Is it possible for an item to have good usability and a positive user experience for one core group of users and, by virtue of that design, make the experience worse for another core group of users? Does this then mean that the overall usability is actually poor?
I thought of this because I had borrowed and subsequently (accidentally) stolen a pen from a friend who rather enjoyed writing with it because it wrote smoothly, made dark lines, and was comfortable to hold. Using this pen was awful for me as a left-hander; the ink smeared all over the paper, it was hard to read what I had written, and I had to throw away several checks I was trying to write with it (seriously, who doesn't accept online bill pay yet?)
My question is: does this pen have good usability? A positive user experience? Should the evaluation be restricted to a definite user (i.e. right-hander) or should it be evaluated against all possible users? What other situations (software or otherwise) could/has this occurred in?