Hey D,
You and I have talked in the past at TFF, remember? I'm not sure exactly how others do it, but just an FYI, our password management is a little more "enterprise class" than what you described. If we post *any* change to the user/pass data, our system expects a positive response from the module. If it doesn't get that response, it assumes the post failed and it will post again until it succeeds. I would think the better systems have at least that level of data integrity protection built in.
Also, as you know, using your bandwidth fraud package controls to stop sharing without making life too difficult for your customers is, of course, a balance. We work with our merchants on those strategies. Hit me up if you'd like to discuss.
