Code clone detection has many appealing applications, ranging from maintenance effort reduction to example identification based on similarities. However, traditional approaches to clone detection have a significant feedback loop; hours and even days are necessary to identify clones from a large amount of source code.

Instant code clone search challenges what many have taken for granted: clone detection can be realized in a time period that is sufficiently short enough to convey a feeling of instantaneous feedback. By using advanced multidimensional indexing (R*tree), the approaches proposed by the author are more accurate than http://koders.com and http://www.google.com/codesearch.

The motivation and the benefits are well described, which gives good value to the paper. Instant code clone search requires a strong knowledge of indexing because of the amount of formalism employed. This paper is recommended for those with expertise in database indexing.