That's interesting - the people I was talking with made a point of using the corrected version multiple times over a short space of time, and seemed to assimilate the change. Anedotes aren't data though - if you have evidence to contradict that observation I'd be interested to see it - yes.
Thanks.
Added in edit: you were downvoted - I don't know why, so I've upvoted you.
