Simple. The code came that way from bytecoin, from which MRO was forked (or alternately relaunched), and was supposedly going for two years. So somehow that little gem was "missed" by them for two years. When the Monero team discovered the issue, we released it, and then most of the other cryptonote coins copied it, as I said.
Were it not for the Monero team fixing it, would it have been allowed to continue for another two years?
I firmly believe that noodle should not be part of development team.
Like I stated in IRC, I am not part of the "dev team", I never was. Just so happens I took a look at the code and changed some extremely easy to spot "errors". I then decided to release the binary because I thought MRO would benefit from it. I made this decision individually and nobody else should be culpable, especially the community of individuals who have come together to maintain and foster the software.
By the way, I'm not even a real coder, so whatever changes I made should be easy to spot; especially for experienced developers.Cheers.
Doesn't seem like the monero dev team actually discovered it all. Boolberry certainly didn't copy this un-cripple 'fix' or optimization as you like to call it.
It seems logical that another actor (perhaps relating to XMR team if you're a fan of conspiracy theories, or conceivably someone previously involved in bytecoin, -- who naturally would be aware, OR an entirely independent user who has basic coding knowledge and few minutes to spare glancing over the code) had realised this and was operating at a significant advantage for some time.
The whole thing was a bit of a facepalm moment.

Although it seems to be in the past now.