[Unofficial] Feature Request / Suggestion List

[Fernando]

Just out of curiosity I decided to do a search of what would be the ultimate JP-CN dictionary for Pleco. I came across this gem: Morohashi Tetsuji‘s Dai Kan Wa Jiten (大漢和辭典).

Seems to be the ultimate Borgian labyrinth in Chinese learning and has been fully digitized in 2018. The catch(es) is (are) that it’s windows only, requires input by radical or code and costs nothing less than 140,000 yen. That’s more than a 1000€ or $1200. It comes in a USB stick and has amazing features like zoom in and out for PDFs... I was like, bro, only the Japanese can get away with amazing content in a format straight from the 1990s…

大漢和辞典デジタル版 | 大修館書店 創業100周年記念企画

www.taishukan.co.jp

 

[Shun]

Wow, that must be a work of love! (or art) The printed version spans 15 large volumes.

 

[Fernando]

No doubt about it. That’s a work of love plus Japanese-level dedication.

Only thing is that if I were Japanese and willing to invest that money in a dictionary I’d spend the extra and get the 15 physical volumes. At that price they seem to be targeting mostly institutions with their digital version, rather than individual buyers.

 

[Shun]

Yeah, exactly. It's similar in price to what the Encyclopædia Britannica used to cost—which was even larger when it was still in print. They could have made the digital version cheaper, around 100 USD, but increased the price of the printed version. In the end, they would have made a similar amount of money, and it would have been closer to E.B.'s pricing model.

 

[Fernando]

They could do that, but I don’t know. If they’re really jealous of what they have and find it dishonorable to sell it for cheap I wouldn’t begrudge them. What they could also do is raise the price of the print version somewhat and toss the digital version for free. Then you’d only have it if you got the 15 volumes.

Downside is that the digital version would probably get very few updates if any.

 

[mikelove]

Honestly, if HYDCD weren't aggressively state-sponsored/subsidized it would probably be similarly expensive and hard to get, if it existed at all.

 

[Shun]

They could do that, but I don’t know. If they’re really jealous of what they have and find it dishonorable to sell it for cheap I wouldn’t begrudge them. What they could also do is raise the price of the print version somewhat and toss the digital version for free. Then you’d only have it if you got the 15 volumes.

Downside is that the digital version would probably get very few updates if any.

I very much agree, what I frivolously suggested was based on incomplete knowledge. Also, the price of the printed version was already very high, so it would have been hard to increase it further.

Honestly, if HYDCD weren't aggressively state-sponsored/subsidized it would probably be similarly expensive and hard to get, if it existed at all.

I agree, it must be hard for them to get enough revenue for their dictionaries, even if they are of a very high quality. The market probably just isn't big enough.

 

[Fernando]

Honestly, if HYDCD weren't aggressively state-sponsored/subsidized it would probably be similarly expensive and hard to get, if it existed at all.

The compiling of 漢語大詞典 started in 1979, not long after the end of the Cultural Revolution. I wonder how much a desire for atonement was a factor in it coming into being.

 

[Shun]

The compiling of 漢語大詞典 started in 1979, not long after the end of the Cultural Revolution. I wonder how much a desire for atonement was a factor in it coming into being.

I honestly don't know, but it's an interesting thought. There certainly also is the constant aim to document and perpetuate one's culture.

 

[archimon]

Would it be possible to make it so that entries that simply link to another entry show that linked entry automatically rather than forcing the user to click the link? E.g., in the follow image, could it just show the entry for the other version of 准 rather than making me leave the reader view to view that entry?

 

[mikelove]

In general we should already be doing that - we do in most dictionary entries - so I'll look into why it's not happening here.

 

[jurgen85]

On a related note, when only a part of a definition is a link, it is completely hidden from the reader popup. See e.g. 抄手 in OCC. I'm not sure what type of display would be suitable for those types of entries?

 

[mikelove]

On a related note, when only a part of a definition is a link, it is completely hidden from the reader popup. See e.g. 抄手 in OCC. I'm not sure what type of display would be suitable for those types of entries?

I don't know that there really is one - I suppose we could display those links anyway but in the context of a reader the odds that you're going to be interested in 'see also' definition are not super high anyway.

 

[jurgen85]

I don't mean the "see also" links, I mean where one would expect a definition line, but there is only a link (more "see" than "see also"). OCC has a few of these. Additional examples:

扣[B3]
劃[A4,B]
個[C]

 

[mikelove]

Sorry, it's not tagged as 'see also' but it's similar in that it's referring you to a related entry. In either case you have to follow the link to get the relevant information, which isn't great for a reader - you're better off just switching to another dictionary at that point.

We're looking at doing some more editorial work to optimize dictionaries for use in readers - a different sort of a 'display mode' which expands links like that - but we probably wouldn't be able to apply that to the Oxford dictionary due to restrictions on their end on our modifying the dictionary text.

 

[jurgen85]

Sorry for the confusion. It's very possible I just don't understand the nuances of the meanings, but I think my issue may be that 劃 is a partial variant character of 畫, but it's treated in an inconsistent way from other partial variants like e.g. 惟/唯. That indeed seems like Oxford's problem rather than yours.

The issue with 劃 as it stands is that in the reader popup it doesn't give a visible hint that it can mean "draw" or "stroke of a Chinese character", even by proxy of a link. I can't click a link if there is no trace of a link, no indication that A4 or B exist.

 

[mikelove]

The issue with 劃 as it stands is that in the reader popup it doesn't give a visible hint that it can mean "draw" or "stroke of a Chinese character", even by proxy of a link. I can't click a link if there is no trace of a link, no indication that A4 or B exist.

Yes, my point here is that this makes the OCD a less than optimal dictionary for use in reader popups because of its tendency to hide meanings in this way; other dictionaries with simpler structures like ABC and CC are better suited to quickly looking up meanings of words in documents.

But we can certainly investigate whether we might be able to make it more obvious that this particular OCD definition ought to be viewed full-screen to get all of its meanings.

 

[Toom]

Hello,
I would be nice to have a "I don't know" button in multiple choice tests. Sometimes I don't know the answer and just pick a random one but get the correct one. I then have to go back to the card and then undo the card score.

 

[mikelove]

I would be nice to have a "I don't know" button in multiple choice tests. Sometimes I don't know the answer and just pick a random one but get the correct one. I then have to go back to the card and then undo the card score.

Thanks - we've already added this for 4.0, actually.

 

[asane]

Been browsing through the forums for feature suggestions of the past to organize them in one place. Have a suggestion? Post it here, I'll update the list.

IF you have alternatives for the suggestions made, whether it be dictionaries or third-party alternatives; please post your suggestions here.

A couple suggestions for flashcards.
Problem: Some cards have so many homophones, that it really doesn't make sense to test listening, but still makes sense to test for both reading and speaking. Others have an English meaning that could be way too many different Chinese words. Some users solutions have been to create separate tests and scorefiles for each test direction. However, I was seeing words reappear too close together because the different direction SRS timing wasn't linked, so I didn't have a proper delay in exposure. I would see the same word within days in each direction, and then nothing for a long time. Not very efficient, so a single alternating test is best.

1) Limit which directions some cards are tested in alternating tests. For example, I can flag cards with way too many Chinese homophones to only alternate between reading and speaking.

2) Hide multiple choice. Currently to solve the "there are many solutions to this question" problem, I do all tests as multiple choice, but not every card is so vague to really need it. I would like to have the option to have the multiple choices hidden at the start of each new card to see if I can answer correctly without any hints. Currently I cover the screen to avoid catching the answer out of the corner of my eye.

3) Within each score file, keep separate track for each testing direction. Set the SRS so that directions with lower win rates or quality scores get shown more frequently. Also if score quality is recorded, only tie it to the direction that was tested. Currently, I'm always afraid to mark "6" when I may only be a "4" in another direction. It would make seeing the card again in the harder direction less likely to happen.

4) Have alternating test direction repeat the same test direction if a card is answered incorrectly the next time it is due.