Outlier Dictionary of Chinese Characters

J

jenerback

Member
Not sure if this request is best directed at Ash or Mike, but would it be possible to make the ancient forms go full-screen when tapped? In line with the behavior of alternative fonts under the character tab.

Also, I remember reading about a system tab. Is that still in the works?

In any case, thanks for a super interesting dictionary!
 
Ash

Ash

举人
Glad you like the dictionary!
The system tab is definitely still in the works. In fact, we've made some progress on it. Actually, we already have the data mostly done thanks to a friend of ours that wrote code to analyze our raw data. As I mentioned above, I'm almost finished with the Expert Entry Expeditor (EEE -- the software that will help us creating expert entries quickly). Should be completed really soon, since most of it is already working. Once that's done, there's another software project that is mostly complete, but that I need to finish up. Next in line is getting the system data ready to send to Pleco, which entails some editing and getting the data into a format that Pleco can use.
As far as your request, that's something Mike will have to answer, so I'll pass the mic to him (ba dum tsss!).
 
mikelove

mikelove

皇帝
Staff member
Full screen ancient forms would be a bit problematic at the moment since we don’t currently support vector embedded images; Outlier’s embedded characters wouldn’t scale well.

This is mostly Apple’s fault; they still don’t support decoding SVG images directly and we’ve been reluctant to spend time on adding independent SVG support since we kept expecting they’d fix that. But since it remains missing in iOS 13 and we’re unlikely to require anything newer than that anytime soon it may be time to bite the bullet and add it.
 
alex_hk90

alex_hk90

状元
Ash said:
Expert entries
trad: 攵變攴高藍立春土剪持屮路尚黑能美也各堂是會監需足射艸舞前草艹造軟無得
simp: 攵攴无高立春监土剪持屮路尚变软黑能美会也蓝䒑各堂是需足射艸舞前草艹造得

Also, I'm getting REALLY close to finishing the software that will allow us to create the Expert entries fairly quickly (by automating basically everything except the actual content creation). Once it's done, we'll put out another 100 or 200 Expert entries within a month. I'll post here once the software is complete.
Click to expand...
Just had a look at a few of these expert entries and I'm really impressed by the detail. :D

Looking forward to the next batch of Expert entries! :)
 
  • Like
Reactions: Ash
羊肉泡馍

羊肉泡馍

Member
Ash said:
Expert entries
trad: 攵變攴高藍立春土剪持屮路尚黑能美也各堂是會監需足射艸舞前草艹造軟無得
simp: 攵攴无高立春监土剪持屮路尚变软黑能美会也蓝䒑各堂是需足射艸舞前草艹造得

Also, I'm getting REALLY close to finishing the software that will allow us to create the Expert entries fairly quickly (by automating basically everything except the actual content creation). Once it's done, we'll put out another 100 or 200 Expert entries within a month. I'll post here once the software is complete.
Click to expand...
I just noticed that there was an update recently with 100 new expert entries. Any chance you could post the list of characters here?
 
ACardiganAndAFrown

ACardiganAndAFrown

状元
羊肉泡馍 said:
I just noticed that there was an update recently with 100 new expert entries. Any chance you could post the list of characters here?
Click to expand...
Mini Update: 100 new Expert entries have been released
Simplified: 的不在我他为地到以之才弋王士止童辛又日寸寺时京尤丵对⻏⻖贝过中就舌(舌)舌()音言骨弓殳发另冎乍用首道作刀另肉月(肉)月(月)夕别史吏使事同凡来去多外卜回没水有民牛臣目罒(目)看见视玉现望示社工巨矩兴耳亡人麦夂斤向宀了卩亻儿口


Traditional: 的不在我他為地到以之才弋王士止童辛又日寸寺時京尤丵對⻏⻖貝咼過中就舌(舌)舌()音言骨弓殳發另冎乍用首道作刀肉月夕別史吏使事同凡來去多外卜回沒水有民牛臣目罒(目)看見視玉現望示社工巨矩興耳亡人麥夂夊斤向宀亻儿了卩口
Click to expand...
 
Ash

Ash

举人
@alex_hk90 and @羊肉泡馍 , glad you like it! The main part of the wait was me coding up the software tools that allow us to make the entries in a reasonable time with minimal errors. The code is 99.7% done. Still needs a few changes here and there. We're trying to get to a place where we can put a predictable number of Essentials and Expert entries on a 4 or 6 week cycle. It'll be another few months before we get there as there a couple of internal things that we have to get taken care of first (one of those is getting the system level data ready). The main part of the waiting over the years has been due to the creation of software tools: 1) for testing data before releasing it to Pleco, 2) for converting paleographic analyses to entries, 3) for creating meaning trees and 4) for creating Expert entries. Those are all pretty much done now.
 
jurgen85

jurgen85

探花
Hi, I got the essentials version and I'm absolutely loving it! Here are a few of my humble suggestions:
  1. Explain the phonetic loan icon　○　in "How to use this info" and how it affects subsequent lines.
  2. "How to use this info" link could be merged into the respective section title (COMPONENTS, MEANINGS) to be less cluttered and repetitive.
  3. Take into account dark mode for popups.
  4. Consistent font in popups (it's currently larger than others, such as grammar label popups in ABC).
  5. The etymology arrows　→ ⇒ ⇛ ⭆　use inconsistent fonts.
  6. Integrate your data into searches, so that searching something like　又:e　would match　难　and other characters where　又　is an empty (or form/meaning/sound/any) component.
 
Ash

Ash

举人
jurgen85 said:
Hi, I got the essentials version and I'm absolutely loving it! Here are a few of my humble suggestions:
  1. Explain the phonetic loan icon　○　in "How to use this info" and how it affects subsequent lines.
  2. "How to use this info" link could be merged into the respective section title (COMPONENTS, MEANINGS) to be less cluttered and repetitive.
  3. Take into account dark mode for popups.
  4. Consistent font in popups (it's currently larger than others, such as grammar label popups in ABC).
  5. The etymology arrows　→ ⇒ ⇛ ⭆　use inconsistent fonts.
  6. Integrate your data into searches, so that searching something like　又:e　would match　难　and other characters where　又　is an empty (or form/meaning/sound/any) component.
Click to expand...
Glad you like it!
To comment on your suggestions:
1. Yes, this would be a good idea. We've been meaning to do that (and for the arrows as well), but it has slipped through the cracks up to now.
2. I can talk to Pleco about that, but I'd have to see actual images of how it would look first.
3. Can you be more specific? Are images not coming out correctly in dark mode or what?
4. This is definitely a Pleco thing. As far as I know, our data doesn't request fonts.
5. Same as #4.
6. We are going to be publishing system-level data in the near future, though I don't know when exactly. We have some internal stuff to take care of first, but once that's done, we'll be working on the system-level data. We already have the data, we just need to edit it and talk to Pleco about how it will be displayed. So, for the 又 entry, you'd click on the "system" tab and it would show lists like: 又 as a semantic component, 又 as a sound component, 又 as an empty component, 又 as any component, etc. You can check out our original demo, which gives system-level data for three entries (they are marked).
As far as searching (i.e., as a separate issue from system data), that would be a Pleco thing.
 
jurgen85

jurgen85

探花
I don't know if it's an Outlier thing or a Pleco thing:

Regardless of the main character set, the explanations for many secondary set characters do not show up unless I go to the Chars tab and select the one from the other set, or search for that character which lists an additional lone Outlier entry. Same goes for characters with multiple readings, only some readings have the Outlier section. This leads to some confusion (no other dictionary lists 师 as a traditional variant or 師 as simplified, but with Outlier enabled it looks like it). Examples:

师、師
单、單
会、會
医、醫
将、將
带、帶
离、離
义、義
国、國
连、連
产、產
着、著 + various pron.
穴 xue2, xue4
行 xing2, hang2, hang4, heng4
重 chong2, zhong4
更 geng1, geng4
漂 piao1, piao3, piao4
仔 zi3, zai3
突 tu1, tu2
劲 jing4, jin4
调 tiao2, diao4
稍 shao1, shao4

——————————

Edit: I just noticed that a flashcard with the hanzi 这[這] will not be tested at all with "Remap cards to dicts" set to only the Outlier dicts.

See also related post.
 
Last edited:
mikelove

mikelove

皇帝
Staff member
Thanks, we'll add these to our correction list. There aren't official SC/TC maps from Outlier yet, we generated our own quick-and-dirty ones for search indexing purposes but didn't want to put too much more into it than that since we expect they probably will add them eventually.

But just in general I'd suggest that you stick to the simplified Outlier for simplified character searches and the traditional Outlier for traditional, the information in one won't necessarily even be accurate/relevant for the other (e.g. in cases of one-to-many mappings).

Actually come to think of it we may not even be including simplified/traditional mappings for Outlier in Pleco 4.0 at all if we can get a particular bit of our new entry merging system working in time.
 
Ash

Ash

举人
We just released 700+ new Essentials entries, bringing the total number of completed entries to over 2700!! All of the stub entries have been filled in too. So, be sure to check that out.

If you already have the dictionary, you should be prompted to download the update automatically when you open Pleco. If not, just go to Menu > Add-ons > Updated and you should be able to download it there.

We'll be adding another few dozen Expert entries within a month or so.
 
ACardiganAndAFrown

ACardiganAndAFrown

状元
Ash said:
We just released 700+ new Essentials entries, bringing the total number of completed entries to over 2700!! All of the stub entries have been filled in too. So, be sure to check that out.

If you already have the dictionary, you should be prompted to download the update automatically when you open Pleco. If not, just go to Menu > Add-ons > Updated and you should be able to download it there.

We'll be adding another few dozen Expert entries within a month or so.
Click to expand...
What percentage of the project has been completed now? Expert was slated to be: 2000 characters? And essentials was going to be: 3500 characters? Is that right?
 
Ash

Ash

举人
Expert and Essentials will both be 4000 (though for some of the more modern characters, there isn't much interesting to say Expert wise).
Essentials is 67.5% finished. However, this isn't a good metric for learners trying to figure out if the dictionary is ready enough. The vast majority of
people learning Chinese will never get over 2500 characters, so from that perspective, Essentials is basically complete for the vast majority of learners. 67.5% IS a good metric of how much work we have done and left to do! Since there's only ~135 Expert entries, it's not worth crunching the numbers, but those 135 entries do make for some interesting reading!
 
ACardiganAndAFrown

ACardiganAndAFrown

状元
Ash said:
Expert and Essentials will both be 4000 (though for some of the more modern characters, there isn't much interesting to say Expert wise).
Essentials is 67.5% finished. However, this isn't a good metric for learners trying to figure out if the dictionary is ready enough. The vast majority of
people learning Chinese will never get over 2500 characters, so from that perspective, Essentials is basically complete for the vast majority of learners. 67.5% IS a good metric of how much work we have done and left to do! Since there's only ~135 Expert entries, it's not worth crunching the numbers, but those 135 entries do make for some interesting reading!
Click to expand...
But I'm assuming the process to build the expert entries is more or less streamlined now, as keeps getting mentioned?
 
You must log in or register to reply here.
Top