Comscore will be measuring "heavy, medium, and light" users separately, which means we'll get a better read of data than an average where one or the other extreme of user can skew the average.
Hard to see how this can be a bad thing. Not sure that's true of Nielsen's move to go to "time spent on site" though objections over people keeping browsers open for hours without viewing will be taken care of. For example, they can stop counting (and already do) if a page has been open for more than an hour, and say they'll also soon only count whatever page is on top. So if you hide something, or minimize it, it's not counted as being consumed.