The more I work in the dynamic languages here, the more I think we need a 'efficient' vote as well as best practices and clever. While best practices and cleverness are good, I see too many answers that are O(n^2) solutions for questions that have a very clear O(n) solution that are voted to the top because they are a single line and use some cool language features.

I think these answers are great and they show you some really neat aspects of certain languages but they are not solutions that would be production ready and often overshadow the most efficient solutions

Is there any request for a efficent voting mechanism? I would love that.

All? short circuts.

More stats. How long did I take to solve it? Avereage solve time. A kata breakdown(how I earned what, it seems I randomly got 40 kata). What languages do I spend the most time solving? What langauges do I spend the least time? Change in leadership position over time. Any more tracking would be cool to look at.

It doesn't provide a way to submit tests though. You need to submit a solution and see that output.
Also the user documentation seems out of date (the returned output does not match what they say should be returned) so it is resonable to assume the rest could be out of date.

No test cases with words with apostrophe.

Comparing an array like that would be n, linear.

I really like this solution, your process_items is very clean and short. I'm curious why you decided to use two lists instead of a hash though? Also, don't you need to reset your @identify to proc {|obj| obj} in your reset? Otherwise you would be using the identify from the previous batch job.

Regex usually isn't the best for readability but it's a lot cleaner than how I removed the _ and =. I'm new to ruby so bare with me when I ask, are ternary operators prefered? I see them everywhere in the solutions but I've never liked the way read vs a regular if else.

Haha yeah, ruby has a lot of hidden gems(tee hee) in the collections docs

Sorting isn't really needed, you can do it a lot faster comparing every two elements.