It could just be that more people are shooting digital and looking at their images at 100% or greater.
Most of the older replies are probably from shooting negative film and checking prints. In this case, determining how good the 28-135/IS is relative to even an L lens on relatively small prints is negligable.
There's a saying, "measure with a micrometer, mark with chalk, and cut with an axe," which is more true now that many people are shooting digital than ever before. In other words, many people are looking inanely close at their images and worrying about artifacts that'll never see in even a decent size print. Whereas, in the past many people were judging lenses by comparing prints or looking at chromes with a loupe thats magnification approximated their intended print size.
It becomes a double edge sword. Sure, it's good (and I do) to look at images at 100% to judge quality. The flip side is the realization of what size prints you intend to make. Or in the extreme, if you only plan to view the full image, 1280x1024, on the screen.
I own this lens and like it a lot. Sure it's not as good as my L lenses or EX lenses. However, even now that I'm about 100% digital and it's field of view has changed on my 10d, it serves a purpose.