It is fascinating that the discussion here seems to go into very different directions, though. A lot less "study made this and that mistake" comments here.

Gah! I really wish the HN submission feature could borrow some ideas from Stack Overflow's duplicate question detection.

(FWIW, the full title is "An Efficiency Comparison of Document Preparation Systems Used in Academic Research and Development", which is too long for HN, so the previous submission also changed it.)

We're planning to work on better duplicate detection in the new year.

I like how it works now organically. If the discussion was broadly seen, there won't be any up vote interest, and if not, then it clearly hasn't taken its course.

We like that too, and it's by design. A story that hasn't been broadly seen won't count as a duplicate in the new system either.