At first, I used it indoors, photographing family. Yeah, it's not good for portraits, but in the low light and capturing a wider view, it works well. As long as I don't pixel-peep, the results look good. Since the camera also used ISO 1600, I figure it doesn't matter that it's not terribly sharp. (Am I going to have sharper noise?
Later, when I try to take some quick test photos, I find that f2.8 is a bit soft, as they say. The corners look really bad. And yet, I didn't notice this on my indoor pics, even though I took quite a few (most at f2.8 and high ISO). I guess I just didn't manage to force a subject all the way into the corner of the frame, ever. Or have any detail that would highlight the blur.
So, I took test shots outside at the woods. Corners remain poor until around f8 where I'm not sure if they're quite "fixed" yet, but are at least no longer obviously blurry. My copy of the lens seem to be slightly worse on one side, but perhaps not by much. But F5 looks pretty good, corners aside, and I think already beats the 18-55 in contrast and sharpness. I guess I should do more testing, but I think the ultimate proof is whether you end up with good looking photos (that aren't too distracted by blurriness in the corners).
So, outside as a travel lens, it should be ok I think? Although, at f8, I shot in kind of overcast morning light, and with a 1/30 (?) shutter speed, managed to get enough blur to negate all of the quality that f8 was supposed to bring to me.
I guess my concern is that I don't want to end up with a lot of photos that seem too flawed just because I wanted a slightly smaller lens. If anything, a prime should do better! I think it does, with some caveats.