2.0.1 New Features

[renovator]

[quote="mikeloverenovator - oh it's not about the programming, it's about the interface - there are way too many options / screens / etc for people to deal with as is. Letting people customize the sort order of cards in categories, for example, is another thing I'd love to consider for 2.0.1, but it's a terrible idea in terms of simplifying the interface, so for that reason alone it's going to need to wait until 2.1. Thus my one-checkbox comment at the start of this thread. This particular feature might not require many new checkboxes, but that extra "1" is going to generate a massive amount of e-mail (just like the *s after ABC headwords and the numbers before them do now) and general confusion so it would have to be added very cautiously.[/quote]

To keep the interface as simple as possible. Have a default of an unchecked box and to engage the feature give us a box in flashcards Advanced Settings>Display between Defn size and Buttons with a check box: Duplicate headword sequence

I am sure you will find that anyone doing 100 or more cards per day will really value this option.

 

[mikelove]

Well that would keep it from bothering people who don't want to use it, but I'm still not sure how usable it is in general - aside from raising a whole lot of confusing questions about flashcard organization (which card does the score get recorded on if you've popped up a different one, do we include those alternate cards if they're anywhere in the system or only if they're in the categories eligible for this session - another checkbox or two needed just to resolve that, do we count those alternate cards as reviewed if people look at them, etc), it seems like embedding this right in the interface would make the nature of the card itself somewhat ambiguous - better to have, say, a "search for alternates" button option at the bottom of the screen, so that those other cards are presented in a clearly different interface and there's no question about whether they're part of this session too and how they're supposed to be handled / scored / etc.

 

[renovator]

The cards come up now, so just keep bringing up the cards and scoring them the way they do now. Make no changes there. By making this extra check box, it only identifies specific card order and changes nothing. It would now allow me to see which card with the identical chinese character headword is being presented. At this time it is a crap shoot to figure out which card is being presented. Actually not a total crap shoot because I can sort of guess based on the card score if it is the card I know best that will be shown or the card I screwed up the most often. This is really not a very good situation. By identifying the duplicate cards with the same chinese characters first by total # of tones in ascending order would take care of the bulk of these duplicates. 亲奶奶 would have two possible cards qin1nai3nai or qing4nai3nai so qin1nai3nai would be identified as card 1 as it has 1+3=4 tones as lower than qing4nai3nai 4+3=7 tones identifying it as card 2. 不是 bu4shi with a total of 4 tones would be identified as card 1 and bu4shi4 with a total of 8 tones card 2. Only clash I can see is where tones are equal as in 校 where both jiao4 and xiao4 both score a total of 4 on total tones. In case of tie use lowest letter in alphabet first. J before x so Jiao4 would be identified as card 1.

So really only 2 variables which I am sure anyone that does 100 or more cards per day would really appreciate so they don't introduce a bunch of bogus scoring within their card deck.

 

[hleperlier]

• I learned some Cantonese a while ago and my friends decided to make me read Cantonese Characters without using any romanization system. I found this considerably helped my pronunciation and fluency (or lack of?!).
I haven't installed the sound files, not for lack of space, but because the zhuyin fuhao are enough for me. I would welcome (and install) the sound files if I could hear the characters pronounced while reviewing flash cards and Chinese definitions (in Send to Reader) without having to read any pinyin or zhuyin fuhao; being able to have the Chinese definition in the Guifa dict. would be very helpful in a flashcard.
Maybe it's possible now but I'm not sure how to make it work.
Thanks Mike for a great programme.

• still getting the occasional crashes, not sure why. It seems to happen when I change functions, such as highlighting a character or word then trying to cancel or erase the input line or vice versa

Henry

stoney - so the audio would only be played on the button press and not as soon as the card comes up? You can already map the audio play to a button, actually (from the Flashcards section of Preferences), though that won't stop it from playing audio as soon as the card comes up. You can reveal parts of cards separately if you map hardware buttons to the "screen left" / center / right commands and enabled the reveal-separately option in the Modes section of Advanced Settings.

what I want to do with flashcards is, without auto play enabled;
1. See the 'headword' ie, the chinese characters, but no pinyin
2. say the tones out load ( or in my mind)
3. press a button, soft or hard, but preferably hard to play the audio
4. listen to see if I was correct

but now, I think the sound will only play if the pinyin is first revealed. I would like to be able to see the characters first ( therefore no auto play ) and think of the sound without prompting from the pinyin, then play the audio. I don't want to associate the sound with any pinyin, just with the characters.
Now you can hear the sound first with autoplay or hear the sound with a button but only after seeing the pinyin.

--------------------------------
Ok. I got the "reveal separately" part figured out now. Thanks.

 

[mikelove]

renovator - ah, now I get it. But I'm still not wild about that approach, particularly as those numbers can easily change - if you add another card with the same characters that's sorted earlier tones-wise, all of the other cards with those characters would be shifted down and hence you'd have to relearn all of the numbers, or if you delete a duplicate card with those characters then everyone gets shifted up. (duplicate cards seem to throw a wrench in this scheme in general)

Numbers based on the actual tones (e.g. list the sum of the tones in the corner) would stay consistent, but that's problematic since they'd give you a hint as to the correct tones. Perhaps we could provide an option to list the internal flashcard IDs, those would at least remain consistent as long as you didn't export / reimport the list - bit complicated to remember, though, since most of them would be 4 or 5 digits.

hleperlier - if you set the "Mandarin pron" in the Display panel of Preferences to "Skip," that will get rid of pronunciation in most dictionary contexts (including the reader, I believe). With flashcards, using "Head + Audio" in a flashcard test (configure that in the Show field of a Test Setup) would give you the characters and audio but not show you the pronunciation until you revealed the card.

Not sure about those crashes - are you running a lot of other programs in the background on your TX? (like CJKOS)

 

[renovator]

renovator - ah, now I get it. But I'm still not wild about that approach, particularly as those numbers can easily change - if you add another card with the same characters that's sorted earlier tones-wise, all of the other cards with those characters would be shifted down and hence you'd have to relearn all of the numbers, or if you delete a duplicate card with those characters then everyone gets shifted up. (duplicate cards seem to throw a wrench in this scheme in general)

Numbers based on the actual tones (e.g. list the sum of the tones in the corner) would stay consistent, but that's problematic since they'd give you a hint as to the correct tones. Perhaps we could provide an option to list the internal flashcard IDs, those would at least remain consistent as long as you didn't export / reimport the list - bit complicated to remember, though, since most of them would be 4 or 5 digits.
quote]

There are many ways to do this, I was just looking for the most simple. You could just bring up a message on a duplicate card and let us input whatever text we want for example but that seems like more of a pain. I understand the shuffle you are concerned about but it is not as bad as you think because if a card is removed, then it is no longer a duplicate (therefore not a problem of confusion) and I have not yet found triplicate occurances which could make this a problem. If you add a card, now there would be a notification system in place that would show it as a duplicate with its proper sequence number. I don't think you will find any perfect system to resolve this problem, but if we wait for perfection in life we get nowhere. With this identification system I really think it will handle virtually all permutations.

When you export and reimport a list, why can't the program just go back through the master list, check for duplicate headwords and identify the duplicate cards per this system? Each card is now only once in the database so this should not be too hard to do. Nothing really changes from what you are doing now as all this system will do is put some form of identification on duplicate headword cards and the identification will have a little sequence intelligence so that we have a clue which card is actually being shown.

They way the cards come up now is a real pain and butchers the score of these cards as well. This proposed system would cure, I believe at least 95% plus of the problem.

 

[feryl]

Dear Mike, how about add a keyboard shortcut for "Add/Edit Custom Flashcard" ? Adding custom flashcard takes several steps: open modules/open flashcards/manage flashcards/tap plus button/tap Simp field . It would be very nice if I could skip those steps and add custom flashcard directly like Add/Edit Custom Entry. Thanks for your hard work.

 

[renovator]

renovator - if you add another card with the same characters that's sorted earlier tones-wise, all of the other cards with those characters would be shifted down and hence you'd have to relearn all of the numbers, or if you delete a duplicate card with those characters then everyone gets shifted up.

Not quite true. The schema I propose is just for you to put the cards in proper sequence. I think for the designations shown on card, since there would only be 2 cards would be Dup1 and Dup2. Which card is Dup1 or Dup2 would be determined by the propesed schema. The proposed schema would be shown in your manual so everyone that wants to use this option know how the cards are sequenced.

 

[mikelove]

renovator - there are some characters with as many as 5 different pronunciations, like everybody's favorite 阿 - I guarantee you there are more than a few Pleco users who have all 5 pronunciations for that in their flashcards. And I don't think this is that severe a problem given that we've had hardly any complaints about it in the 3-and-a-half years since we released PlecoDict 1.0 (which had the same potential for duplicates).

Counting any pronunciation as correct as long as there's some card with the same characters that has that pronunciation would really screw up the scoring system - you'd end up reviewing / marking as reviewed a card that wasn't actually due to be studied yet, and not reviewing one that you were supposed to study. And you might end up not practicing some pronunciations at all, since you'd likely keep entering the one that you remembered the most easily.

Really I think the best solution to this is just to show the definition along with the headword in pronunciation tests - that lets you disambiguate between words based on the actual differences between them. We could add some other features to deal with this problem, like an option to search for cards with duplicated headwords so that you can make sure they're assigned to different categories and not included in the same session, or an option to reveal the definition only in cases where there were multiple headwords, but adding what would have to be semi-arbitrary numbers to cards (numbers that don't even remain the same between different people's flashcard decks, since some might have duplicates and others might not) just doesn't feel like a very clean solution - nothing else I can think of works as well as simply showing the definition.

feryl - good idea; you can already do this to some extent by assigning it to a button in "Button Commands," but having a regular menu shortcut would certainly help too.

 

[renovator]

renovator - we released PlecoDict 1.0 (which had the same potential for duplicates).

OK, thanks. PlecoDict 1.0 did not have the same problem. In 1.0 I could count the card correct with either pinyin as both pinyin were shown on a card with the same headword. In 2.0 with QWERTY input and automatic scoring, the card is counted as wrong if I input the pinyin from the alternate card to the one that is shown.

Showing the definition along with headword on all cards will be the solution for now. I do like your idea to reveal definition only on cards with duplicate headwords and that would be great if you ever have time to implement.

After valiant effort, I give up on this one for now.

Happy Thanksgiving!!!!! and thanks again for this wonderful product

 

[grabolus]

Hi,

a minor feature could be a possibility to input german umlaute. On my treo I can not write those into pleco (we discussed this in another thread). Maybe an additional input keyboard would be possible?
Maybe you only need to duplicate the code for the english input keyboard and add some german umlaute.

I am not sure if this is a treo only issue... I hope the german community is growing...

Bye

 

[mikelove]

renovator - good point, 1.0 didn't have a mode where this mattered as much. Anyway, selective definition reveals are certainly a possibility for a future release, and if we come up with another solution we like we'll certainly think about it - I agree it's a problem, just not one with a straightforward solution. Thanks for your persistence, really helps with designing future versions to be pushed to think about this stuff.

grabolous - good point - would a standard German/Austrian QWERTZ layout be best, or do we need something with a ß too?

 

[grabolus]

It would be great if ß were available as this appears in HanDeDict, too. This letter is used with a probability of 0.3% in German. It would be possible to live without it, but if you include it and if you include Umlaute "äöüÄÖÜ" then you have all possible letters in german language.

 

[mikelove]

OK, we'll see about ß - should be room, I think. Though if we do add a German keyboard would probably appear only on the separate Input screen, at least initially (since it would be a lot harder to rearrange those E/P/Z/EN/PY/ZY buttons to add a German option in the main screen).

 

[sinoreen]

What I'd love to see already in 2.0.1.: The option to show a button in the toolbar that lets you directly jump to the Edit Card Screen from any currently displayed dict entry.

 

[mikelove]

Would this be for a newly-created flashcard or for an existing one (if it found one that linked to the dictionary entry)?

 

[sinoreen]

#1 I was just thinking about the best way to implement this. How about that: The Add to FC-Card button in the normal dict view changes depending on whether a card with the same headword and pinyin as the currently displayed dict entry already exists in one of my FC-Lists or not (maybe change symbol AND colour). This would have the advantage that looking up a word in any dict, I would immediately know whether the entry already exists in my FC-Lists and if it does, the button would have changed to an Edit Card button and by tapping it I could directly go to the Edit Card Screen (maybe nevertheless continue to allow to simply add another category by tapping and holding the Edit Card button).
However, in Preferences, for Duplicate cards option set to "allow", an option for a separate Edit Card button in the toolbar would be necessary. (Still, also in this mode, the Add to FC button could change depending on whether a card with the same headword and pinyin as the currently displayed dict entry already exists or not)

In short, no matter how you finally implement it, the two important functions are:
a. Being able to identify whether a card with the same headword and pinyin as the currently displayed dict entry already exists or not (for example by changing the colour of the Add to FCs button or by changing the style/colour of a separate Edit Card button, which would for example be a faded colour [and therefore also not work] if no card exists yet and change to full colour if a card does already exist).
b. Have the possibilty to go to the Edit Card screen for any currently displayed dict entry for which a card with the same headword and pinyin already exists in my FCs, no matter whether the Card has just newly been created or already a while ago.

#2 One thing that I didn't know where to post and didn't want to start a new thread for is concerning this forum. How about adding the possibility to rate other users posts (for example on a scale from 1 to 6 and also show a count for the total number of ratings). By doing so it would be easy for you to find out how many users actually agree with a specific feature suggestion or likewise how many users are experiencing similar bugs etc. Only registered users should be allowed to rate a post, in order to make sure that one person can only rate one single post once.

 

[renovator]

#2 One thing that I didn't know where to post and didn't want to start a new thread for is concerning this forum. How about adding the possibility to rate other users posts (for example on a scale from 1 to 6 and also show a count for the total number of ratings). By doing so it would be easy for you to find out how many users actually agree with a specific feature suggestion or likewise how many users are experiencing similar bugs etc. Only registered users should be allowed to rate a post, in order to make sure that one person can only rate one single post once.

How about charging a very nominal fee for an upgrade? People always want all sorts of stuff half of which they will actually never use, but ask for it because they get it free of charge. If a person thinks they want something bad enough, they should be willing to pay $5 or $10 for the upgrade. The requested upgrades with the highest total $ would be programmed first. This would help schedule programming priorities as well as help defray programming costs. Most top of the line software on the market charges something for an upgraded version.

 

[mikelove]

sinoreen - changing the Add to Flash icon is a definite no-go, unfortunately, because people are extremely sensitive to how quickly dictionary entries come up and searching for cards that match the current entry would create a noticeable lag (much more than changing the audio icon, for example) - even for someone like yourself who'd get a lot of benefit out of the icon changing, the slowdown would be significant enough that I think you'd end up disabling it. Same goes with updating the potential Edit Card icon.

About the only two options that would be practical for now are either to include an Edit Card button in the duplicate card popup, or add an Edit Card button to the toolbar which would do nothing if the card isn't already a duplicate - since that could get kind of annoying without an icon to indicate whether or not that duplicate existed, I think the first option would probably be the best bet for now, assuming we do in fact upgrade the duplicate alert in general in 2.0.1.

As far as rating posts, to be honest I don't think that would generate enough data to be useful - a dozen or so ratings is not a statistically meaningful sample, particularly given that PlecoForums users tend to be a lot more advanced / use a lot more of the software's features than our average customer. We are planning to do a customer survey (probably along with releasing 2.0.1) to try to get some information about larger-scale new features (big things like text-to-speech, desktop / online integration and the like), but for smaller ones I think we pretty much have to rely on our own judgment.

renovator - interesting idea, but it would be tough to get working in practice. The idea of charging for software version upgrades in general is something we've gone back and forth about - the last two big ones were accompanied by a couple of big expensive new dictionaries the sales of which should quite handily cover our development costs, but that's not necessarily going to be the case in future upgrades. At the same time, making upgrades free gets a lot more people using a new version right away, which means they're also showing it to their friends / reviewing it on their blogs and generating a whole lot of word-of-mouth sales as a result of that.

As far as charging for specific features, though, with that I worry that it would both make things too complicated (we already have WAY too many upgrade options, even with most of them just add-on dictionaries, and people quite reasonably find sorting through all of them very confusing) and would also make us appear very stingy and penny-pinching - people don't like to buy a piece of software only to find that when they try to do something it immediately tells them they have to buy such-and-such add-on pack to use it. The accounting software QuickBooks, for example, from the moment you start using it begins aggressively hawking various expansion packs / add-ons, payroll service and the like, and I for one find all of those ads very annoying, so I wouldn't want to have a situation in Pleco where you have to shell out $10 to enable some cool new flashcard panel.

 

[renovator]

- interesting idea, but it would be tough to get working in practice. The idea of charging for software version upgrades in general is something we've gone back and forth about - the last two big ones were accompanied by a couple of big expensive new dictionaries the sales of which should quite handily cover our development costs, but that's not necessarily going to be the case in future upgrades. At the same time, making upgrades free gets a lot more people using a new version right away, which means they're also showing it to their friends / reviewing it on their blogs and generating a whole lot of word-of-mouth sales as a result of that.

As far as charging for specific features, though, with that I worry that it would both make things too complicated (we already have WAY too many upgrade options, even with most of them just add-on dictionaries, and people quite reasonably find sorting through all of them very confusing) and would also make us appear very stingy and penny-pinching - people don't like to buy a piece of software only to find that when they try to do something it immediately tells them they have to buy such-and-such add-on pack to use it.

As long as the upgraded versions are inexpensive I would think $20 or less should be no problem for a new upgraded version. For those reviewing on blogs, etc. just set up a comp list as it is easy enough for you to spider google and figure out who those people are and give it to them no charge. It has always worked for the magazine industry, should work for you as well.

As far as specific features, I was thinking more of scheduling your programming time. The more people wiling to pay a nominal fee to have a feature appear in the next version, the more interested the general public would be in that feature. If people were willing to pay $5 or $10, you would charge that to anyone that wants the specific feature in the next version release. The mere fact of paying a nominal fee, any amount for that matter, shows that someone is seriously interested in that specific feature. These special features would all be included in the next release 2.0.2, 2.03, etc and people that had paid for any of the specific programmed features get the next version for free or at a discount. I did not intend to suggest that you would order a special version with just the options an individual wants.