Ordered Hashes in Ruby 1.9

In Ruby 1.9, hash tables preserve insertion order by maintaining a doubly-linked list. This seems like a dubious feature. It only accords with the principle of least surprise if you don’t know how a hash table works. I don’t think I’ve ever needed a hash to do this, and I worry that making the order mostly predictable will lead to fragile code that accidentally relies on the ordering where it shouldn’t. All of this comes at the cost of increased memory use. Overall, it’s faster than Ruby’s previous hash implementation, but presumably it would be even faster without the ordering feature.