Sorting is one of the most studied problems in computer science, with a rich history spanning from the mechanical tabulating machines of the 1890 U.S. Census to modern parallel sorting algorithms. The fundamental task—arranging elements in a specified order—underpins countless applications from database indexing and binary search to data presentation and human comprehension. Donald Knuth devoted an entire volume of "The Art of Computer Programming" to sorting and searching, reflecting the algorithmic depth beneath this seemingly simple operation.

The distinction between lexicographic (text) sorting and numeric sorting is a common source of confusion and errors in data handling. Lexicographic ordering compares strings character by character from left to right using character code values; in this scheme, "9" sorts after "10" because the character "9" has a higher code point than "1". Numeric sorting converts values to numbers before comparison, correctly placing 9 before 10. Mixed-type columns—containing both text and numeric values—require heuristic decisions about which sorting mode to apply, and incorrect detection produces confusing results where some values appear out of order.

Multi-level sorting, also called hierarchical or compound sorting, applies multiple sort keys in priority order. The primary sort divides data into groups (e.g., by department), and the secondary sort orders records within each group (e.g., by salary within each department). Additional sort levels further refine the ordering. SQL's ORDER BY clause supports this through comma-separated column specifications, and spreadsheet applications implement it through multi-level sort dialogs. Multi-level sorting is essential for creating structured reports where data is organized hierarchically.

Sort stability—whether records with equal sort keys maintain their original relative order—is an important but often overlooked property. A stable sort preserves the input order among equal elements, which matters when applying sequential sorts: sorting first by name and then stably by department produces records grouped by department with names alphabetically ordered within each group. An unstable sort may scramble the name ordering within departments. Most modern sorting algorithms used in practice, including TimSort (used by Python and Java) and merge sort, are stable by default.

Date sorting requires parsing date strings into comparable temporal values, a task complicated by the vast diversity of date formats used worldwide. The American MM/DD/YYYY, European DD/MM/YYYY, and ISO 8601 YYYY-MM-DD formats can produce ambiguous dates—is "01/02/2024" January 2nd or February 1st? Correct date sorting depends on accurate format detection, which may require analyzing multiple values in a column to disambiguate the format. The ISO 8601 format, with its year-first ordering, has the unique property of sorting correctly under simple lexicographic comparison, which is one reason it is recommended as the standard date format for data interchange.