases.

Already answering my earlier question... could do both of these through a similar system, though, and in both cases we've basically got the data we need already. (though there'd need to be a way to hand-tweak it / add custom entries / etc)

 

[Entropy]

That makes sense, though at the same time I worry that it'd be a bit of a kludge if we alerted you on an incorrect first stroke but not any other; how about some sort of an "accuracy meter" that gave you a rough idea of how well you were doing as you drew the character? If you saw it way down in the red after your first stroke you'd know you had it wrong, but this would also give you some of that fine-grained feedback that's the major advantage of this system over an auto-improving one.

I like the accuracy meter idea a lot. Not sure what you mean by auto-improving? But I still want something that lets me know quickly that the first stroke doesn't follow the rule about component parts. Perhaps just flagging *that* would be useful.

 

[mikelove]

I like the accuracy meter idea a lot. Not sure what you mean by auto-improving? But I still want something that lets me know quickly that the first stroke doesn't follow the rule about component parts. Perhaps just flagging *that* would be useful.

Well if you suddenly find yourself in the red after just one stroke then you know that one stroke was wrong, but I see your point - maybe some sort of flash or other alert after a very / completely wrong stroke?

 

[Entropy]

Well if you suddenly find yourself in the red after just one stroke then you know that one stroke was wrong, but I see your point - maybe some sort of flash or other alert after a very / completely wrong stroke?

If it really pegged the meter, that would do. But I think it would be useful to mark those characters in general, so that when you look one up you know it's got something funny in its stroke order.

So what did you mean by auto-improving?

~ Kiran

 

[mikelove]

So what did you mean by auto-improving?

Something like Skritter or PopupChinese, where after you draw the stroke (or something reasonably close to it) the software morphs / changes your stroke into the correct one; it "feels" great, almost seductive in a way, but I'm still trying to figure out whether it's actually a good or a bad thing for one's character retention.