3.0 New Dictionary Bug Report / Feedback Thread

[TimBenderSLS]

Makes sense - definitely does not seem worth the money and effort to try to track such things down one by one (nor does it seem valuable - any compound not present in the modern dictionaries is going to be questionable enough in the pronunciation that I wouldn't place much stock in the decision of any particular person you hired, unless it came from the original editors/publication, and even then I'd check it for reasonableness). I was just wondering if it was a bug in incorporating already existing data, which I now know it is not. Much rather you guys spend the effort developing new tools and licensing more dictionaries (the point of hydcd is not really to get pronunciations of compound words anyway)

One more thing I noticed today: I think there might be something odd going in with radicals and simplified/traditional conversion - 言 in hydcd shows a simplified version of 讠 (so, for example, if you look 言, the integrated entries for all other dictionaries are separated from the results from HYDCD in a separate tab)

 

[mikelove]

Well at the very least we'd like to make sure the Pinyin matches up with other dictionaries that cover the same word, so that merged searches work correctly - that one's a basic functional issue.

Good note on 言, definitely shouldn't be converted.

 

[heiban]

• When looking up a character, is it possible to make it so dragging your finger changes the tab view between DICT, STROKE, CHARS,WORDS,SENTS?I know dragging your finger from the edge makes it go backwards, but dragging in between the edges doesn't do anything right now. This feature just make it more intuitive..

Edit; I loaned the screenshot from one of the replies on page #1 so you know what view I'm talking about:

• When pushing "play" to view strokes, why not make it go continually? The fast forward button is there to make it go to the end anyway if we want it to stay at the end. It easier to write it down on the paper when its continually repeating it with out having to press play all the time. Also, make it "play" immediately when we push the "STROKE" view, because that's the point with the stroke view after all right? To see the strokes?

 

[TimBenderSLS]

I am very curious - what is the nature of the 2010 supplement currently being incorporated? Is it just additional entries, or also corrections of known errors? For example, HYDCD's entry for 汉奸 quotes a Song dynasty text 《玉照新志》 for "檜 既陷此, 無以自存, 乃日侍於 漢 奸 戚悟室 之門。" - when in fact 汉奸 appears nowhere in any of the various versions of that text that we have, one of the editors apparently just wrote it down wrong (it wasn't until the late Qing dynasty that 汉奸 took on any connotation of 'traitor', certainly not in the Song). Since this is a well known misquote, it would seem if they were fixing any errors they might fix this one - is it still the same in the new supplement data?

Though, even if the supplement leaves in place the very occasional error - HYDCD's inclusion within pleco remains one of the greatest things to happen this year. One of my colleagues here told me that it 'changed his life'. I'm just really excited to see what cool new things the supplement introduces....

 

[mikelove]

heiban - to be honest, adding a separate behavior for scrolling within the screen feels confusing to me; too close to the regular back gesture and consequently too easy to trigger the wrong one by mistake. We also tend to be very nervous about assigning gestures because we might want them for something else later (and any time we take away something that some people found useful there tends to be a lot of blowback) - a swipe gesture for example might be useful with a future version of the stroke order tab that allowed switching between several different fonts.

Auto-playing on entering STROKE might make sense, looping continuously is the sort of behavior that might be useful but is going to feel like a bug to anyone who doesn't find it useful. At the very least there'd have to be a separate optional "loop" button or option - if we did it by default we'd get a *lot* of complaints.

TimBenderSLS - it's mostly additions rather than removals; they do include some removals / corrections but the vast majority of the entries are additions. This particular 汉奸 misquote does not seem to be covered, but since they pretty much let us do our own thing editorially speaking, we can make a note to apply that correction to our version anyway.

 

[TimBenderSLS]

Mike,

Oh, interesting. If you do decide to insert a pleco correction, here is a link to an article that discusses the mistake: http://www.cuhk.edu.hk/ics/21c/supplem/essay/0401034g.htm . In the section with the heading "清代以前无「汉奸」？" they go over the 7 different surviving versions of this text, showing that none match the HYDCD quote, and none include 汉奸. (Scholars of Qing often talk about this error, though this was the best thing I could quickly pull up online)

Hope everyone has a great thanksgiving,

 

[BrianTung]

A small thing: In the usage notes section for the 亂七八糟 entry in 多功能成語辭典, we have: "當形容 ‘事物糟糕，污穢，惡劣’ 時，宜用亂七八糟，如 ‘種種烏七八糟的傳聞’。" Shouldn't that first 亂 be a 烏? Are content errors bugs?

 

[mikelove]

TimBenderSLS - thanks.

BrianTung - content errors like this are often bugs, yes - in this case it appears that our conversion script may have failed to take into account that ~s are sometimes used to replace the second chengyu in a synonym comparison instead of the first.

 

[mikeo]

mikeo - hmm, that actually seems more like a software bug - thanks!

Any word on 3.01/fixing CY problem posted earlier ?

 

[mikelove]

Still working through the very long list of items - the basic problem is that there aren't really any super-high-priority ones that would necessitate a quick update to fix only those, there's a very long list of minor issues and we'd rather get through them all in one go then do it incrementally and leave the users whose particular issues weren't being fixed increasingly frustrated. (also more efficient time-wise this way since we only have to package up one new release)

 

[mikeo]

Still working through the very long list of items - the basic problem is that there aren't really any super-high-priority ones that would necessitate a quick update to fix only those, there's a very long list of minor issues and we'd rather get through them all in one go then do it incrementally and leave the users whose particular issues weren't being fixed increasingly frustrated. (also more efficient time-wise this way since we only have to package up one new release)

Makes sense, thanks.

Another instance of probably the same underlying (indexing?) issue with CY, screenshots attached. When you click " change dict entry" on first screen, following screen displays different, unrelated root phrase.

 

[dymtang]

21C dictionary, hither entry, "hither and you" should be "hither and yon" (just need to flip the 'u' upside down).

 

[mikelove]

Thanks.

 

[Ricky1194]

My ipad 3 is running 3.0, it is my iphone and on my iPad. On my ipad it keeps randomly crashing (or exiting, as it still shows up in open items) when doing flashcards..(at least that is where I have noticed it). Not sure what is causing it, but a direction to go with is the syncing as I have been going back and forth quite a bit between phone and pad a lot for flashcards.

 

[mikelove]

What version of iOS are you running on your iPad and on your iPhone? Could you send me your crash logs? This page has instructions for how to obtain those.

 

[Ricky1194]

What version of iOS are you running on your iPad and on your iPhone? Could you send me your crash logs? This page has instructions for how to obtain those.

I am running 7.0.4. I have attached the logs. Let me know if you need anything else. Thanks!

 

[mikelove]

Thanks - this is indeed an iCloud crash, it'll be fixed in 3.0.1 but in the meantime I'd recommend that you wait to make sure your flashcards have synced before starting a test and not make any changes on the other device while the test is running.

 

[dcarpent]

In going through a set of flashcards I came to a card for 修. After revealing the meaning I tapped the character because I wanted to go to the definition screen to look at some sample sentences. But as soon as I hit the arrow to move to that screen Pleco crashed. I restarted the same stack of flashcards, arrived at this same card, and did the same: Pleco crashed. I've re-created this about five times in a row, so I think it is a bug. I've tried doing the same with other characters on other cards in the same stack and there is no problem. Could this be specific to this character?

 

[mikelove]

Could you post crash logs? (see above for how to generate those)

 

[dymtang]

I purchased early access GZHFYCD dictionary and the Cantonese Audio module. When I click on the speaker icon of the sample sentences under GZHFYCD, am I supposed to hear it say in Cantonese? Right now, the sentences are rendered in (rather funny) Mandarin.