OCR!

[Entropy]

In the middle third of the display. IOW, divide the current camera space in half, recognizer on top, recognized characters underneath.

That's a pretty thin band left for the camera preview - are you sure you'd be able to see enough to find where you were on the page?

Well, it's pretty hard to do that now, so I don't think it would get worse for me anyway.

On the iPad, you could divide the screen into panes to allow the user to edit text while recognizing.

Maybe we could display the image alongside the recognized text, but "while recognizing" is a bit of a stretch.

Not literally while the recognizer is doing its work. I was thinking that one would scroll the image, select text, recognize, edit (if there were errors) and then proceed with another portion of the image.

A specialized menu filter might help, but as I noted above we can't do the filtering at the engine level so we'd have to screen the character results to try to find menu words.

That doesn't seem like it would be a problem to me. Of course, compiling the list of menu characters is something I should get to work on, I suppose.

Is there a technical reason I need to switch between lookup, flashcard and document capture modes? Modes are bad. I'd rather just have three buttons in the toolbar.

Which toolbar would you propose adding them to? (not like we have a lot of extra room, and this a relatively infrequently-made switch, not to mention the fact that we might vary the user interface between modes more in future releases)

The one that currently has three ridiculously oversized buttons in it? (Focus/Pause/&c)

In capture mode, turning on ""hide unused chars" leads to *no* characters being displayed, green or blue. There's still a translation, though.

That might actually be preferable under the circumstances.

Except that I might be a better judge of whether it's gotten the right character. Which suggests to me that I might want a "try again" button that eliminates the wrong match. I bet that's a technical problem, though.

I don't want to have a bunch of English characters recognized, even if they're written on the menu.

So you'd rather have it leave those spaces blank? Difficult to filter since occasionally one or two English characters are recognized as Chinese ones.

Yes, at least if those are on separate lines. If there's something inline like XO, I guess that should be captured. But for the most part, it seems to introduce random junk to the translated text.

If there's formatting, I want captured text to reflect that. (I don't know whether it already does.) I'd like to be able to capture a whole menu page at one time, and get individual items on individual lines. In fact, it would be great if Pleco could recognize vertical text and rearrange it as horizontal lines and vice versa (the latter in the editor, not OCR.)

Can't do much for formatting, but we do recognize / rearrange vertical text even with still image capture.

I just want newlines at the ends of short lines. I'm not sure what happens now since I can't get past the camera shake problems to recognise a large block of text.

Desktop version may actually be forthcoming, I'm thinking a $20 or $30 Chinese OCR app for Mac would be an interesting little way of experimenting with native desktop Mac development without committing to bringing the full version of Pleco to it. Along with perhaps a $10 to $20 drop-in replacement for Apple's trackpad handwriting recognizer that used our more accurate engine and two-finger clear gesture.

If you could make it work with an external Magic Trackpad, while allowing the internal trackpad and/or mouse to continue working, that would be a huge win. Apple's implementation of HWR on the Mac is annoying to use, because you can't do anything else while the HWR is active. $30 seems like the right price to pay for a simple two-trick pony. And of course you have a LOT more processing power to work with on the desktop, even on a mini MacBook Air.

~ Kiran <entropy@io.com>

 

[mikelove]

I just bought OCR for my iPod Touch 4g, and guess what - it's not all gloom and doom - I'm really happy with it! It's true that OCR would be much better with auto-focus to identify some of the smaller more complicated characters. However, OCR still works decently in my opinion even without auto-focus. Perhaps part of the reason I'm happy is that all of Mr. Love's scary disclaimers set my expectations really really low for OCR on the iPod Touch 4g; however, despite all that - I really think OCR is still useful and good to have as part of Pleco on an iPod Touch 4g.

Great! I'd rather be too pessimistic than promise too much, but I'm certainly delighted to see some iPod users finding it helpful.

Question 1: Sometimes while I am looking up a word with OCR, the green characters OCR is trying to lock on to are changing too fast for me to pause on the correct one. Is there a way to make the characters change more slowly, so I don't feel like I'm playing Whack'a-Mole.

If you make the box bigger and turn off the the "shrink big images" option in Settings / OCR / Lookup Words, that should help matters a little, but we're working on a much better fix for 2.2.(2?). Ultimately this is a function of the poor focus, though; a less sharp character makes the recognizer less confident in its match's accuracy and hence more likely to encounter a false positive.

Question 2: I looked in the instruction manual, but I didn't find an explanation for the OCR settings "Word detect samples" and "Word detect compare length." Can you tell me what those settings do?

It's a little hard to find but it's in the Settings section:

Word detect samples: configure how many frames in a row the recognizer has to see the same text before "locking on" to it; increasing this number will make the definition change more slowly / become less "jittery" in general, while decreasing it will make it more responsive.

Word detect compare length: configure the number of characters at the start of the word that have to match in order for the system to "lock on"; consider increasing this number if you capture a lot of long words.

mikelove wrote:
That's a pretty thin band left for the camera preview - are you sure you'd be able to see enough to find where you were on the page?

Well, it's pretty hard to do that now, so I don't think it would get worse for me anyway.

Makes sense.

Not literally while the recognizer is doing its work. I was thinking that one would scroll the image, select text, recognize, edit (if there were errors) and then proceed with another portion of the image.

So you'd basically be selecting a block of text in the image and tapping on a button to append it to your text file? Would you want this inserted at the insertion point location or the end of the file?

That doesn't seem like it would be a problem to me. Of course, compiling the list of menu characters is something I should get to work on, I suppose.

We might actually want to support arbitrary character lists for this - people can define their own using the flashcard system (or something like it).

The one that currently has three ridiculously oversized buttons in it? (Focus/Pause/&c)

They're oversized to make them easy to hit - we've gotten very few complaints, anyway, and even fewer about the equally large buttons in flashcards.

Yes, at least if those are on separate lines. If there's something inline like XO, I guess that should be captured. But for the most part, it seems to introduce random junk to the translated text.

Doesn't that junk often also take the form of Chinese characters?

I just want newlines at the ends of short lines. I'm not sure what happens now since I can't get past the camera shake problems to recognise a large block of text.

You should already get those in Capture Text, actually.

If you could make it work with an external Magic Trackpad, while allowing the internal trackpad and/or mouse to continue working, that would be a huge win. Apple's implementation of HWR on the Mac is annoying to use, because you can't do anything else while the HWR is active. $30 seems like the right price to pay for a simple two-trick pony. And of course you have a LOT more processing power to work with on the desktop, even on a mini MacBook Air.

That would be the idea, basically, a much better version of trackpad handwriting optimized for use with a Magic Trackpad or on a laptop trackpad alongside an external mouse. (though we could probably also add a gesture to quickly turn it on / off and keep using the pad for mousing)

 

[kun4]

When reading a book or magazine with OCR, what is the usual (or suggested) distance between book and iPhone? I'm merely looking for a ballpark figure; a distance most people feel comfortable with.

 

[Entropy]

I was thinking that one would scroll the image, select text, recognize, edit (if there were errors) and then proceed with another portion of the image.

So you'd basically be selecting a block of text in the image and tapping on a button to append it to your text file? Would you want this inserted at the insertion point location or the end of the file?

Good question, but I was thinking end of the file. Basically, I want to be able to check the recognizer's work, since some people claim still image rec is potentially unreliable. I guess this will need three panes, one for the HWR for correcting the text in question. (Of course, if there's a Mac version I could just use that for this kind of translation.) Using an iPhone as a HWR input device for the iPad would be helpful here.

Doesn't that junk often also take the form of Chinese characters?

Huh. So it does. But it also gets Western characters from adjacent lines. I guess you could filter them out when the text is actually captured, rather than in the recognizer.

optimized for use with a Magic Trackpad or on a laptop trackpad alongside an external mouse. (though we could probably also add a gesture to quickly turn it on / off and keep using the pad for mousing)

Three finger tap to turn it off and on! Or maybe an eleven finger click :-|

But the killer feature will be to figure out how to let me write on one input device while using my computer normally in every other way. If you can do that, you could also let me attach a second keyboard for pinyin input.

increasing this number will make the definition change more slowly / become less "jittery" in general, while decreasing it will make it more responsive.

I turned this up to 5 in the live mode and it seems much nicer now. (Also turned it up in the other modes.) It's still plenty responsive. BTW, it might be nice to have a mode that shows pinyin in place of the definition.

They're oversized to make them easy to hit - we've gotten very few complaints, anyway, and even fewer about the equally large buttons in flashcards.

That wasn't meant as a complaint. I think they'd be easy enough to hit even if there were 4 instead of 3--I want to capture the very text I'm currently reading (the whole text--capturing a flashcard while in pause mode would presumably only add the first word.) And, in pause mode I want to be able to tap a character or word and perform all the usual actions, as if i were in the reader.

5 buttons might be too much, I have go admit. But if I capture text, I can presumably make flashcards later, so that seems like the obvious choice. Anyway, why not just tap anywhere on the screen to pause/capture/&c? Or, tap (two-finger tap?) on the screen to move to the next word in a block of live text.

BTW, I think that "Docs+OCR" would be better than "Read+OCR" since the documentation is there. And, the manual should be downloaded by the app so it's available offline, at least when you purchase and download another add-on.

~ Kiran <entropy@io.com>

 

[tennonn8]

I tried OCR on my iPod Touch 4G recently. I tried the demo version first. (As I said before, I was determined to buy it anyway, but I was just curious to see what the demo version looked like.) After I tried the demo, I returned to the Add-Ons page and clicked the "Buy" button. At this point Pleco froze completely, didn't respond to any clicks. It could be caused partially by a network problem, because after I restarted Pleco, it said that the network wasn't accessible, and the App Store didn't work as well. So I rebooted the device and restarted the WiFi network and then I was able to buy the OCR successfully.

I think it was basically a bug. Even though I said above that it could be partially intensified by a network problem, well, usually, when a network is inaccessible, we see that "rotating asterisks" or whatever it is called, indicating that something is going on, and then after a while an error message is displayed. But in my case, everything just froze.

As for the OCR itself, I agree with the user "nihao pengyou" here, that after our expectations were set so low before, the actual performance is really impressive. For me, it worked very nice with subtitles, you just have to get used to it a little bit. It has problems when subtitles are shown over a heterogeneous, speckled background or when the character looks "out of focus" by itself already, when you can hardly see the spaces between the strokes. But I would never expect any OCR program to be any good under such circumstances! But otherwise, the OCR proved to be very effective with subtitles. Sometimes, I was even quite surprised that the OCR recognized the characters correctly even though they looked pretty fuzzy to me. When I looked closer, I realized that the fuzziness was caused by a rather strong gradient in the color intensity along the width of the stroke. Apparently, it was not a problem for the OCR as long as the spaces between the strokes remained clear.

All in all, the overall impression is very positive, it seems to be a very useful new tool even on an "unsupported" iPod Touch. Thanks guys!

 

[candice]

Hello Mike,

I just purchased the complete bundle for iOS, but the OCR add-on does not appear as an add-on which I am able to download and purchase.

I have an Iphone 4 that is running on a China Unicom contract in mainland China. Could that have sth to do with it, or the fact that i am seeking to purchase it within the Mainland .... or that the phone itself somehow does not allow me to download the OCR, because it is linked to a specific carrier, or problems with Itunes for purchases from within China?...can't seem to figure it out

Just seems odd and I really would like the OCR feature...one of my key criteria for purchasing ....as i already have an old non-camera iphone with pleco installed (transferred from my windows mobile).

Your feedback would be greatly appreciated.

Thanks

 

[mikelove]

When reading a book or magazine with OCR, what is the usual (or suggested) distance between book and iPhone? I'm merely looking for a ballpark figure; a distance most people feel comfortable with.

FWIW, I'd say that mostly depends on the size of the font in the book you're reading, and on how steadily you can hold your phone (steadier -> able to keep it pointed at text with a larger zoom factor -> farther away).

Good question, but I was thinking end of the file. Basically, I want to be able to check the recognizer's work, since some people claim still image rec is potentially unreliable. I guess this will need three panes, one for the HWR for correcting the text in question. (Of course, if there's a Mac version I could just use that for this kind of translation.) Using an iPhone as a HWR input device for the iPad would be helpful here.

Makes sense. Quite well aware of your interest in using an iPhone for iPad handwriting input

Huh. So it does. But it also gets Western characters from adjacent lines. I guess you could filter them out when the text is actually captured, rather than in the recognizer.

I suppose, but the presence of Western characters might make it easier to quickly spot a mis-recognized line... I don't see much benefit to skipping them, anyway, it's rare that you'd have a mis-recognized document where the mis-recognized bits exclusively took the form of out-of-place A-Z letters and there weren't also few stray Chinese characters mixed in. (heck, even when recognizing English text the system sometimes has a tendency to misinterpret a few of them as Chinese characters)

But the killer feature will be to figure out how to let me write on one input device while using my computer normally in every other way. If you can do that, you could also let me attach a second keyboard for pinyin input.

That would be the benefit to the iPhone-as-external-Mac-input-device, I suppose, though the other app that did that with the iPhone's built-in recognizer doesn't seem to have done very well.

I turned this up to 5 in the live mode and it seems much nicer now. (Also turned it up in the other modes.) It's still plenty responsive. BTW, it might be nice to have a mode that shows pinyin in place of the definition.

OK, we might want to consider increasing the default... and only Pinyin, no definition? I guess it would be a bridge to reading Chinese for people who were better on spoken Chinese than on characters, though we'd have to accept a certain degree of inaccuracy in ambiguous cases (multiple pronunciations / meanings for the same character, with the definition the easiest way to disambiguate).

That wasn't meant as a complaint. I think they'd be easy enough to hit even if there were 4 instead of 3--I want to capture the very text I'm currently reading (the whole text--capturing a flashcard while in pause mode would presumably only add the first word.) And, in pause mode I want to be able to tap a character or word and perform all the usual actions, as if i were in the reader.

Maybe we could have three little tabs on the side like we do in flashcards?

5 buttons might be too much, I have go admit. But if I capture text, I can presumably make flashcards later, so that seems like the obvious choice. Anyway, why not just tap anywhere on the screen to pause/capture/&c? Or, tap (two-finger tap?) on the screen to move to the next word in a block of live text.

We actually did have people tap on the screen to pause capture in Beta 1, but we changed it because it was making it too difficult to resize the box - now we have extremely generous tap areas for the box corners, you can touch almost everywhere and have it anchor your touch to the nearest dot.

BTW, I think that "Docs+OCR" would be better than "Read+OCR" since the documentation is there. And, the manual should be downloaded by the app so it's available offline, at least when you purchase and download another add-on.

That might make it harder to find the reader, though - we really need a way to label all three. Though if we replace the tab bar with a screen that contains more icons (an idea I'm liking more and more) we can fix the problem easily that way by not combining too many things into each tab.

The manual already can be downloaded for offline use, see the "Free" tab in Add-ons. (new feature in 2.2)

I think it was basically a bug. Even though I said above that it could be partially intensified by a network problem, well, usually, when a network is inaccessible, we see that "rotating asterisks" or whatever it is called, indicating that something is going on, and then after a while an error message is displayed. But in my case, everything just froze.

That's odd... basically all we do is tell iTunes the user has initiated a purchase and then put up that spinning icon until it responds, so it seems like it must have frozen in the middle of that initiate-purchase command. Could be an Apple bug, actually, the OCR bug on 3GSes running 4.2 took exactly that form (call into an Apple function that's supposed to run in the background and it hangs). I suspect they've been busily updating different parts of iOS to improve its multitasking support, in anticipation of the arrival of dual-core mobile processors in 2011 - problems like that are far and away the most common / infuriating type of bug in multithreaded programming.

As for the OCR itself, I agree with the user "nihao pengyou" here, that after our expectations were set so low before, the actual performance is really impressive. For me, it worked very nice with subtitles, you just have to get used to it a little bit. It has problems when subtitles are shown over a heterogeneous, speckled background or when the character looks "out of focus" by itself already, when you can hardly see the spaces between the strokes. But I would never expect any OCR program to be any good under such circumstances! But otherwise, the OCR proved to be very effective with subtitles. Sometimes, I was even quite surprised that the OCR recognized the characters correctly even though they looked pretty fuzzy to me. When I looked closer, I realized that the fuzziness was caused by a rather strong gradient in the color intensity along the width of the stroke. Apparently, it was not a problem for the OCR as long as the spaces between the strokes remained clear.

All in all, the overall impression is very positive, it seems to be a very useful new tool even on an "unsupported" iPod Touch. Thanks guys!

Great! It's sounding more and more like we want to change this from totally unofficial to semi-official - still require jumping through a few more hoops, but maybe not be quite as discouraging of it as we are now. (I'll have to do a database check in a few days to see what percentage of people activating their iPods to enable OCR actually ended up purchasing it)

Hello Mike,

I just purchased the complete bundle for iOS, but the OCR add-on does not appear as an add-on which I am able to download and purchase.

I have an Iphone 4 that is running on a China Unicom contract in mainland China. Could that have sth to do with it, or the fact that i am seeking to purchase it within the Mainland .... or that the phone itself somehow does not allow me to download the OCR, because it is linked to a specific carrier, or problems with Itunes for purchases from within China?...can't seem to figure it out

Just seems odd and I really would like the OCR feature...one of my key criteria for purchasing ....as i already have an old non-camera iphone with pleco installed (transferred from my windows mobile).

First off, are you running the latest version of Pleco (2.2.0)? You can check on that in the Settings tab / About - OCR won't work at all in earlier versions. If you're running an older version, you can update via the "Updates" tab in App Store.

If you are running the latest version, I'm not sure why OCR wouldn't be coming up in the file catalog - the catalog listing is actually generated entirely by our server, not by Apple's (they just fill in the prices), so there's not really any possibility that a particular iTunes store / phone localization / etc might be getting in the way. Do other catalog items (e.g. those in the "Free" section of Add-ons) still show up correctly?

 

[x2o]

I'll have to do a database check in a few days to see what percentage of people activating their iPods to enable OCR actually ended up purchasing it

I tried out the demo on my touch and was definitely pleased with the recognition capabilities, given sufficient lighting and reasonably sized characters. The only thing stopping me from purchasing it on the spot was that the disclaimer warned that future updates might make OCR completely incompatible for ipod touches and the iphone 3g - how likely is this? I know you have to cover all bases, considering that OCR is intended for devices with autofocus, but I'm more than willing to buy OCR if there is a good chance it will remain functional for my device.

Thanks!

 

[Entropy]

But the killer feature will be to figure out how to let me write on one input device while using my computer normally in every other way. If you can do that, you could also let me attach a second keyboard for pinyin input.

That would be the benefit to the iPhone-as-external-Mac-input-device, I suppose

Hah, in *this* case I was referring to either a Magic Trackpad, Magic Mouse, or for that matter drawing tablets, which Apple's HWR completely ignores.

only Pinyin, no definition? I guess it would be a bridge to reading Chinese for people who were better on spoken Chinese than on characters,

I'm thinking of the case where I recognise the characters, but don't know how to pronounce them, such as 水煮牛. Yummy!

Maybe we could have three little tabs on the side like we do in flashcards

Flashcards have tabs? Oh, so they do! I like that a lot.

The manual already can be downloaded for offline use, see the "Free" tab in Add-ons. (new feature in 2.2)

Right, I'm just saying this should happen as invisibly as possible.

~ Kiran <entropy@io.com>

 

[jiacheng]

I have some suggestions for the UI for capturing blocks of text on either the capture text mode or any of the still image capture modes.

Basically, whatever image is being OCR'd could be shown with or without the the green recognized character overlays. You could zoom in and see what the original character or word that was recognized in that block of text. This way you could pick out mistakes more easily at a glance, and tap a button that would give alternative suggestions, or allow you to enter a custom correction.

Perhaps there should also be a way to highlight the characters that the OCR system was less confident about and call these to the attention of the user and allow more efficient scanning for potential mistakes.

Or maybe a split screen mode in the reader showing the text as well as the image of the currently highlighted word/character.

 

[mikelove]

I tried out the demo on my touch and was definitely pleased with the recognition capabilities, given sufficient lighting and reasonably sized characters. The only thing stopping me from purchasing it on the spot was that the disclaimer warned that future updates might make OCR completely incompatible for ipod touches and the iphone 3g - how likely is this? I know you have to cover all bases, considering that OCR is intended for devices with autofocus, but I'm more than willing to buy OCR if there is a good chance it will remain functional for my device.

It's very unlikely we'd reduce compatibility on the iPod Touch 4G - the iPhone 3G can barely use OCR now due to memory and processor-related constraints, so it's quite possible that in a future release the memory or processor requirements might increase to a point where OCR would no longer run at all on the 3G (we couldn't really justify devoting a lot of time to lowering them for the sake of a handful of 3G OCR users), but the iPod has as much RAM as the 3GS (albeit with a higher-resolution screen) and a faster processor, so the odds of us running into a constraint on the iPod that didn't also affect our officially-supported 3GS users are pretty low.

I'm thinking of the case where I recognise the characters, but don't know how to pronounce them, such as 水煮牛. Yummy!

Makes sense, though we'd run into the usual not-always-clear-what-the-Pinyin-for-a-particular-character-is problem.

Flashcards have tabs? Oh, so they do! I like that a lot.

Thanks! That one has generated a few complaints, actually, mostly concerning the particular functions we've chosen to stick in the other two tabs, but I think it'll remain the default even if we add an option for people who want the back / undo / card info / category / etc functions available all the time to make them so.

mikelove wrote:
The manual already can be downloaded for offline use, see the "Free" tab in Add-ons. (new feature in 2.2)

Right, I'm just saying this should happen as invisibly as possible.

I suppose we could add a prompt to automatically download the manual in the background when they switch to it, but I'm not entirely happy with that idea - it interrupts the user when they're trying to do something and it adds a confusing alternate way to invoke the add-on download system.

Basically, whatever image is being OCR'd could be shown with or without the the green recognized character overlays. You could zoom in and see what the original character or word that was recognized in that block of text. This way you could pick out mistakes more easily at a glance, and tap a button that would give alternative suggestions, or allow you to enter a custom correction.

Perhaps there should also be a way to highlight the characters that the OCR system was less confident about and call these to the attention of the user and allow more efficient scanning for potential mistakes.

Or maybe a split screen mode in the reader showing the text as well as the image of the currently highlighted word/character.

We won't actually be able to get confidence values out of the system for another month or so (they're working on an update for us that adds support for retrieving them), but it certainly would make sense to highlight low-confidence once we can recognize them.

Combining the image of the character with the recognized character is definitely something we're planning, though it's not a perfect fix for recognition errors since sometimes the problem is that the system mis-identifies the boundaries of a character (combines multiple characters into one, or splits a single character into two) - we'd still need a way to manually edit the finished text output too.

I do think we need to something different with the UI for Capture Text - right now it feels a bit too slow to update even though it's a good bit faster than still-image capture (and needs to be in order to be useful). Maybe we should replace the pause button with a button to capture / instantly(?) recognize the text that's currently in the box and not recognize it until then? But that would lose some of the auto-correcting benefits of seeing that overlay live as it is now...

 

[Entropy]

I suppose we could add a prompt to automatically download the manual in the background when they switch to it, but I'm not entirely happy with that idea

"You seem to be using Pleco for the first time. Would you like to download the manual? Cool! You'll find it in the "Reader" tab, which you get to by..." and so forth.

Many users probably don't realize it isn't built-in until they can't use it due to lack of the network.

~ Kiran <entropy@io.com>

 

[mikelove]

"You seem to be using Pleco for the first time. Would you like to download the manual? Cool! You'll find it in the "Reader" tab, which you get to by..." and so forth.

Many users probably don't realize it isn't built-in until they can't use it due to lack of the network.

Well there are other ways to fix that - I'm just a little worried about "alert fatigue," we can get away with maybe one or two startup alerts before people start ignoring them - pinning any kind of vital functionality to any alert after the first one is probably a bad idea.

 

[character]

Well there are other ways to fix that - I'm just a little worried about "alert fatigue," we can get away with maybe one or two startup alerts before people start ignoring them - pinning any kind of vital functionality to any alert after the first one is probably a bad idea.

An introductory dialog (doesn't have to look like an alert) could point out the major features and offer to download the manual. I agree having the manual on board would help a lot of people; I can understand not bundling it with the app due to size.

 

[character]

Mike, Pleco.com doesn't feature OCR first thing -- you might want to embed your video instead of (or in between) the screenshots.

 

[mikelove]

An introductory dialog (doesn't have to look like an alert) could point out the major features and offer to download the manual. I agree having the manual on board would help a lot of people; I can understand not bundling it with the app due to size.

We have one when the app is first run; it's already as long as it can be without scrolling, but its main job is simply to direct people to the very important fan icon (without which they're not going to be able to access help or anything else outside of the main dictionary). I think the best fix for this will probably just be to have a dedicated help button in the new "wall of icons" thing in 2.3.

Mike, Pleco.com doesn't feature OCR first thing -- you might want to embed your video instead of (or in between) the screenshots.

Honestly I hate embedded video on web pages, at least when you're not expecting it - slows page loading way down and disrupts the whole layout. An OCR screenshot would probably be a good idea though.

 

[tanyahart]

It's beautiful! I just saw the video on the ChinesePod mailing list and ran off to buy Pleco. I didn't realise it was out for iPhone yet. I live in Hong Kong and I used it on a street sign last night (I'm so excited i'm showing it off to everyone i ever met). I'll try to take a photo if you like. It also read the pale-green-on-white text in a brochure (but didn't recognise 'Latvia'... maybe when I've got a proper dictionary installed).

Hey, I'm totally new to Pleco. I've downloaded the demo of the basic package and I can't find the Flash cards. Are they disabled in the demo? I'm planning to buy it, just wanted to try it first.

And another off-topic question: Are any of the dictionaries Cantonese? With Yale or Jyutping? I've been using Qing Wen app which is very simple but has audio option for Cantonese.

Soooo cool.

Best, Tanya

PS On the green issue, it's ugly, but it stands out against all the background stuff, it's fine. It's less aesthetics than clarity that you need.

 

[mikelove]

It's beautiful! I just saw the video on the ChinesePod mailing list and ran off to buy Pleco. I didn't realise it was out for iPhone yet. I live in Hong Kong and I used it on a street sign last night (I'm so excited i'm showing it off to everyone i ever met). I'll try to take a photo if you like. It also read the pale-green-on-white text in a brochure (but didn't recognise 'Latvia'... maybe when I've got a proper dictionary installed).

Thanks! Really happy to hear that... we haven't gotten that many "field reports," actually, most people just seem to use it for reading novels and such (which to be honest is the main thing we designed it for). Photos are certainly welcome, we should probably set up a Facebook album or something for that...

Hey, I'm totally new to Pleco. I've downloaded the demo of the basic package and I can't find the Flash cards. Are they disabled in the demo? I'm planning to buy it, just wanted to try it first.

They're not available in the demo yet, unfortunately - haven't figured out a good way to demo them without giving away the whole thing, though we're committed to adding one in 2.3 (along with redesigning a lot of the flashcard configuration UI).

And another off-topic question: Are any of the dictionaries Cantonese? With Yale or Jyutping? I've been using Qing Wen app which is very simple but has audio option for Cantonese.

Not yet - we didn't want to rely too much on the character-by-character approach that they use since it's not all that accurate, though you can pull up Cantonese pronunciations for individual characters in the Character Info screen; tap on a character in a dictionary entry, tap on the 字 button at the top right corner of the screen, tap "Details," scroll down and tap on "Add New Field...," then select "Cantonese" and after that you'll see Cantonese included on that "Details" page.

We are hard at work on licensing a proper Cantonese dictionary, but it's been very slow going since there aren't that many out there - we think we've got a tentative deal for one now, though, so hopefully we'll finally get that signed / get started on development in early 2011...

 

[tanyahart]

Here you go.
http://gallery.me.com/tanya_hart#100104&bgcolor=black&view=mosaic&sel=1
(Can't see a way to upload a pic, bit new to forums).

Feel free to download them and stick them on facebook or wherever. Any requests?

I'm having fun testing it out on unreasonable surface and light situations. So far it even managed to get the fancy calligraphy style text high up on the outer wall of the school. The main problem was holding the iphone steady from 150m. The black on silver lift signs were hard to read where the black has chipped off. That's pretty reasonable.

I have a little obsession with signs, especially the ones that give instructions. I love that reassuring authoritative tone and occasional absurdist style. 'Please wrap spittle'.

The main thing I need it for is to help read my kid's Kindy homework. My Cantonese is struggling to keep up with 'K2' level. Hey, you have at least one Canto-lexicographer fan, he's the one who told me about Pleco back when your website said 'iPhone users can just forget it' or something equally discouraging that had me not check back for 3 years... I did find that tab for Cantonese but it's so far awaaaaay... so many taps.

Best, Tanya.

 

[mikelove]

Here you go.
http://gallery.me.com/tanya_hart#100104 ... saic&sel=1
(Can't see a way to upload a pic, bit new to forums).

Thanks!

I'm having fun testing it out on unreasonable surface and light situations. So far it even managed to get the fancy calligraphy style text high up on the outer wall of the school. The main problem was holding the iphone steady from 150m. The black on silver lift signs were hard to read where the black has chipped off. That's pretty reasonable.

Cool! We really need to find some way to put an iPhone on a Steadicam mount, I think (actually they already seem to offer one though it's probably rather expensive)

I have a little obsession with signs, especially the ones that give instructions. I love that reassuring authoritative tone and occasional absurdist style. 'Please wrap spittle'.

I imagine that's a challenge for Chinese sign makers, actually - it's difficult to word something like that in Chinese in anything other than a flat / authoritative tone without making the language very frilly...

The main thing I need it for is to help read my kid's Kindy homework. My Cantonese is struggling to keep up with 'K2' level. Hey, you have at least one Canto-lexicographer fan, he's the one who told me about Pleco back when your website said 'iPhone users can just forget it' or something equally discouraging that had me not check back for 3 years... I did find that tab for Cantonese but it's so far awaaaaay... so many taps.

Oh we've got at least a few of them, we've just been apologizing to them for the lack of a Cantonese dictionary for years and years... and yeah, at one point we were saying iPhone was a no-go for the same reason that we're saying that about BlackBerry now; there simply wasn't a good way to put something like Pleco on iPhone in the year or so before they released their first native-code software development kit.