I must respectfully disagree quite strongly! It's parsecs that tell you something useful, not light years - and the trick is combining them with arcseconds, the other unit you unfairly malign. If something is N parsecs distant, then one arcsecond of angular size corresponds to N AU of physical size. So for instance, when I discovered a circumstellar disk 0.8 arcsec across 1000 pc away, I could immediately and easily know the disk was actually 800 AU in diameter. I don't really care that I'm looking at photons emitted 3260 years ago - that's not interesting from a standpoint of figuring out the astrophysics. So light years are pretty useless. From a physical perspective, arcseconds and parsecs are great units since they make it so easy to convert between angular and physical sizes.
Now, this all gets harder for extragalactic distances since the relation goes nonlinear due to relativistic spacetime expansion. But none of that extragalactic stuff is very interesting, anyway.