Board Thread:Code Review/@comment-1757994-20151023054248/@comment-1757994-20151024122227

Wow, actual metrics. You inspired me to do some of my own. I ran both sets of code on Community Central. Then I used
 * number of files = $('#mw-dupimages h3').length
 * number of duplicates = $('#mw-dupimages li').length

I also ran a code snippet to test uniqueness of the lists


 * 1) Original
 * 415 files with 484 duplicates (415 + 484 = 899)
 * 167 files are not unique
 * i.e., it really found only 248 files with ??? duplicates
 * 1) Updated
 * 477 files with 569 duplicates (477 + 569 = 1046)
 * 0 files are not unique

The updated code found about twice as many files with duplicates for only about 30% more network traffic.

I'll ponder why it took twice as many requests, though.