On Wed, 15 Sep 2004 22:58:48 -0700, Mark Williamson node.ue@gmail.com wrote:
Since the wrong character and the right character in Traditional are the same character in Simplified, it won't have any effect on Simplified users who will continue browsing it as it is. However, the error *will* show up to Traditional users, who will then correct it. This correction will have no effect on the apperance of the text to a Simplified user, but it will make it so it uses the correct character for a Traditional user.
This eliminates the need for special semantic markup.
--Jin Junshu/Mark
As someone pointed out before, the problem is not just on the character-to-character mapping. Some concepts are expressed entirely differently, for example, 电脑(electronic brain) vs. 计算机(calculator or computer). A second example will be translations of foreign names, for example, Croatia is translated in Mainland China as 克罗地亚，but 克罗埃西亚 in Taiwan. This kind of difference can be arbitrary, and will likely evolve along time. It is mainly this kind of difference that requires a special markup.
In fact, I think the character ambiguity is less of an issue, because most of them (although not all) can be distinguished by looking at the phrase that a character is in. Including phrases that contains ambiguous characters in the conversion table should eliminate most of the character ambiguity.
However I don't have enough data to proof this. I have implemented this idea on the test site, but since I myself am a Simplified user, I can't really tell if most of the ambiguity on the character level is eliminated or not. I urge Traditional Chinese user on this list to visit the test site, and proof read a few converted articles to see how well/bad the current implementation works. The test site is at http://s87257573.onlinehome.us/wiki/