From: Philippe VERDY (verdy_p@wanadoo.fr)
Date: Sun Mar 20 2005 - 12:52:17 CST
De : "Doug Ewell"
> Philippe VERDY wrote:
>
> >> Absolute code point order happens to work quite well for unpointed
> >> Hebrew.
> >
> > And quite well also for *basic* modern English...
>
> You want lowercase "a" to sort after uppercase "Z"?
Sometimes yes, even for English.
It may be useful to have all proper names, correctly capitalized, sorted after common words (properly not capitalized, like it happens at the beginning of sentences as a matter of grammatical style). This is what happens in many dictionnaries that have two distinct sections from common words that are actually part of the language and proper names for which there are not definitions but articles.
When doing that, the primary and secondary collation levels are merged into a single primary level, with uppercase base letters sorted separately from lowercase letters. This can be compliant with UCA collation, by tailoring (but it still allows sorting together the accents, something that binary sorting cannot perform well as binary sorting only uses a single collation level).
Also, sorting technical terms is often performed by separating leading lowercase and uppercase letters.
Sometimes Some other dictionnaries prefer mixing the two types of entries.
This archive was generated by hypermail 2.1.5 : Sun Mar 20 2005 - 12:53:05 CST