New Chinese Dictionary Project

[ChouDoufu]

Inspired by Albert's (of Laowai Chinese) post about the future for Chinese dictionaries, I started a project to create a useful and humane Chinese dictionary. It's something I've been thinking about for a while (4 years now). Hopefully it will make things easier for people studying Chinese in the future.

If you want to know more about it, or are interested in helping out, read this post and sign up for the project's mailing list.

Oh, and spread the word.

[roddy]

How about you tell us more about it here?

[laurenth]

In addition, the blog post referred to is not accessible - for me at least (504 gateway time out).

[jbradfor]

In what way are other dictionaries inhumane?

[Gleaves]

Good luck with the effort. It's quite an undertaking. Is this a joint effort with Laowai Chinese?

One quick suggestion - I don't think 3000 Hanzi Dictionary quite fits as a name. I realize that is the name of your blog, but it implies the dictionary is limited to 3000 hanzi.

[ChouDoufu]

@Gleaves The actual dictionaries will be called

• "Chinese-English Dictionary by 3000 Hanzi" and
  
• "English-Chinese Dictionary by 3000 Hanzi"
  

That will hopefully stem some of the confusion

@laurenth: thanks for letting me know. Had some memory issues. I've upgraded the server.

@jbradfor: Most dictionaries aren't made for people. The editors compiling them don't spend a lot of time thinking about how people will actually use the dictionary. The approach I'm going for takes into account how people plan to and want to use it. It will have a good UX (user experience). That's a humane dictionary.

@roddy: It's a bit difficult to discuss. I have so much to say about the subject that I don't know where to start. But I'm open to questions and thoughts about dictionaries and about the project.

I'd also love to ask forum members questions too:

Like which should be done first?

A Chinese-English dictionary or an English-Chinese dictionary?

[roddy]

I think you're going to need more detail before anyone gets enthusiastic. Sample entry to look at?

[jbradfor]

Most dictionaries aren't made for people. The editors compiling them don't spend a lot of time thinking about how people will actually use the dictionary.

I assume you're familiar with MDBG? What do you think is wrong with the UI and what do you want to do better?

[cababunga]

I think you're going to need more detail before anyone gets enthusiastic.

My guess is that he meant us to read Albert Wolfe's article linked from the original post, which explores problems in details.

[jbradfor]

I read the article. I think it's stupid (with one exception).

On the C->E side, he's basically whining about the problem every dictionary has. We want more entries, more information, more this, more that. But we want it to be cheap. Or free. As he says "I’m a horrible business man", and that's obviously true.

One the E->C side, what he's asking for is not a dictionary, it's a combination dictionary, usage guide, usage notes, and grammar guide. There is often no 1-to-1 correspondence between words/concepts in different languages, and something as common as "go" (or "run") would not have a 1-to-1 correspondence between languages as different as English and Chinese. How can a dictionary fix this?

The one good point to the article is he points out some information that MDBG is missing that could be added without too much work. A field for "usage notes" would be a good example, better information about frequency is another. And I agree it falls short with information on regional variations (except mainland vs Taiwan).

Overall, I fail to see how his approach is better than what MDBG has now. I certainly agree that some of his ideas could be added to MDBG, but I fail to see why he wants to replace it.

It's a bit difficult to discuss. I have so much to say about the subject that I don't know where to start. But I'm open to questions and thoughts about dictionaries and about the project.

Then I don't think you will get many (productive) comments. If you don't know where to start, how can we know what questions to ask and what thoughts to provide?

[Olle Linge]

On the C->E side, he's basically whining about the problem every dictionary has. We want more entries, more information, more this, more that. But we want it to be cheap. Or free. As he says "I’m a horrible business man", and that's obviously true.

If you read the article again, you'll see that he compares with other languages. If we look at English-Swedish (simply because I'm Swedish), there are excellent dictionaries online, for free. This is not a problem "every dictionary has". If it is, there is a huge difference in how serious it is. If I want to look up a word in English and get the Swedish translation, I'm more or less always able to do that with ease online. That is not true with Chinese-English dictionaries. Also, writing a fairly detailed article to specify problems and suggest solutions isn't exactly "whining".

One the E->C side, what he's asking for is not a dictionary, it's a combination dictionary, usage guide, usage notes, and grammar guide. There is often no 1-to-1 correspondence between words/concepts in different languages, and something as common as "go" (or "run") would not have a 1-to-1 correspondence between languages as different as English and Chinese. How can a dictionary fix this?

Again, compare with dictionaries in other languages. They usually contain quite a lot of information on how the words are used, but that's quite rare among online (and paper) Chinese dictionaries. I don't think anyone expects a perfect description of how the word is used in Chinese, but more than nothing at all would certainly be a good start. A dictionary can't "fix this", but it can definitely help. For instance, most dictionaries I used in English contains information such as "written", "formal", "vulgar" and "rare". Verbs can be marked as "transitivie" or "intransitive", nouns as "countable" or "uncountable" etc.

Overall, I fail to see how his approach is better than what MDBG has now. I certainly agree that some of his ideas could be added to MDBG, but I fail to see why he wants to replace it.

I agree. If we ever hope to accomplish something, spreading ourselves thin over too many projects is a bad idea. Also, I see no real reason why a new dictionary should be started when the foundations already exist.

[cababunga]

Also, I see no real reason why a new dictionary should be started when the foundations already exist.

It may look like it on the first glance, but it isn't. Questions like this popped up in the past and nothing changed for years. It all boils down to the following quote from the post by Dennis:

http://groups.google...23a00b653f0f81f

> I end my post with this question: What is the best way to make major

> changes to a project like CEDICT?

The answer is simple: start a new project. I reckon that saying it is

easier than doing it. I do believe that multiple projects can live

together though, where the new project can take advantage of the old

project until it eventually replaces the old project (if it proves to be

successful that is).

This is (or was four years ago) an opinion of a person who is behind MDBG. If someone wants to implement all infrastructure to enable collaborative work on such a dictionary, I'd say we definitely shouldn't discourage such an undertaking.

[jbradfor]

What is the best way to make major

changes to a project like CEDICT?

The answer is simple: start a new project.

I read that. But the author didn't give any reason why, so I'm totally unconvinced.

If someone wants to implement all infrastructure to enable collaborative work on such a dictionary, I'd say we definitely shouldn't discourage such an undertaking.

Wellllll....... There is limited number of people that are willing to work on such projects without pay, and their time is limited. While it's not a zero-sum game in terms of time working on CEDICT / MDBG and this new project (someone might get fired up to work on something new), I don't think it's independent either. So to the extent that time is spent on a new project duplicates what CEDICT already has, this could be time taken away from other projects.

In thinking about this more, what I don't understand is that with one or two exceptions, everything that the author wants is independent of what CEDICT currently contains. Why not stay with CEDICT as the base C->E dictionary, and add, as separate projects, the other information needed? For example, a collaborative project to look at regional variation in word usage would be invaluable, and could be done totally in parallel to, and entirely complement, CEDICT.