*breaks lurk to comment*
I used
TumblThree, it seems to be the easiest. For a caption blog like orangeorc's, you'll want to make sure to backup photo metadata. Then if you take the generated metadata.txt, make it metadata.html, and open that in your browser, all the stories are there. It also downloads all the pictures, but unfortunately they aren't embedded in the metadata file. I might mess around a little to see if there's an easy way to do that, but for now, this will work to keep the stories.
*re-enters lurk mode*