Regarding traditional <-> simplified conversion

[ChristopherB]

I'm aware it's difficult for software to always perfectly convert one character set to another, due to the lack of a one-to-one mapping between every tradtional and simplified character. My question is, does one way have a better chance of correct conversion over another; that is, is traditional to simplified likely to be more accurate than the other way around?

[skylee]

is traditional to simplified likely to be more accurate than the other way around?

Yes. Because 1 simplified character might represent more than 1 traditional character. Examples include 只, 发, 于, 云, etc.

[trevelyan]

If you actually have to deal with Chinese data processing, you should probably just use existing tools to make the problem go away. Even if you store all of your own data in traditional you'll have to deal with simplified input at some point.

Best solution is to use software that does word and phrase level character conversion instead of character-level conversion. Adso takes care of this and can be downloaded from http://adsotrans.com/downloads/. Google is getting better at it too.

[westmeadboy]

Given one traditional character, is there exactly one corresponding simplified character?

[imron]

Not necessarily. One example I can think off the top of my head is 乾, which is pronounced both gān and qián. For the pronunciation gān the character is simplified as 干. For the pronunciation qián the character maintains its original form.

So, in a given piece of Traditional Chinese text, when converting to Simplified the character 乾 will sometimes be 乾 and sometimes be 干.

[westmeadboy]

How about if you are given a traditional character AND its pinyin?

[imron]

Not sure, however the first question that would spring to my mind is how was that pinyin created? If it was generated, then it might also suffer from inaccuracies.

Just thought of another character too - 麽 which maps to both 么 and 麽 (but also with different pinyin).