Outlier Linguistic Solutions

[OneEye]

Well, our idea is that our dictionary should be usable, no matter what other materials you're using to learn Chinese. So you can basically just dive in to whatever character you need to learn, and if you're unsure of some of our terminology, you can click the term in the dictionary and be taken to an explanatory page. So you learn the system by way of learning individual characters, no matter which order you learn.

 

That said, we've talked about writing a workbook (which can be used with or without the dictionary) that teaches the concepts more explicitly, and also serves as a sort of "intro to learning characters," with tips on how to make your learning more efficient. The workbook would, of course, be a more progressive approach through the content.

 

The journal article is actually about sound components, specifically the sound formulas which will be in the final version of the dictionary. No further word on that front right now, unfortunately. We'll update when we have more news. But for now, I can attach another one of Ash's published articles which is directly related to our project, and which laid much of the groundwork for our way of explaining characters. It's in Chinese, and I've also attached an abstract in English.

A Learning-centered Framework_Abstract.pdf

A Learning-centered Framework.pdf

[vellocet]

The "for experts" area isn't as "inside baseball" as I thought it would be.  It's quite readable and it's nice to see the oracle bones.  Some of it was as I feared, a reference for people who already know all the characters.  But not too much. 

 

In the demo, I only see characters.  Does this dictionary do words?

 

I sadly noticed that, once again, the dictionary does not include the part of speech.  What is it with Chinese dictionaries that they never list this?  How am I supposed to figure out how to place it in a sentence?

 

In general, the descriptions were good and helpful.  Seemed to answer all of those obvious niggling questions I have when learning Chinese.  It's infuriating to read along in Chinese learning material and they just blow by obvious questions the learner will have.  Just as an example, "when two third tones occur in succession, the first third tone becomes a second tone."  Well, what do you do when there's three in a row?  The book never says!  It just tells you a rule, considers its job done and goes on to the next one.

 

I wish the actual dictionary definitions were longer.  Example, 堂 says "hall (of a building)".  Well, what does hall mean, exactly?  A hallway?  A ballroom?  I think more exposition would be helpful here.  I've seen this character in hotels, but never studied it properly.  Something about the way that the character is often used in China would be helpful.  Otherwise it's not really much of a dictionary, more of a portable character reference, I'd think.  It's digital, brevity isn't an issue like the old paper dictionaries.  I see a section for "sentences" but that's not really what I'm talking about.  If you can be chatty and conversational in the rest of the entry, why skimp on the actual definition? 

 

A definition in Chinese would be cool, too. 

 

Some of the entries are rather long, and I'm worried that loading all that might overtax my el cheapo Hongmi phone.  How is application performance on lower-end smartphones?  A lot of laowai I know have Meizu, Xiaomi, etc. 

 

Otherwise it looks good and I'll probably buy it when it comes out if it costs $30 or so.  I wouldn't mind having the "expert's edition" even if I'm not an expert, especially if it can be collapsed by default in the options.  Is that the only difference, one section missing? 

[Demonic_Duck]

For some unfathomable reason, bit.ly is blocked in mainland China. Would suggest switching the bit.ly links on here or on your company's site with direct links to the content itself.

[OneEye]

Hi vellocet,

 

I just spoke with Ash, and we will have a list of common words using each character the final release. We won't have full entries for words (though you'll be able to tap on the words to look them up in whatever other dictionaries you have installed in Pleco), but each character entry will contain a list, and probably parts of speech for the words as well. We just don't have that data in the demo version. Also, Pleco's "Words" tab is very useful for this sort of thing.

 

The "Sentences" tab is a Pleco feature. The demo is set up to look and feel like it will when it's released through Pleco, as much as possible, so not everything you see on screen is Outlier stuff.

 

Application performance on lower-end phones is something you'll have to ask Pleco about. I'll see if I can get Mike to pop in to this thread and answer that for you.

 

Thanks for the feedback! I'll pass your comments on to the rest of the team. We definitely want to hear this kind of stuff, especially now when it's still early, so we can implement some of the feedback in the actual release. We'll also make some incremental updates later based on user feedback, so keep it coming!

[OneEye]

Thanks for the heads up, Demonic_Duck! I'll get those fixed ASAP.

[Demonic_Duck]

A definition in Chinese would be cool, too.

 

To clarify (@OneEye), will the first full release include all of Chinese Trad, Chinese Simp and English? Also, is/will there be a demo version of the Chinese version(s) to play with before that?

[OneEye]

@vellocet,

 

I just spoke with Ash, and we will have a list of common words using each character the final release. We won't have full entries for words, but each character entry will contain a list, and probably parts of speech for the words as well. We just don't have that data in the demo version.

 

@Demonic_Duck,

 

The full release will be bilingual: Traditional and Simplified Chinese, as well as either English or German. Hopefully we'll add other languages later as well. We don't have a Chinese demo ready yet, but we're working on one. I'll post it as soon as it's ready. We do have a German one (only a few entries, unfortunately) which should be ready very soon.

[Demonic_Duck]

The full release will be bilingual: Traditional and Simplified Chinese, as well as either English or German.

 

So all three languages (four if you include Trad/Simp as different languages) will be released at once, but you have to choose between English or German for any given licence?

 

Sounds good to me!

[OneEye]

Yeah, that's it. We wanted to make something learners can grow into when their Chinese gets to that point, not to mention something that will be useful to native-speaking teachers who might be more comfortable reading in Chinese but want the information available in their students' base language as well.

[Demonic_Duck]

I notice this isn't in the demo version, but is there any chance you'll include information on common colloquial names for certain components? For example, barely anyone except for serious academics will know that the phonetic value of “攵” is [pū], but as soon as you tell them “反文” they'll know exactly what you're talking about.

 

Of course, there are other resources available with this information, but I'd imagine many people would find it useful to have handy whenever they look up a component.

[Shelley]

Maybe a dumb question but what is memory chunking?

 

I think it looks great and has all the info you would need. Really like the demo.

[tysond]

So, I said I would support the kickstarter and I did - $199 buckeroos.  Even before I opened the demo.

 

I will think long and hard about which character I will sponsor :-) :-) :-)  Hope all the rude ones are not gone by the time I get there.

 

Some feedback:

 

- Generally, based on the demo, I can get quite a bit of value out of this already.  I am at an upper-intermediate sort of kind of getting to advanced level, but sometimes it's super useful to get a better grounding in even simpler characters.  Having an annotated insightful history, *clear* explanation about phonetic components (even if they are all complex and stuff) and comparisons between simplified and traditional is super useful.  Some of the phonetic descriptions are very hard for a layman to understand.  I've read many books about languages and language sound drift / evolution so I totally get it.  But if the essentials version says "this is a phonetic component" but it's actually way different in modern pronunciation, who does it help except the experts?  

 

- In the demo I can't click on components to see them - e.g. 草 - doesn't let me click on 艸 .  I assume this will work eventually. 

 

- It ... umm... reeks ... of academia.  I think you need to be didactic (see I can be academic) with some kind of introduction / walkthrough / daily method of how to use this kind of information to learn Chinese characters better.  Otherwise it's just a reference for people who have already gone through a lot of the slog and want to bounce around getting some deeper understanding or drilling in on something that's been hard to memorize.  Don't get me wrong, it's a very nice reference.   I have been through the slog and would love a reference like this.  But I am 大手大脚 so I will buy lots of stuff before worrying about the cost.  For a cash strapped student it's a hard sell if they just want to plough through and learn.  I know it's just a demo and you have other stuff in the works (so maybe my comments are irrelevant), but I think your business depends on getting the value proposition right for the mass of students.  Maybe a character learning "mill" or "tracker" that suggests a sequence of learning?  Something that takes advantage of the fact that this is software not just a book, helps you sort by frequency/common lists/etc?  Guides you through a process of learning?  Suggests characters to learn next?  To be honest one of Heisigs greatest values was that he sorted the characters into an order and grouped them.  Just follow the order and you'll get there.  What is the goal of using this app?  Or is it a Pleco dictionary?  (I have bought most of them too :-) )

 

- App looks very nice and usable generally.  I really like the meaning derivation arrows and circles.  Lots of references leave you with 27 meanings and it's up to you to figure out which are root, which are derived and what's just kind of left-field.  This is very valuable.  There are some colors here I don't understand.  For 谢 the most common meaning is To Thank - that's in purple - is that color significant? 

 

- I agree that having the common chinese names of each component is highly useful. 

 

- Yes, words sorted by frequency would be cool.

 

- Hope the brevity can be expanded a bit over time.  The Experts part is very nice and has lots of cool stories and insight.  The Essentials feels light.  Back to.... what's the value to essentials learners?  Feels a bit like the essentials is just an upsell to experts for academics.

 

- Sorry if I have been blunt.  You have my support already.  Now that you are attracting customers via the Kickstarter I hope the mass of the feedback will help you fine tune.

 

- Good luck.  I support folk who build Chinese learning resources because I hope that you can make a living and we can benefit.  

[realmayo]

Gosh it could be at least a year away, hadn't realised that. Still I suppose that's the point of kickstarter. 

 

(I'll probably have learned all of Chinese by then. But just in case....)

[vellocet]

Wait, what?  I thought this was standalone software.  It's actually some kind of Pleco plugin?  Wow, I didn't expect that.  So the market is just going to be a subset of people who have Pleco installed?  I know a lot of people with the other dictionaries installed.  Well, actually I suppose most of them aren't serious about learning Chinese.  Good thing I already have Pleco registered. That was going to be one of my issues, how is it going to be going back and forth between this and Pleco.  If it's already in the software then that's great.

[mikelove]

Covering some Pleco questions:

 

The plan is for this to be integrated into our regular app as an add-on. The exact UI is still TBD, but in general we're moving towards making the current tabs-and-sections interface more customizable (can add / remove / rename tabs, mix and match data from whatever databases you want in whatever tabs you want, etc), so that should allow for seamless jumping between Outlier's dictionary and any other data in Pleco you're fond of using.

 

So this means that if you own the ABC dictionary in Pleco you'll be able to get ABC dictionary classifications in the dictionary just by tapping on the "DICT" tab, not to mention parts of speech and such from other dictionaries (our free PLC dictionary will be adding parts of speech in a few months, so you'll be able to get those without buying any other Pleco add-ons). Along with Cantonese, example sentences, flashcards, compound words, etc. Basically, if there's another dictionary in Pleco that offers a particular piece of information then you'll be one tap away from that information in Outlier's dictionary.

 

We're also adding a dedicated UI for single dictionaries (think of an amped-up version of the current Browse Entries screen with a search box), so if you don't care for our software in general and just want to be able to get at this particular dictionary, there'll be an easy way to do that. It should also be possible to get at entries from this dictionary directly via link from another app, if, say, you want to be able to hop into them from an Anki test; we just added direct URL links to the definition screen for our 3.2.5 update on iOS and will shortly have them on Android too.

 

As far as performance, we spent a bunch of time last year optimizing our Android app to display longer entries quickly - pre-2014 it was noticeably slower at loading them than our iOS app, but it no longer is - so the dictionary should work fine on even a low-end phone.

 

As far as character learning systems, we haven't had any specific discussions with Outlier on that but we're pretty excited about this dictionary in general and if we can come up with a way to work its philosophy into a new flashcard test type or set of premade lists or some such we'd very much like to do so.

[Shelley]

Maybe a dumb question but what is memory chunking?

 

Its ok, I looked it up, I now understand.

[OneEye]

re: colloquial names for components, I think that's a really cool idea. I'll talk to the others about it and see if it's something we can implement. However, the correct pronunciation of such components are still important to know as a learner, especially if they serve as sound components in other characters. I know you guys know that, I'm just reiterating.

 

@tysond,

 

Thanks so much for your support, and all the feedback.

 

 

- Generally, based on the demo, I can get quite a bit of value out of this already.  I am at an upper-intermediate sort of kind of getting to advanced level, but sometimes it's super useful to get a better grounding in even simpler characters.  Having an annotated insightful history, *clear* explanation about phonetic components (even if they are all complex and stuff) and comparisons between simplified and traditional is super useful.  Some of the phonetic descriptions are very hard for a layman to understand.  I've read many books about languages and language sound drift / evolution so I totally get it.  But if the essentials version says "this is a phonetic component" but it's actually way different in modern pronunciation, who does it help except the experts? 

 

One thing we regret not having ready for the demo is the "sound formulas." They're a simple set of rules that will help make sense of sound variation in the phonetic series. They tell you which initials and finals are related to each other. For a simple example, b, p, m are related because they're pronounced in the same part of the mouth (the lips). Some related sounds don't seem to be related because of how the language has changed over time. An example would be 立's sound series (check the System tab under 立 in the demo). You have initials l- y- q- and w-. That's a pretty messy sound series, but the formula will make it clear how they're related.

 

Something I should emphasize here: the formulas are NOT intended to all be memorized! Of course, you can if you want, but you'll get a feel for possible sound relationships just by learning the characters and glancing at the formulas when they don't make sense. The point is that you can then look at a character like 當 and know that 尚 is the sound component because using our system has helped you develop a very acute feel for how sound components work. We've surveyed native speakers to gauge their feel for sound components, and our system should help learners identify sound relationships better than the average native speaker.

 

 

 

- In the demo I can't click on components to see them - e.g. 草 - doesn't let me click on 艸 .  I assume this will work eventually.

 

Yes, all functionality that Pleco normally has will be available. Our original demo had everything clickable as much as possible, but it ended up being several hundred megabytes, and very slow (Invision is essentially a more flexible version of PowerPoint, so it's all just image files with clickable links to other image files). We removed some of the nonessential functionality in order to get it down to about 12 MB.

 

 

 

- It ... umm... reeks ... of academia.  I think you need to be didactic (see I can be academic) with some kind of introduction / walkthrough / daily method of how to use this kind of information to learn Chinese characters better.  Otherwise it's just a reference for people who have already gone through a lot of the slog and want to bounce around getting some deeper understanding or drilling in on something that's been hard to memorize.  Don't get me wrong, it's a very nice reference.   I have been through the slog and would love a reference like this.  But I am 大手大脚 so I will buy lots of stuff before worrying about the cost.  For a cash strapped student it's a hard sell if they just want to plough through and learn.  I know it's just a demo and you have other stuff in the works (so maybe my comments are irrelevant), but I think your business depends on getting the value proposition right for the mass of students.  Maybe a character learning "mill" or "tracker" that suggests a sequence of learning?  Something that takes advantage of the fact that this is software not just a book, helps you sort by frequency/common lists/etc?  Guides you through a process of learning?  Suggests characters to learn next?  To be honest one of Heisigs greatest values was that he sorted the characters into an order and grouped them.  Just follow the order and you'll get there.  What is the goal of using this app?  Or is it a Pleco dictionary?  (I have bought most of them too :-) )

 

Well, it is a Pleco dictionary first and foremost, because we want people to be able to use it without being married to any particular order. While Heisig's order is quite brilliant (I used his books when I was a beginner), the limiting factor is that you pretty much have to start at the beginning.

 

That being said, we do have some ideas for how to guide people through it, not least of which is the workbook I mentioned in #81 above. We've thrown other ideas around, like a video course, but it will depend on whether we can get the funding to make it happen. Hopefully I'll be able to announce something soon, but rest assured that we don't want to just throw this thing out there and leave people without any guidance as to how to make use of it most efficiently.

 

 

 

- App looks very nice and usable generally.  I really like the meaning derivation arrows and circles.  Lots of references leave you with 27 meanings and it's up to you to figure out which are root, which are derived and what's just kind of left-field.  This is very valuable.  There are some colors here I don't understand.  For 谢 the most common meaning is To Thank - that's in purple - is that color significant?

 

Yes, we've highlighted what we call the "Basic Meaning” in each entry. That's the meaning that most native speakers would think if they saw the character out of context — the basic meaning in modern Mandarin. Sometimes that's the same as the original meaning (as is the case with 立), sometimes it's not (各).

 

One of the problems with many dictionaries is, as you mentioned, that they throw a ton of different definitions at you without any guidance as to which ones are most important to know. Sometimes editors will put extremely obscure ones in there just to differentiate their dictionary from others on the market. We're taking the opposite approach — reduction. The point of the Essentials Edition — and this is perhaps why its value may not be immediately obvious — is to give you the info you need to master a character, and no more. We're trying to avoid overwhelming the learner by including a bunch of stuff that they don't need to know. Generally, if meaning Y is easy to get to intuitively from meaning X already listed, there's no reason to add meaning Y to the list. The exception to that is if there's an obscure or archaic meaning which is an important link between two other meanings, in which case it will be there, but grayed out so that you know it's not necessary to memorize it.

 

 

 

- Yes, words sorted by frequency would be cool.

 

Agreed! It's important to learn characters in the context of learning words rather than in a vacuum. An easy and effective way to do that is to learn a few words containing that character at the same time you learn the character, so you get a feel for how it's used in actual language. So that's definitely something that will be there, just not in the demo.

 

 

 

- Hope the brevity can be expanded a bit over time.  The Experts part is very nice and has lots of cool stories and insight.  The Essentials feels light.  Back to.... what's the value to essentials learners?  Feels a bit like the essentials is just an upsell to experts for academics.

 

The value of the Essentials version is subtle: a correct analysis of each character form. Naturally, people will wonder why they should go with this over whatever else is out there, and that's the reason. The more clearly a learner understands how characters work, and the more accurate the explanations in their learning materials, the more efficient their learning will be. It may not be flashy, but it is super important.

 

 

- Sorry if I have been blunt.  You have my support already.  Now that you are attracting customers via the Kickstarter I hope the mass of the feedback will help you fine tune.

 

Not at all! We're glad to have honest, thoughtful feedback like this. It will help us to refine our thought process and hopefully make the product more useful for more people. So thanks!

 

[OneEye]

@vellocet,

 

I've known people who have used other dictionary software for some reason who are, in fact, serious about learning Chinese. Why they don't use Pleco is beyond me.   But our thinking was: Pleco is a free platform and it's the best thing out there, so naturally that's who we wanted to work with if at all possible. The fact that Mike saw value in our project was confirmation that we were on the right path.

 

@realmayo,

 

We're hoping it won't take that long. We're sticking to the principle of "under-promise and over-deliver." Our plan is to have the data finished and sent to Pleco by the end of this year. We have a workflow already set up that should theoretically allow for that to happen. But the best-laid plans....you know.

[Demonic_Duck]

I know plenty of Japanese people who are serious about learning Chinese and who don't use Pleco, they use those little portable dictionaries. I think that's mainly because Pleco is heavily geared towards English speakers (not necessarily native English speakers, but many Japanese folks no very little English at all).

[OneEye]

Ah, true enough. I'm not sure how much of a market there might be for our dictionary amongst Japanese learners of Chinese though — they tend not to have any trouble remembering characters! Of course, if the demand is there it would be a cool product to do, but I'd be surprised.