Casing (upper vs lower)

By Cathy Wissink & Michael S.Kaplan - Windows Globalization,Microsoft Corporation

Casing is that aspect in certain writing systems (e.g., Latin, Cyrillic, Greek), where there are two forms of a letter: a big (or uppercase) variant, and a small (or lowercase) variant. For example, there are two variants of Latin letter P (U+0050 or "P", and U+0070 or "p"), and there are two variants of Cyrillic letter Pe (U+041F or "П", and U+043F or "п").

Casing does impact sorting, but to a lesser extent than the linguistic characters of a language do; the weighting of case has less of an influence than linguistic characters. In other words, while the linguistic characters of a language group words into collections within a list, casing influences how a character may sort relative to a different version of that character.

To contrast the concepts of linguistic characters vs. casing in sorting, let's go back to the sorting example given a few paragraphs earlier. As noted earlier, the primary weight of the linguistic characters creates a grouping of "A" words, followed by "B" words.

Now compare the difference between "apple" and "Apple". The words are identical, except for their case. From a collation perspective, you need to give the concept of case a weight, such that one comes before the other. However, you do not want to weight the case of the letter A so much that it ends up being ordered with the B words. The weight of case (of the first letter of the string) therefore needs to be less than the weight of character.