Thanks, Ken. This is very helpful information, and I think we can consider this question answered, with the possible exception of what is the best default behavior.
Our product is very configurable, and it will fall to me to help the customer get what they want. But if it were a clear-cut case of right and wrong, we would start by trying to educate them. And, there remains the issue of what should be the default behavior. (This is more the developer's concern.) It sounds (correct me if I'm wrong) like we would probably do well to leave the default alone; that is, it is at least as acceptable as the alternative (plus there is value in not changing things unless they were truly wrong before). Best regards, Gary At 02:49 PM 4/14/2004 -0700, Kenneth Whistler wrote: >Gary asked: > >> Thanks, Frank. I hadn't found the grid index and it looks >> worth remembering. But now I'm beginning to think I didn't >> ask the right question. >> >> Judging by what we saw in the back of the Unicode 2.0 book, >> we would tend to say that it is correct that (in an index) >> 21333 (0x5355) is sorting under 21313 (0x5341) instead of >> 20843 (0x516b). > >That is an incorrect assumption. Dictionaries (and indexes to >character lists in general) make different assignments to >radicals in these marginal cases. There is no one right >answer for every circumstance. Certainly, you cannot take >the Unicode *2.0* radical/stroke index as definitive for >anything. > >If you look at the Unicode *4.0* radical/stroke index, you will >find that U+5355 is listed *twice* in the index, once under >the U+5341 radical and once under the U+516B radical, >precisely to make it easier for people to find the >character in question, no matter what their assumption might >be regarding what part represents "the" radical for the >character. This is a not uncommon situation, particularly >for simplified characters which don't always have obvious >radicals. Note that the traditional form of this character, >U+55AE, is listed under the 'mouth' radical, U+53E3, and >not under U+5341 or U+516B. > >> I am looking for some table of radicals >> that I can show our customer to help support that claim. > >I think that rather than arguing with the customer on the >basis of an old radical/stroke index in Unicode 2.0, your >best course might simply be to provide the customer with >the behavior they desire. :-) > >> Perhaps I should start by asking for opinions on the above >> sorting, and for guidelines on how best to govern such >> decisions, > >Please your customer. > >--Ken > >> though I'll admit I know less than the development >> engineer involved, and so may be asking a less educated >> question than we would have. --- Gary Grosso Arbortext, Inc. Ann Arbor, MI, USA
