Sorting text seems trivial until you encounter real-world data. Lexicographic (dictionary) sort compares strings byte by byte using character code values, which produces counterintuitive results: "Banana" sorts before "apple" (uppercase B has a lower ASCII value than lowercase a), and "file10.txt" sorts before "file2.txt" (the character "1" comes before "2", regardless of the "0" that follows). These behaviors surprise users who expect case-insensitive, numerically-aware sorting.

Natural sort order — also called "human sorting" — handles embedded numbers intelligently. Instead of comparing "file2" and "file10" character by character (where "1" < "2"), natural sort extracts the numeric segments and compares them as integers (2 < 10). This produces the order users expect: file1, file2, file10, file20. The algorithm splits each string into alternating text and number chunks, then compares corresponding chunks with the appropriate method.

Locale-aware sorting (Unicode collation) is essential for multilingual data. In Swedish, "ö" sorts after "z" (it is a separate letter at the end of the alphabet), while in German, "ö" sorts alongside "o" (it is treated as a variant). The Unicode Collation Algorithm (UCA) and its locale-specific tailorings handle these differences, but most simple sorting tools use byte-level comparison that ignores locale rules entirely.

Stable vs. unstable sorting matters when lines are equal under the primary comparison but have a natural original ordering you want to preserve. A stable sort guarantees that equal elements retain their relative order from the input. JavaScript's Array.sort() is guaranteed stable since ES2019, which means ties preserve input order — important when sorting by one criterion while maintaining a previous sort's order.