Imagine an apatosaurus with a long, elephant-like snout. Plenty of people have. That's because the nostril placement on sauropod dinosaurs is, in some ways, remarkably similar to that of trunked animals that live today. In both cases, the nostrils are large, and they're located up around what we'd call the forehead, kind of smack between the eyes.
On the one hand, this is one of those things that it's really hard to ever know for certain. We don't have preserved soft tissue, so when we make models of what dinosaurs might have looked like we're really going on clues from the bones and comparisons to living animals with similar bone structure. Because of that, it is somewhat reasonable to suggest that hey, maybe, sauropods really did look like grumpy diplodocus in the image above. It's fun to speculate.
But not all speculations are created equal. In a fascinating post at the Tetrapod Zoology blog, Darren Naish explains why a superficial similarity to trunked animals isn't enough to counteract the much-more prevalent evidence against sauropod trunks. One of the more interesting lines of evidence he points out is the fact that dinosaurs apparently lacked the facial which form the trunk in living animals. We know this partly because muscles leave their signature on bone, and Naish says there's no evidence sauropods had the right facial muscles. It's further bolstered by the fact that the animals most closely related to sauropods don't have those facial muscles, either.
Naish's piece reminds me of the last time we talked about sauropod biology here. That, too, dealt with the fact that superficial similarities aren't enough to infer that two animals must have identical biology. Only, in that case, we were talking about the differences between the long necks of giraffes and the long necks of sauropods.