Text case conversion and the Turkish I problem

Many programs upper-case Turkish text incorrectly, turning “istanbul” into “ISTANBUL” instead of “İSTANBUL”. The cause is that Turkish has four distinct i letters.

Four i's: i, İ, ı, I

In English the capital of i is I. In Turkish the dotted i becomes İ and the dotless ı becomes I. A program that ignores the language applies the English mapping, producing errors in both directions.

Where it causes trouble

In headings, product names and email subjects it looks unprofessional. In software it is more serious: lower-casing “FILE” with Turkish rules yields “fıle”, so file-name or command comparisons fail. Developers should state the locale explicitly for language-independent comparisons.

German ß

In German, ß is usually upper-cased as “SS”. The reverse conversion does not always give ß back, which needs care in automatic processing.

Converting correctly

Convert with the text's language in mind. For technical fields such as URLs and file names, a slug that simplifies letters is safer.

The text case converter applies Turkish, English and German rules separately; for URLs use the slug generator.