Hey everyone, Jonathan from Decode here. Two years ago, we started as a university research project with the goal of drastically reducing the time and barriers to fluency in Chinese.
Today, we’re excited to finally go public with a platform that pushes the boundaries of digital language learning. Decode draws on the wisdom of the crowd to comprehensively map out languages in an incredibly detailed and structured way. With Taiwan and mainland Mandarin Chinese as the first supported languages, the Decode platform is built upon a powerful, data-driven linguistic framework that will supercharge a new generation of learning services.
Imagine a platform that understands the full complexity of your target language, with vocab, pronunciations, and grammar all tagged by region, formality, and frequency. Imagine a multilingual dictionary that links specific senses of words to precise translations in countless other languages. Imagine a collaborative, ever-expanding language database where contributors are rewarded and recognized for their hard work. Imagine SRS flashcards that are rapidly customizable and truly personalized, created instantly through accessing a common pool of comprehensive, reliable data. Decode is all of this, and more.
Learn more about Decode on our website or our Kickstarter page (coming soon), and let us know your thoughts and questions here!
Wow, overall it was a bit overwhelming to take in but the few things that stood out for me are: it’s data-driven, you tried to piece together Traditional Chinese as individual characters and allowed flexibility on that, and that you guys spent a lot of time trying to look at this whole thing from a foreign learner’s perspective.
I still don’t get how this can be different from Pleco (because honestly that’s how it all came off to me at first) but I am very interested and excited to see how this will be. Is it going to be like Duolingo? Are you expanding on the concepts of Google Translate? Entirely different beast here? Guess we’ll see.
These are my initial reviews felt like you guys deserved to know since you seem to be on to something really good here. Congrats and add oil! Lol.
My dream. One day, foreign people in Taiwan will not be assumed by default to be illiterate imbeciles who don’t know left from right. No more sign language or mime play, no more ‘bring a Taiwanese along with you’, just a plain conversation, as if you were, shudder, an equal.
I hope Decode Chinese helps us achieve this. Godspeed Decode Chinese.
Hello! We have updated the post to include our video that helps to explain what we are doing differently. Hope it helps clarify and please let us know if there is something we should improve or explain better. https://www.decodeapp.io/
We are different than Pleco in multiple ways, I will try to explain some of them and later on I will try to make a video to help clarify this in the future.
We have both Taiwanese Mandarin and Mainland Mandarin as different languages on our platform (dictionary, grammar, sentences, and articles)
Mandarin Chinese is a diverse language with hundreds of regional dialects. We help to understand differences with clear labeling and explanation. Imagine trying to learn English but you just throw in Irish, British, Australian, and US variations without clear labels or explanation. That is the problem with learning Chinese today but multiplied by a factor of 100.
For instance, most Chinese learners in Taiwan have heard that the words for taxi and bicycle are said differently in Taiwan vs the mainland. But when you search on taxi on Pleco, 計程車 is the fourth result, and there’s an unhelpful tag “DIALECT” next to it (see the image below for a simplified example of one out of thousands of examples on our platform)
We do not simply switch between traditional and simplified characters because that can never capture the complexity of the problem and will result in loss of data after each conversion. Instead we have mapped out thousands of cross-strait words and different variations of characters on a deep level manually
Sometimes, simplified to traditional is one-to-many or many-to-one
simplified 复/覆 vs traditional 覆 (fu4)
simplified 烟 (yan1) vs traditional 菸/煙
Sometimes, there are multiple variants of a character with slightly different meanings
We are data-driven by default which makes it easy to keep track on everything you have learned to customize and merge different types of resources without duplicates.
This means that you can import multiple different course book (word list and labels) or other resources and we can automatically help to filter out the things you already studied. But there are a tons of other things that we can apply, for instance just a simple thing as filter out search results and generating example sentences that are personalized just for you are some of the cool things we can do.
Our platform allows users to dynamically connect flash cards to resources on our platform and then update their cards when new information has been added.
For instance, new translations in different languages or improved sentences for your cards. You can either decide to automatically update with the recommended settings by the community or detailed decide what information to synchronize without losing your progress.
Thank you for your support. We have been working intensively to reduce barriers and time to reach fluency in Mandarin Chinese. Please sign up to our early access on our website to receive more information if you have not already done so
Not always. In fact quite unreliably in my experience. Sometimes there is a Taiwan Mandarin afterthought sure. But look up, for example the translation for ‘authentic’, it’s the Chinese version you will see first. There is no explicit label for the Taiwan Mandarin version on my dictionary, just something generically labelled ‘variant’’ right at the bottom of the page. However I haven’t updated the dictionary or app in a while, maybe they’ve fixed it.
Also a lot of the translations are littered with short tones, which I don’t believe are used to the same extent here.